Imported Upstream version 0.19.7

[platform/upstream/gettext.git] / gettext-tools / doc / gettext.texi

\input texinfo          @c -*-texinfo-*-
@c %**start of header
@setfilename gettext.info
@c The @ifset makeinfo ... @end ifset conditional evaluates to true in makeinfo
@c for info and html output, but to false in texi2html.
@ifnottex
@ifclear texi2html
@set makeinfo
@end ifclear
@end ifnottex
@c The @documentencoding is needed for makeinfo; texi2html 1.52
@c doesn't recognize it.
@ifset makeinfo
@documentencoding UTF-8
@end ifset
@settitle GNU @code{gettext} utilities
@finalout
@c Indices:
@c   am = autoconf macro  @amindex
@c   cp = concept         @cindex
@c   ef = emacs function  @efindex
@c   em = emacs mode      @emindex
@c   ev = emacs variable  @evindex
@c   fn = function        @findex
@c   kw = keyword         @kwindex
@c   op = option          @opindex
@c   pg = program         @pindex
@c   vr = variable        @vindex
@c Unused predefined indices:
@c   tp = type            @tindex
@c   ky = keystroke       @kindex
@defcodeindex am
@defcodeindex ef
@defindex em
@defcodeindex ev
@defcodeindex kw
@defcodeindex op
@syncodeindex ef em
@syncodeindex ev em
@syncodeindex fn cp
@syncodeindex kw cp
@ifclear texi2html
@firstparagraphindent insert
@end ifclear
@c %**end of header

@include version.texi

@ifinfo
@dircategory GNU Gettext Utilities
@direntry
* gettext: (gettext).                          GNU gettext utilities.
* autopoint: (gettext)autopoint Invocation.    Copy gettext infrastructure.
* envsubst: (gettext)envsubst Invocation.      Expand environment variables.
* gettextize: (gettext)gettextize Invocation.  Prepare a package for gettext.
* msgattrib: (gettext)msgattrib Invocation.    Select part of a PO file.
* msgcat: (gettext)msgcat Invocation.          Combine several PO files.
* msgcmp: (gettext)msgcmp Invocation.          Compare a PO file and template.
* msgcomm: (gettext)msgcomm Invocation.        Match two PO files.
* msgconv: (gettext)msgconv Invocation.        Convert PO file to encoding.
* msgen: (gettext)msgen Invocation.            Create an English PO file.
* msgexec: (gettext)msgexec Invocation.        Process a PO file.
* msgfilter: (gettext)msgfilter Invocation.    Pipe a PO file through a filter.
* msgfmt: (gettext)msgfmt Invocation.          Make MO files out of PO files.
* msggrep: (gettext)msggrep Invocation.        Select part of a PO file.
* msginit: (gettext)msginit Invocation.        Create a fresh PO file.
* msgmerge: (gettext)msgmerge Invocation.      Update a PO file from template.
* msgunfmt: (gettext)msgunfmt Invocation.      Uncompile MO file into PO file.
* msguniq: (gettext)msguniq Invocation.        Unify duplicates for PO file.
* ngettext: (gettext)ngettext Invocation.      Translate a message with plural.
* xgettext: (gettext)xgettext Invocation.      Extract strings into a PO file.
* ISO639: (gettext)Language Codes.             ISO 639 language codes.
* ISO3166: (gettext)Country Codes.             ISO 3166 country codes.
@end direntry
@end ifinfo

@ifinfo
This file provides documentation for GNU @code{gettext} utilities.
It also serves as a reference for the free Translation Project.

@copying
Copyright (C) 1995-1998, 2001-2015 Free Software Foundation, Inc.

This manual is free documentation.  It is dually licensed under the
GNU FDL and the GNU GPL.  This means that you can redistribute this
manual under either of these two licenses, at your choice.

This manual is covered by the GNU FDL.  Permission is granted to copy,
distribute and/or modify this document under the terms of the
GNU Free Documentation License (FDL), either version 1.2 of the
License, or (at your option) any later version published by the
Free Software Foundation (FSF); with no Invariant Sections, with no
Front-Cover Text, and with no Back-Cover Texts.
A copy of the license is included in @ref{GNU FDL}.

This manual is covered by the GNU GPL.  You can redistribute it and/or
modify it under the terms of the GNU General Public License (GPL), either
version 2 of the License, or (at your option) any later version published
by the Free Software Foundation (FSF).
A copy of the license is included in @ref{GNU GPL}.
@end copying
@end ifinfo

@titlepage
@title GNU gettext tools, version @value{VERSION}
@subtitle Native Language Support Library and Tools
@subtitle Edition @value{EDITION}, @value{UPDATED}
@author Ulrich Drepper
@author Jim Meyering
@author Fran@,{c}ois Pinard
@author Bruno Haible

@ifnothtml
@page
@vskip 0pt plus 1filll
@c @insertcopying
Copyright (C) 1995-1998, 2001-2012 Free Software Foundation, Inc.

This manual is free documentation.  It is dually licensed under the
GNU FDL and the GNU GPL.  This means that you can redistribute this
manual under either of these two licenses, at your choice.

This manual is covered by the GNU FDL.  Permission is granted to copy,
distribute and/or modify this document under the terms of the
GNU Free Documentation License (FDL), either version 1.2 of the
License, or (at your option) any later version published by the
Free Software Foundation (FSF); with no Invariant Sections, with no
Front-Cover Text, and with no Back-Cover Texts.
A copy of the license is included in @ref{GNU FDL}.

This manual is covered by the GNU GPL.  You can redistribute it and/or
modify it under the terms of the GNU General Public License (GPL), either
version 2 of the License, or (at your option) any later version published
by the Free Software Foundation (FSF).
A copy of the license is included in @ref{GNU GPL}.
@end ifnothtml
@end titlepage

@c Table of Contents
@contents

@ifset makeinfo
@node Top, Introduction, (dir), (dir)
@top GNU @code{gettext} utilities

This manual documents the GNU gettext tools and the GNU libintl library,
version @value{VERSION}.

@menu
* Introduction::                Introduction
* Users::                       The User's View
* PO Files::                    The Format of PO Files
* Sources::                     Preparing Program Sources
* Template::                    Making the PO Template File
* Creating::                    Creating a New PO File
* Updating::                    Updating Existing PO Files
* Editing::                     Editing PO Files
* Manipulating::                Manipulating PO Files
* Binaries::                    Producing Binary MO Files
* Programmers::                 The Programmer's View
* Translators::                 The Translator's View
* Maintainers::                 The Maintainer's View
* Installers::                  The Installer's and Distributor's View
* Programming Languages::       Other Programming Languages
* Conclusion::                  Concluding Remarks

* Language Codes::              ISO 639 language codes
* Country Codes::               ISO 3166 country codes
* Licenses::                    Licenses

* Program Index::               Index of Programs
* Option Index::                Index of Command-Line Options
* Variable Index::              Index of Environment Variables
* PO Mode Index::               Index of Emacs PO Mode Commands
* Autoconf Macro Index::        Index of Autoconf Macros
* Index::                       General Index

@detailmenu
 --- The Detailed Node Listing ---

Introduction

* Why::                         The Purpose of GNU @code{gettext}
* Concepts::                    I18n, L10n, and Such
* Aspects::                     Aspects in Native Language Support
* Files::                       Files Conveying Translations
* Overview::                    Overview of GNU @code{gettext}

The User's View

* System Installation::         Questions During Operating System Installation
* Setting the GUI Locale::      How to Specify the Locale Used by GUI Programs
* Setting the POSIX Locale::    How to Specify the Locale According to POSIX
* Installing Localizations::    How to Install Additional Translations

Setting the Locale through Environment Variables

* Locale Names::                How a Locale Specification Looks Like
* Locale Environment Variables::  Which Environment Variable Specfies What
* The LANGUAGE variable::       How to Specify a Priority List of Languages

Preparing Program Sources

* Importing::                   Importing the @code{gettext} declaration
* Triggering::                  Triggering @code{gettext} Operations
* Preparing Strings::           Preparing Translatable Strings
* Mark Keywords::               How Marks Appear in Sources
* Marking::                     Marking Translatable Strings
* c-format Flag::               Telling something about the following string
* Special cases::               Special Cases of Translatable Strings
* Bug Report Address::          Letting Users Report Translation Bugs
* Names::                       Marking Proper Names for Translation
* Libraries::                   Preparing Library Sources

Making the PO Template File

* xgettext Invocation::         Invoking the @code{xgettext} Program

Creating a New PO File

* msginit Invocation::          Invoking the @code{msginit} Program
* Header Entry::                Filling in the Header Entry

Updating Existing PO Files

* msgmerge Invocation::         Invoking the @code{msgmerge} Program

Editing PO Files

* KBabel::                      KDE's PO File Editor
* Gtranslator::                 GNOME's PO File Editor
* PO Mode::                     Emacs's PO File Editor
* Compendium::                  Using Translation Compendia

Emacs's PO File Editor

* Installation::                Completing GNU @code{gettext} Installation
* Main PO Commands::            Main Commands
* Entry Positioning::           Entry Positioning
* Normalizing::                 Normalizing Strings in Entries
* Translated Entries::          Translated Entries
* Fuzzy Entries::               Fuzzy Entries
* Untranslated Entries::        Untranslated Entries
* Obsolete Entries::            Obsolete Entries
* Modifying Translations::      Modifying Translations
* Modifying Comments::          Modifying Comments
* Subedit::                     Mode for Editing Translations
* C Sources Context::           C Sources Context
* Auxiliary::                   Consulting Auxiliary PO Files

Using Translation Compendia

* Creating Compendia::          Merging translations for later use
* Using Compendia::             Using older translations if they fit

Manipulating PO Files

* msgcat Invocation::           Invoking the @code{msgcat} Program
* msgconv Invocation::          Invoking the @code{msgconv} Program
* msggrep Invocation::          Invoking the @code{msggrep} Program
* msgfilter Invocation::        Invoking the @code{msgfilter} Program
* msguniq Invocation::          Invoking the @code{msguniq} Program
* msgcomm Invocation::          Invoking the @code{msgcomm} Program
* msgcmp Invocation::           Invoking the @code{msgcmp} Program
* msgattrib Invocation::        Invoking the @code{msgattrib} Program
* msgen Invocation::            Invoking the @code{msgen} Program
* msgexec Invocation::          Invoking the @code{msgexec} Program
* Colorizing::                  Highlighting parts of PO files
* libgettextpo::                Writing your own programs that process PO files

Highlighting parts of PO files

* The --color option::          Triggering colorized output
* The TERM variable::           The environment variable @code{TERM}
* The --style option::          The @code{--style} option
* Style rules::                 Style rules for PO files
* Customizing less::            Customizing @code{less} for viewing PO files

Producing Binary MO Files

* msgfmt Invocation::           Invoking the @code{msgfmt} Program
* msgunfmt Invocation::         Invoking the @code{msgunfmt} Program
* MO Files::                    The Format of GNU MO Files

The Programmer's View

* catgets::                     About @code{catgets}
* gettext::                     About @code{gettext}
* Comparison::                  Comparing the two interfaces
* Using libintl.a::             Using libintl.a in own programs
* gettext grok::                Being a @code{gettext} grok
* Temp Programmers::            Temporary Notes for the Programmers Chapter

About @code{catgets}

* Interface to catgets::        The interface
* Problems with catgets::       Problems with the @code{catgets} interface?!

About @code{gettext}

* Interface to gettext::        The interface
* Ambiguities::                 Solving ambiguities
* Locating Catalogs::           Locating message catalog files
* Charset conversion::          How to request conversion to Unicode
* Contexts::                    Solving ambiguities in GUI programs
* Plural forms::                Additional functions for handling plurals
* Optimized gettext::           Optimization of the *gettext functions

Temporary Notes for the Programmers Chapter

* Temp Implementations::        Temporary - Two Possible Implementations
* Temp catgets::                Temporary - About @code{catgets}
* Temp WSI::                    Temporary - Why a single implementation
* Temp Notes::                  Temporary - Notes

The Translator's View

* Trans Intro 0::               Introduction 0
* Trans Intro 1::               Introduction 1
* Discussions::                 Discussions
* Organization::                Organization
* Information Flow::            Information Flow
* Translating plural forms::    How to fill in @code{msgstr[0]}, @code{msgstr[1]}
* Prioritizing messages::       How to find which messages to translate first

Organization

* Central Coordination::        Central Coordination
* National Teams::              National Teams
* Mailing Lists::               Mailing Lists

National Teams

* Sub-Cultures::                Sub-Cultures
* Organizational Ideas::        Organizational Ideas

The Maintainer's View

* Flat and Non-Flat::           Flat or Non-Flat Directory Structures
* Prerequisites::               Prerequisite Works
* gettextize Invocation::       Invoking the @code{gettextize} Program
* Adjusting Files::             Files You Must Create or Alter
* autoconf macros::             Autoconf macros for use in @file{configure.ac}
* Version Control Issues::      
* Release Management::          Creating a Distribution Tarball

Files You Must Create or Alter

* po/POTFILES.in::              @file{POTFILES.in} in @file{po/}
* po/LINGUAS::                  @file{LINGUAS} in @file{po/}
* po/Makevars::                 @file{Makevars} in @file{po/}
* po/Rules-*::                  Extending @file{Makefile} in @file{po/}
* configure.ac::                @file{configure.ac} at top level
* config.guess::                @file{config.guess}, @file{config.sub} at top level
* mkinstalldirs::               @file{mkinstalldirs} at top level
* aclocal::                     @file{aclocal.m4} at top level
* acconfig::                    @file{acconfig.h} at top level
* config.h.in::                 @file{config.h.in} at top level
* Makefile::                    @file{Makefile.in} at top level
* src/Makefile::                @file{Makefile.in} in @file{src/}
* lib/gettext.h::               @file{gettext.h} in @file{lib/}

Autoconf macros for use in @file{configure.ac}

* AM_GNU_GETTEXT::              AM_GNU_GETTEXT in @file{gettext.m4}
* AM_GNU_GETTEXT_VERSION::      AM_GNU_GETTEXT_VERSION in @file{gettext.m4}
* AM_GNU_GETTEXT_NEED::         AM_GNU_GETTEXT_NEED in @file{gettext.m4}
* AM_GNU_GETTEXT_INTL_SUBDIR::  AM_GNU_GETTEXT_INTL_SUBDIR in @file{intldir.m4}
* AM_PO_SUBDIRS::               AM_PO_SUBDIRS in @file{po.m4}
* AM_XGETTEXT_OPTION::          AM_XGETTEXT_OPTION in @file{po.m4}
* AM_ICONV::                    AM_ICONV in @file{iconv.m4}

Integrating with Version Control Systems

* Distributed Development::     Avoiding version mismatch in distributed development
* Files under Version Control::  Files to put under version control
* Translations under Version Control::  Put PO Files under Version Control
* autopoint Invocation::        Invoking the @code{autopoint} Program

Other Programming Languages

* Language Implementors::       The Language Implementor's View
* Programmers for other Languages::  The Programmer's View
* Translators for other Languages::  The Translator's View
* Maintainers for other Languages::  The Maintainer's View
* List of Programming Languages::  Individual Programming Languages
* List of Data Formats::        Internationalizable Data

The Translator's View

* c-format::                    C Format Strings
* objc-format::                 Objective C Format Strings
* sh-format::                   Shell Format Strings
* python-format::               Python Format Strings
* lisp-format::                 Lisp Format Strings
* elisp-format::                Emacs Lisp Format Strings
* librep-format::               librep Format Strings
* scheme-format::               Scheme Format Strings
* smalltalk-format::            Smalltalk Format Strings
* java-format::                 Java Format Strings
* csharp-format::               C# Format Strings
* awk-format::                  awk Format Strings
* object-pascal-format::        Object Pascal Format Strings
* ycp-format::                  YCP Format Strings
* tcl-format::                  Tcl Format Strings
* perl-format::                 Perl Format Strings
* php-format::                  PHP Format Strings
* gcc-internal-format::         GCC internal Format Strings
* gfc-internal-format::         GFC internal Format Strings
* qt-format::                   Qt Format Strings
* qt-plural-format::            Qt Plural Format Strings
* kde-format::                  KDE Format Strings
* boost-format::                Boost Format Strings
* lua-format::                  Lua Format Strings
* javascript-format::           JavaScript Format Strings

Individual Programming Languages

* C::                           C, C++, Objective C
* sh::                          sh - Shell Script
* bash::                        bash - Bourne-Again Shell Script
* Python::                      Python
* Common Lisp::                 GNU clisp - Common Lisp
* clisp C::                     GNU clisp C sources
* Emacs Lisp::                  Emacs Lisp
* librep::                      librep
* Scheme::                      GNU guile - Scheme
* Smalltalk::                   GNU Smalltalk
* Java::                        Java
* C#::                          C#
* gawk::                        GNU awk
* Pascal::                      Pascal - Free Pascal Compiler
* wxWidgets::                   wxWidgets library
* YCP::                         YCP - YaST2 scripting language
* Tcl::                         Tcl - Tk's scripting language
* Perl::                        Perl
* PHP::                         PHP Hypertext Preprocessor
* Pike::                        Pike
* GCC-source::                  GNU Compiler Collection sources
* Lua::                         Lua
* JavaScript::                  JavaScript
* Vala::                        Vala

sh - Shell Script

* Preparing Shell Scripts::     Preparing Shell Scripts for Internationalization
* gettext.sh::                  Contents of @code{gettext.sh}
* gettext Invocation::          Invoking the @code{gettext} program
* ngettext Invocation::         Invoking the @code{ngettext} program
* envsubst Invocation::         Invoking the @code{envsubst} program
* eval_gettext Invocation::     Invoking the @code{eval_gettext} function
* eval_ngettext Invocation::    Invoking the @code{eval_ngettext} function

Perl

* General Problems::            General Problems Parsing Perl Code
* Default Keywords::            Which Keywords Will xgettext Look For?
* Special Keywords::            How to Extract Hash Keys
* Quote-like Expressions::      What are Strings And Quote-like Expressions?
* Interpolation I::             Invalid String Interpolation
* Interpolation II::            Valid String Interpolation
* Parentheses::                 When To Use Parentheses
* Long Lines::                  How To Grok with Long Lines
* Perl Pitfalls::               Bugs, Pitfalls, and Things That Do Not Work

Internationalizable Data

* POT::                         POT - Portable Object Template
* RST::                         Resource String Table
* Glade::                       Glade - GNOME user interface description
* GSettings::                   GSettings - GNOME user configuration schema
* AppData::                     AppData - freedesktop.org application description
* Preparing ITS Rules::         Preparing Rules for XML Internationalization

Concluding Remarks

* History::                     History of GNU @code{gettext}
* References::                  Related Readings

Language Codes

* Usual Language Codes::        Two-letter ISO 639 language codes
* Rare Language Codes::         Three-letter ISO 639 language codes

Licenses

* GNU GPL::                     GNU General Public License
* GNU LGPL::                    GNU Lesser General Public License
* GNU FDL::                     GNU Free Documentation License

@end detailmenu
@end menu

@end ifset

@node Introduction, Users, Top, Top
@chapter Introduction

This chapter explains the goals sought in the creation
of GNU @code{gettext} and the free Translation Project.
Then, it explains a few broad concepts around
Native Language Support, and positions message translation with regard
to other aspects of national and cultural variance, as they apply
to programs.  It also surveys those files used to convey the
translations.  It explains how the various tools interact in the
initial generation of these files, and later, how the maintenance
cycle should usually operate.

@cindex sex
@cindex he, she, and they
@cindex she, he, and they
In this manual, we use @emph{he} when speaking of the programmer or
maintainer, @emph{she} when speaking of the translator, and @emph{they}
when speaking of the installers or end users of the translated program.
This is only a convenience for clarifying the documentation.  It is
@emph{absolutely} not meant to imply that some roles are more appropriate
to males or females.  Besides, as you might guess, GNU @code{gettext}
is meant to be useful for people using computers, whatever their sex,
race, religion or nationality!

@cindex bug report address
Please send suggestions and corrections to:

@example
@group
@r{Internet address:}
    bug-gnu-gettext@@gnu.org
@end group
@end example

@noindent
Please include the manual's edition number and update date in your messages.

@menu
* Why::                         The Purpose of GNU @code{gettext}
* Concepts::                    I18n, L10n, and Such
* Aspects::                     Aspects in Native Language Support
* Files::                       Files Conveying Translations
* Overview::                    Overview of GNU @code{gettext}
@end menu

@node Why, Concepts, Introduction, Introduction
@section The Purpose of GNU @code{gettext}

Usually, programs are written and documented in English, and use
English at execution time to interact with users.  This is true
not only of GNU software, but also of a great deal of proprietary
and free software.  Using a common language is quite handy for
communication between developers, maintainers and users from all
countries.  On the other hand, most people are less comfortable with
English than with their own native language, and would prefer to
use their mother tongue for day to day's work, as far as possible.
Many would simply @emph{love} to see their computer screen showing
a lot less of English, and far more of their own language.

@cindex Translation Project
However, to many people, this dream might appear so far fetched that
they may believe it is not even worth spending time thinking about
it.  They have no confidence at all that the dream might ever
become true.  Yet some have not lost hope, and have organized themselves.
The Translation Project is a formalization of this hope into a
workable structure, which has a good chance to get all of us nearer
the achievement of a truly multi-lingual set of programs.

GNU @code{gettext} is an important step for the Translation Project,
as it is an asset on which we may build many other steps.  This package
offers to programmers, translators and even users, a well integrated
set of tools and documentation.  Specifically, the GNU @code{gettext}
utilities are a set of tools that provides a framework within which
other free packages may produce multi-lingual messages.  These tools
include

@itemize @bullet
@item
A set of conventions about how programs should be written to support
message catalogs.

@item
A directory and file naming organization for the message catalogs
themselves.

@item
A runtime library supporting the retrieval of translated messages.

@item
A few stand-alone programs to massage in various ways the sets of
translatable strings, or already translated strings.

@item
A library supporting the parsing and creation of files containing
translated messages.

@item
A special mode for Emacs@footnote{In this manual, all mentions of Emacs
refers to either GNU Emacs or to XEmacs, which people sometimes call FSF
Emacs and Lucid Emacs, respectively.} which helps preparing these sets
and bringing them up to date.
@end itemize

GNU @code{gettext} is designed to minimize the impact of
internationalization on program sources, keeping this impact as small
and hardly noticeable as possible.  Internationalization has better
chances of succeeding if it is very light weighted, or at least,
appear to be so, when looking at program sources.

The Translation Project also uses the GNU @code{gettext} distribution
as a vehicle for documenting its structure and methods.  This goes
beyond the strict technicalities of documenting the GNU @code{gettext}
proper.  By so doing, translators will find in a single place, as
far as possible, all they need to know for properly doing their
translating work.  Also, this supplemental documentation might also
help programmers, and even curious users, in understanding how GNU
@code{gettext} is related to the remainder of the Translation
Project, and consequently, have a glimpse at the @emph{big picture}.

@node Concepts, Aspects, Why, Introduction
@section I18n, L10n, and Such

@cindex i18n
@cindex l10n
Two long words appear all the time when we discuss support of native
language in programs, and these words have a precise meaning, worth
being explained here, once and for all in this document.  The words are
@emph{internationalization} and @emph{localization}.  Many people,
tired of writing these long words over and over again, took the
habit of writing @dfn{i18n} and @dfn{l10n} instead, quoting the first
and last letter of each word, and replacing the run of intermediate
letters by a number merely telling how many such letters there are.
But in this manual, in the sake of clarity, we will patiently write
the names in full, each time@dots{}

@cindex internationalization
By @dfn{internationalization}, one refers to the operation by which a
program, or a set of programs turned into a package, is made aware of and
able to support multiple languages.  This is a generalization process,
by which the programs are untied from calling only English strings or
other English specific habits, and connected to generic ways of doing
the same, instead.  Program developers may use various techniques to
internationalize their programs.  Some of these have been standardized.
GNU @code{gettext} offers one of these standards.  @xref{Programmers}.

@cindex localization
By @dfn{localization}, one means the operation by which, in a set
of programs already internationalized, one gives the program all
needed information so that it can adapt itself to handle its input
and output in a fashion which is correct for some native language and
cultural habits.  This is a particularisation process, by which generic
methods already implemented in an internationalized program are used
in specific ways.  The programming environment puts several functions
to the programmers disposal which allow this runtime configuration.
The formal description of specific set of cultural habits for some
country, together with all associated translations targeted to the
same native language, is called the @dfn{locale} for this language
or country.  Users achieve localization of programs by setting proper
values to special environment variables, prior to executing those
programs, identifying which locale should be used.

In fact, locale message support is only one component of the cultural
data that makes up a particular locale.  There are a whole host of
routines and functions provided to aid programmers in developing
internationalized software and which allow them to access the data
stored in a particular locale.  When someone presently refers to a
particular locale, they are obviously referring to the data stored
within that particular locale.  Similarly, if a programmer is referring
to ``accessing the locale routines'', they are referring to the
complete suite of routines that access all of the locale's information.

@cindex NLS
@cindex Native Language Support
@cindex Natural Language Support
One uses the expression @dfn{Native Language Support}, or merely NLS,
for speaking of the overall activity or feature encompassing both
internationalization and localization, allowing for multi-lingual
interactions in a program.  In a nutshell, one could say that
internationalization is the operation by which further localizations
are made possible.

Also, very roughly said, when it comes to multi-lingual messages,
internationalization is usually taken care of by programmers, and
localization is usually taken care of by translators.

@node Aspects, Files, Concepts, Introduction
@section Aspects in Native Language Support

@cindex translation aspects
For a totally multi-lingual distribution, there are many things to
translate beyond output messages.

@itemize @bullet
@item
As of today, GNU @code{gettext} offers a complete toolset for
translating messages output by C programs.  Perl scripts and shell
scripts will also need to be translated.  Even if there are today some hooks
by which this can be done, these hooks are not integrated as well as they
should be.

@item
Some programs, like @code{autoconf} or @code{bison}, are able
to produce other programs (or scripts).  Even if the generating
programs themselves are internationalized, the generated programs they
produce may need internationalization on their own, and this indirect
internationalization could be automated right from the generating
program.  In fact, quite usually, generating and generated programs
could be internationalized independently, as the effort needed is
fairly orthogonal.

@item
A few programs include textual tables which might need translation
themselves, independently of the strings contained in the program
itself.  For example, @w{RFC 1345} gives an English description for each
character which the @code{recode} program is able to reconstruct at execution.
Since these descriptions are extracted from the RFC by mechanical means,
translating them properly would require a prior translation of the RFC
itself.

@item
Almost all programs accept options, which are often worded out so to
be descriptive for the English readers; one might want to consider
offering translated versions for program options as well.

@item
Many programs read, interpret, compile, or are somewhat driven by
input files which are texts containing keywords, identifiers, or
replies which are inherently translatable.  For example, one may want
@code{gcc} to allow diacriticized characters in identifiers or use
translated keywords; @samp{rm -i} might accept something else than
@samp{y} or @samp{n} for replies, etc.  Even if the program will
eventually make most of its output in the foreign languages, one has
to decide whether the input syntax, option values, etc., are to be
localized or not.

@item
The manual accompanying a package, as well as all documentation files
in the distribution, could surely be translated, too.  Translating a
manual, with the intent of later keeping up with updates, is a major
undertaking in itself, generally.

@end itemize

As we already stressed, translation is only one aspect of locales.
Other internationalization aspects are system services and are handled
in GNU @code{libc}.  There
are many attributes that are needed to define a country's cultural
conventions.  These attributes include beside the country's native
language, the formatting of the date and time, the representation of
numbers, the symbols for currency, etc.  These local @dfn{rules} are
termed the country's locale.  The locale represents the knowledge
needed to support the country's native attributes.

@cindex locale categories
There are a few major areas which may vary between countries and
hence, define what a locale must describe.  The following list helps
putting multi-lingual messages into the proper context of other tasks
related to locales.  See the GNU @code{libc} manual for details.

@table @emph

@item Characters and Codesets
@cindex codeset
@cindex encoding
@cindex character encoding
@cindex locale category, LC_CTYPE

The codeset most commonly used through out the USA and most English
speaking parts of the world is the ASCII codeset.  However, there are
many characters needed by various locales that are not found within
this codeset.  The 8-bit @w{ISO 8859-1} code set has most of the special
characters needed to handle the major European languages.  However, in
many cases, choosing @w{ISO 8859-1} is nevertheless not adequate: it
doesn't even handle the major European currency.  Hence each locale
will need to specify which codeset they need to use and will need
to have the appropriate character handling routines to cope with
the codeset.

@item Currency
@cindex currency symbols
@cindex locale category, LC_MONETARY

The symbols used vary from country to country as does the position
used by the symbol.  Software needs to be able to transparently
display currency figures in the native mode for each locale.

@item Dates
@cindex date format
@cindex locale category, LC_TIME

The format of date varies between locales.  For example, Christmas day
in 1994 is written as 12/25/94 in the USA and as 25/12/94 in Australia.
Other countries might use @w{ISO 8601} dates, etc.

Time of the day may be noted as @var{hh}:@var{mm}, @var{hh}.@var{mm},
or otherwise.  Some locales require time to be specified in 24-hour
mode rather than as AM or PM.  Further, the nature and yearly extent
of the Daylight Saving correction vary widely between countries.

@item Numbers
@cindex number format
@cindex locale category, LC_NUMERIC

Numbers can be represented differently in different locales.
For example, the following numbers are all written correctly for
their respective locales:

@example
12,345.67       English
12.345,67       German
 12345,67       French
1,2345.67       Asia
@end example

Some programs could go further and use different unit systems, like
English units or Metric units, or even take into account variants
about how numbers are spelled in full.

@item Messages
@cindex messages
@cindex locale category, LC_MESSAGES

The most obvious area is the language support within a locale.  This is
where GNU @code{gettext} provides the means for developers and users to
easily change the language that the software uses to communicate to
the user.

@end table

@cindex locale categories
These areas of cultural conventions are called @emph{locale categories}.
It is an unfortunate term; @emph{locale aspects} or @emph{locale feature
categories} would be a better term, because each ``locale category''
describes an area or task that requires localization.  The concrete data
that describes the cultural conventions for such an area and for a particular
culture is also called a @emph{locale category}.  In this sense, a locale
is composed of several locale categories: the locale category describing
the codeset, the locale category describing the formatting of numbers,
the locale category containing the translated messages, and so on.

@cindex Linux
Components of locale outside of message handling are standardized in
the ISO C standard and the POSIX:2001 standard (also known as the SUSV3
specification).  GNU @code{libc}
fully implements this, and most other modern systems provide a more
or less reasonable support for at least some of the missing components.

@node Files, Overview, Aspects, Introduction
@section Files Conveying Translations

@cindex files, @file{.po} and @file{.mo}
The letters PO in @file{.po} files means Portable Object, to
distinguish it from @file{.mo} files, where MO stands for Machine
Object.  This paradigm, as well as the PO file format, is inspired
by the NLS standard developed by Uniforum, and first implemented by
Sun in their Solaris system.

PO files are meant to be read and edited by humans, and associate each
original, translatable string of a given package with its translation
in a particular target language.  A single PO file is dedicated to
a single target language.  If a package supports many languages,
there is one such PO file per language supported, and each package
has its own set of PO files.  These PO files are best created by
the @code{xgettext} program, and later updated or refreshed through
the @code{msgmerge} program.  Program @code{xgettext} extracts all
marked messages from a set of C files and initializes a PO file with
empty translations.  Program @code{msgmerge} takes care of adjusting
PO files between releases of the corresponding sources, commenting
obsolete entries, initializing new ones, and updating all source
line references.  Files ending with @file{.pot} are kind of base
translation files found in distributions, in PO file format.

MO files are meant to be read by programs, and are binary in nature.
A few systems already offer tools for creating and handling MO files
as part of the Native Language Support coming with the system, but the
format of these MO files is often different from system to system,
and non-portable.  The tools already provided with these systems don't
support all the features of GNU @code{gettext}.  Therefore GNU
@code{gettext} uses its own format for MO files.  Files ending with
@file{.gmo} are really MO files, when it is known that these files use
the GNU format.

@node Overview,  , Files, Introduction
@section Overview of GNU @code{gettext}

@cindex overview of @code{gettext}
@cindex big picture
@cindex tutorial of @code{gettext} usage
The following diagram summarizes the relation between the files
handled by GNU @code{gettext} and the tools acting on these files.
It is followed by somewhat detailed explanations, which you should
read while keeping an eye on the diagram.  Having a clear understanding
of these interrelations will surely help programmers, translators
and maintainers.

@ifhtml
@example
@group
Original C Sources ───> Preparation ───> Marked C Sources ───╮
                                                             │
              ╭─────────<─── GNU gettext Library             │
╭─── make <───┤                                              │
│             ╰─────────<────────────────────┬───────────────╯
│                                            │
│   ╭─────<─── PACKAGE.pot <─── xgettext <───╯   ╭───<─── PO Compendium
│   │                                            │              ↑
│   │                                            ╰───╮          │
│   ╰───╮                                            ├───> PO editor ───╮
│       ├────> msgmerge ──────> LANG.po ────>────────╯                  │
│   ╭───╯                                                               │
│   │                                                                   │
│   ╰─────────────<───────────────╮                                     │
│                                 ├─── New LANG.po <────────────────────╯
│   ╭─── LANG.gmo <─── msgfmt <───╯
│   │
│   ╰───> install ───> /.../LANG/PACKAGE.mo ───╮
│                                              ├───> "Hello world!"
╰───────> install ───> /.../bin/PROGRAM ───────╯
@end group
@end example
@end ifhtml
@ifnothtml
@example
@group
Original C Sources ---> Preparation ---> Marked C Sources ---.
                                                             |
              .---------<--- GNU gettext Library             |
.--- make <---+                                              |
|             `---------<--------------------+---------------'
|                                            |
|   .-----<--- PACKAGE.pot <--- xgettext <---'   .---<--- PO Compendium
|   |                                            |              ^
|   |                                            `---.          |
|   `---.                                            +---> PO editor ---.
|       +----> msgmerge ------> LANG.po ---->--------'                  |
|   .---'                                                               |
|   |                                                                   |
|   `-------------<---------------.                                     |
|                                 +--- New LANG.po <--------------------'
|   .--- LANG.gmo <--- msgfmt <---'
|   |
|   `---> install ---> /.../LANG/PACKAGE.mo ---.
|                                              +---> "Hello world!"
`-------> install ---> /.../bin/PROGRAM -------'
@end group
@end example
@end ifnothtml

@cindex marking translatable strings
As a programmer, the first step to bringing GNU @code{gettext}
into your package is identifying, right in the C sources, those strings
which are meant to be translatable, and those which are untranslatable.
This tedious job can be done a little more comfortably using emacs PO
mode, but you can use any means familiar to you for modifying your
C sources.  Beside this some other simple, standard changes are needed to
properly initialize the translation library.  @xref{Sources}, for
more information about all this.

For newly written software the strings of course can and should be
marked while writing it.  The @code{gettext} approach makes this
very easy.  Simply put the following lines at the beginning of each file
or in a central header file:

@example
@group
#define _(String) (String)
#define N_(String) String
#define textdomain(Domain)
#define bindtextdomain(Package, Directory)
@end group
@end example

@noindent
Doing this allows you to prepare the sources for internationalization.
Later when you feel ready for the step to use the @code{gettext} library
simply replace these definitions by the following:

@cindex include file @file{libintl.h}
@example
@group
#include <libintl.h>
#define _(String) gettext (String)
#define gettext_noop(String) String
#define N_(String) gettext_noop (String)
@end group
@end example

@cindex link with @file{libintl}
@cindex Linux
@noindent
and link against @file{libintl.a} or @file{libintl.so}.  Note that on
GNU systems, you don't need to link with @code{libintl} because the
@code{gettext} library functions are already contained in GNU libc.
That is all you have to change.

@cindex template PO file
@cindex files, @file{.pot}
Once the C sources have been modified, the @code{xgettext} program
is used to find and extract all translatable strings, and create a
PO template file out of all these.  This @file{@var{package}.pot} file
contains all original program strings.  It has sets of pointers to
exactly where in C sources each string is used.  All translations
are set to empty.  The letter @code{t} in @file{.pot} marks this as
a Template PO file, not yet oriented towards any particular language.
@xref{xgettext Invocation}, for more details about how one calls the
@code{xgettext} program.  If you are @emph{really} lazy, you might
be interested at working a lot more right away, and preparing the
whole distribution setup (@pxref{Maintainers}).  By doing so, you
spare yourself typing the @code{xgettext} command, as @code{make}
should now generate the proper things automatically for you!

The first time through, there is no @file{@var{lang}.po} yet, so the
@code{msgmerge} step may be skipped and replaced by a mere copy of
@file{@var{package}.pot} to @file{@var{lang}.po}, where @var{lang}
represents the target language.  See @ref{Creating} for details.

Then comes the initial translation of messages.  Translation in
itself is a whole matter, still exclusively meant for humans,
and whose complexity far overwhelms the level of this manual.
Nevertheless, a few hints are given in some other chapter of this
manual (@pxref{Translators}).  You will also find there indications
about how to contact translating teams, or becoming part of them,
for sharing your translating concerns with others who target the same
native language.

While adding the translated messages into the @file{@var{lang}.po}
PO file, if you are not using one of the dedicated PO file editors
(@pxref{Editing}), you are on your own
for ensuring that your efforts fully respect the PO file format, and quoting
conventions (@pxref{PO Files}).  This is surely not an impossible task,
as this is the way many people have handled PO files around 1995.
On the other hand, by using a PO file editor, most details
of PO file format are taken care of for you, but you have to acquire
some familiarity with PO file editor itself.

If some common translations have already been saved into a compendium
PO file, translators may use PO mode for initializing untranslated
entries from the compendium, and also save selected translations into
the compendium, updating it (@pxref{Compendium}).  Compendium files
are meant to be exchanged between members of a given translation team.

Programs, or packages of programs, are dynamic in nature: users write
bug reports and suggestion for improvements, maintainers react by
modifying programs in various ways.  The fact that a package has
already been internationalized should not make maintainers shy
of adding new strings, or modifying strings already translated.
They just do their job the best they can.  For the Translation
Project to work smoothly, it is important that maintainers do not
carry translation concerns on their already loaded shoulders, and that
translators be kept as free as possible of programming concerns.

The only concern maintainers should have is carefully marking new
strings as translatable, when they should be, and do not otherwise
worry about them being translated, as this will come in proper time.
Consequently, when programs and their strings are adjusted in various
ways by maintainers, and for matters usually unrelated to translation,
@code{xgettext} would construct @file{@var{package}.pot} files which are
evolving over time, so the translations carried by @file{@var{lang}.po}
are slowly fading out of date.

@cindex evolution of packages
It is important for translators (and even maintainers) to understand
that package translation is a continuous process in the lifetime of a
package, and not something which is done once and for all at the start.
After an initial burst of translation activity for a given package,
interventions are needed once in a while, because here and there,
translated entries become obsolete, and new untranslated entries
appear, needing translation.

The @code{msgmerge} program has the purpose of refreshing an already
existing @file{@var{lang}.po} file, by comparing it with a newer
@file{@var{package}.pot} template file, extracted by @code{xgettext}
out of recent C sources.  The refreshing operation adjusts all
references to C source locations for strings, since these strings
move as programs are modified.  Also, @code{msgmerge} comments out as
obsolete, in @file{@var{lang}.po}, those already translated entries
which are no longer used in the program sources (@pxref{Obsolete
Entries}).  It finally discovers new strings and inserts them in
the resulting PO file as untranslated entries (@pxref{Untranslated
Entries}).  @xref{msgmerge Invocation}, for more information about what
@code{msgmerge} really does.

Whatever route or means taken, the goal is to obtain an updated
@file{@var{lang}.po} file offering translations for all strings.

The temporal mobility, or fluidity of PO files, is an integral part of
the translation game, and should be well understood, and accepted.
People resisting it will have a hard time participating in the
Translation Project, or will give a hard time to other participants!  In
particular, maintainers should relax and include all available official
PO files in their distributions, even if these have not recently been
updated, without exerting pressure on the translator teams to get the
job done.  The pressure should rather come
from the community of users speaking a particular language, and
maintainers should consider themselves fairly relieved of any concern
about the adequacy of translation files.  On the other hand, translators
should reasonably try updating the PO files they are responsible for,
while the package is undergoing pretest, prior to an official
distribution.

Once the PO file is complete and dependable, the @code{msgfmt} program
is used for turning the PO file into a machine-oriented format, which
may yield efficient retrieval of translations by the programs of the
package, whenever needed at runtime (@pxref{MO Files}).  @xref{msgfmt
Invocation}, for more information about all modes of execution
for the @code{msgfmt} program.

Finally, the modified and marked C sources are compiled and linked
with the GNU @code{gettext} library, usually through the operation of
@code{make}, given a suitable @file{Makefile} exists for the project,
and the resulting executable is installed somewhere users will find it.
The MO files themselves should also be properly installed.  Given the
appropriate environment variables are set (@pxref{Setting the POSIX Locale}),
the program should localize itself automatically, whenever it executes.

The remainder of this manual has the purpose of explaining in depth the various
steps outlined above.

@node Users, PO Files, Introduction, Top
@chapter The User's View

Nowadays, when users log into a computer, they usually find that all
their programs show messages in their native language -- at least for
users of languages with an active free software community, like French or
German; to a lesser extent for languages with a smaller participation in
free software and the GNU project, like Hindi and Filipino.

How does this work?  How can the user influence the language that is used
by the programs?  This chapter will answer it.

@menu
* System Installation::         Questions During Operating System Installation
* Setting the GUI Locale::      How to Specify the Locale Used by GUI Programs
* Setting the POSIX Locale::    How to Specify the Locale According to POSIX
* Installing Localizations::    How to Install Additional Translations
@end menu

@node System Installation, Setting the GUI Locale, Users, Users
@section Operating System Installation

The default language is often already specified during operating system
installation.  When the operating system is installed, the installer
typically asks for the language used for the installation process and,
separately, for the language to use in the installed system.  Some OS
installers only ask for the language once.

This determines the system-wide default language for all users.  But the
installers often give the possibility to install extra localizations for
additional languages.  For example, the localizations of KDE (the K
Desktop Environment) and OpenOffice.org are often bundled separately,
as one installable package per language.

At this point it is good to consider the intended use of the machine: If
it is a machine designated for personal use, additional localizations are
probably not necessary.  If, however, the machine is in use in an
organization or company that has international relationships, one can
consider the needs of guest users.  If you have a guest from abroad, for
a week, what could be his preferred locales?  It may be worth installing
these additional localizations ahead of time, since they cost only a bit
of disk space at this point.

The system-wide default language is the locale configuration that is used
when a new user account is created.  But the user can have his own locale
configuration that is different from the one of the other users of the
same machine.  He can specify it, typically after the first login, as
described in the next section.

@node Setting the GUI Locale, Setting the POSIX Locale, System Installation, Users
@section Setting the Locale Used by GUI Programs

The immediately available programs in a user's desktop come from a group
of programs called a ``desktop environment''; it usually includes the window
manager, a web browser, a text editor, and more.  The most common free
desktop environments are KDE, GNOME, and Xfce.

The locale used by GUI programs of the desktop environment can be specified
in a configuration screen called ``control center'', ``language settings''
or ``country settings''.

Individual GUI programs that are not part of the desktop environment can
have their locale specified either in a settings panel, or through environment
variables.

For some programs, it is possible to specify the locale through environment
variables, possibly even to a different locale than the desktop's locale.
This means, instead of starting a program through a menu or from the file
system, you can start it from the command-line, after having set some
environment variables.  The environment variables can be those specified
in the next section (@ref{Setting the POSIX Locale}); for some versions of
KDE, however, the locale is specified through a variable @code{KDE_LANG},
rather than @code{LANG} or @code{LC_ALL}.

@node Setting the POSIX Locale, Installing Localizations, Setting the GUI Locale, Users
@section Setting the Locale through Environment Variables

As a user, if your language has been installed for this package, in the
simplest case, you only have to set the @code{LANG} environment variable
to the appropriate @samp{@var{ll}_@var{CC}} combination.  For example,
let's suppose that you speak German and live in Germany.  At the shell
prompt, merely execute 
@w{@samp{setenv LANG de_DE}} (in @code{csh}),
@w{@samp{export LANG; LANG=de_DE}} (in @code{sh}) or
@w{@samp{export LANG=de_DE}} (in @code{bash}).  This can be done from your
@file{.login} or @file{.profile} file, once and for all.

@menu
* Locale Names::                How a Locale Specification Looks Like
* Locale Environment Variables::  Which Environment Variable Specfies What
* The LANGUAGE variable::       How to Specify a Priority List of Languages
@end menu

@node Locale Names, Locale Environment Variables, Setting the POSIX Locale, Setting the POSIX Locale
@subsection Locale Names

A locale name usually has the form @samp{@var{ll}_@var{CC}}.  Here
@samp{@var{ll}} is an @w{ISO 639} two-letter language code, and
@samp{@var{CC}} is an @w{ISO 3166} two-letter country code.  For example,
for German in Germany, @var{ll} is @code{de}, and @var{CC} is @code{DE}.
You find a list of the language codes in appendix @ref{Language Codes} and
a list of the country codes in appendix @ref{Country Codes}.

You might think that the country code specification is redundant.  But in
fact, some languages have dialects in different countries.  For example,
@samp{de_AT} is used for Austria, and @samp{pt_BR} for Brazil.  The country
code serves to distinguish the dialects.

Many locale names have an extended syntax
@samp{@var{ll}_@var{CC}.@var{encoding}} that also specifies the character
encoding.  These are in use because between 2000 and 2005, most users have
switched to locales in UTF-8 encoding.  For example, the German locale on
glibc systems is nowadays @samp{de_DE.UTF-8}.  The older name @samp{de_DE}
still refers to the German locale as of 2000 that stores characters in
ISO-8859-1 encoding -- a text encoding that cannot even accommodate the Euro
currency sign.

Some locale names use @samp{@var{ll}_@var{CC}.@@@var{variant}} instead of
@samp{@var{ll}_@var{CC}}.  The @samp{@@@var{variant}} can denote any kind of
characteristics that is not already implied by the language @var{ll} and
the country @var{CC}.  It can denote a particular monetary unit.  For example,
on glibc systems, @samp{de_DE@@euro} denotes the locale that uses the Euro
currency, in contrast to the older locale @samp{de_DE} which implies the use
of the currency before 2002.  It can also denote a dialect of the language,
or the script used to write text (for example, @samp{sr_RS@@latin} uses the
Latin script, whereas @samp{sr_RS} uses the Cyrillic script to write Serbian),
or the orthography rules, or similar.

On other systems, some variations of this scheme are used, such as
@samp{@var{ll}}.  You can get the list of locales supported by your system
for your language by running the command @samp{locale -a | grep '^@var{ll}'}.

There is also a special locale, called @samp{C}.
@c Don't mention that this locale also has the name "POSIX". When we talk about
@c the "POSIX locale", we mean the "locale as specified in the POSIX way", and
@c mentioning a locale called "POSIX" would bring total confusion.
When it is used, it disables all localization: in this locale, all programs
standardized by POSIX use English messages and an unspecified character
encoding (often US-ASCII, but sometimes also ISO-8859-1 or UTF-8, depending on
the operating system).

@node Locale Environment Variables, The LANGUAGE variable, Locale Names, Setting the POSIX Locale
@subsection Locale Environment Variables
@cindex setting up @code{gettext} at run time
@cindex selecting message language
@cindex language selection

A locale is composed of several @emph{locale categories}, see @ref{Aspects}.
When a program looks up locale dependent values, it does this according to
the following environment variables, in priority order:

@enumerate
@vindex LANGUAGE@r{, environment variable}
@item @code{LANGUAGE}
@vindex LC_ALL@r{, environment variable}
@item @code{LC_ALL}
@vindex LC_CTYPE@r{, environment variable}
@vindex LC_NUMERIC@r{, environment variable}
@vindex LC_TIME@r{, environment variable}
@vindex LC_COLLATE@r{, environment variable}
@vindex LC_MONETARY@r{, environment variable}
@vindex LC_MESSAGES@r{, environment variable}
@item @code{LC_xxx}, according to selected locale category:
@code{LC_CTYPE}, @code{LC_NUMERIC}, @code{LC_TIME}, @code{LC_COLLATE},
@code{LC_MONETARY}, @code{LC_MESSAGES}, ...
@vindex LANG@r{, environment variable}
@item @code{LANG}
@end enumerate

Variables whose value is set but is empty are ignored in this lookup.

@code{LANG} is the normal environment variable for specifying a locale.
As a user, you normally set this variable (unless some of the other variables
have already been set by the system, in @file{/etc/profile} or similar
initialization files).

@code{LC_CTYPE}, @code{LC_NUMERIC}, @code{LC_TIME}, @code{LC_COLLATE},
@code{LC_MONETARY}, @code{LC_MESSAGES}, and so on, are the environment
variables meant to override @code{LANG} and affecting a single locale
category only.  For example, assume you are a Swedish user in Spain, and you
want your programs to handle numbers and dates according to Spanish
conventions, and only the messages should be in Swedish.  Then you could
create a locale named @samp{sv_ES} or @samp{sv_ES.UTF-8} by use of the
@code{localedef} program.  But it is simpler, and achieves the same effect,
to set the @code{LANG} variable to @code{es_ES.UTF-8} and the
@code{LC_MESSAGES} variable to @code{sv_SE.UTF-8}; these two locales come
already preinstalled with the operating system.

@code{LC_ALL} is an environment variable that overrides all of these.
It is typically used in scripts that run particular programs.  For example,
@code{configure} scripts generated by GNU autoconf use @code{LC_ALL} to make
sure that the configuration tests don't operate in locale dependent ways.

Some systems, unfortunately, set @code{LC_ALL} in @file{/etc/profile} or in
similar initialization files.  As a user, you therefore have to unset this
variable if you want to set @code{LANG} and optionally some of the other
@code{LC_xxx} variables.

The @code{LANGUAGE} variable is described in the next subsection.

@node The LANGUAGE variable,  , Locale Environment Variables, Setting the POSIX Locale
@subsection Specifying a Priority List of Languages

Not all programs have translations for all languages.  By default, an
English message is shown in place of a nonexistent translation.  If you
understand other languages, you can set up a priority list of languages.
This is done through a different environment variable, called
@code{LANGUAGE}.  GNU @code{gettext} gives preference to @code{LANGUAGE}
over @code{LC_ALL} and @code{LANG} for the purpose of message handling,
but you still need to have @code{LANG} (or @code{LC_ALL}) set to the primary
language; this is required by other parts of the system libraries.
For example, some Swedish users who would rather read translations in
German than English for when Swedish is not available, set @code{LANGUAGE}
to @samp{sv:de} while leaving @code{LANG} to @samp{sv_SE}.

Special advice for Norwegian users: The language code for Norwegian
bokm@ringaccent{a}l changed from @samp{no} to @samp{nb} recently (in 2003).
During the transition period, while some message catalogs for this language
are installed under @samp{nb} and some older ones under @samp{no}, it is
recommended for Norwegian users to set @code{LANGUAGE} to @samp{nb:no} so that
both newer and older translations are used.

In the @code{LANGUAGE} environment variable, but not in the other
environment variables, @samp{@var{ll}_@var{CC}} combinations can be
abbreviated as @samp{@var{ll}} to denote the language's main dialect.
For example, @samp{de} is equivalent to @samp{de_DE} (German as spoken in
Germany), and @samp{pt} to @samp{pt_PT} (Portuguese as spoken in Portugal)
in this context.

Note: The variable @code{LANGUAGE} is ignored if the locale is set to
@samp{C}.  In other words, you have to first enable localization, by setting
@code{LANG} (or @code{LC_ALL}) to a value other than @samp{C}, before you can
use a language priority list through the @code{LANGUAGE} variable.

@node Installing Localizations,  , Setting the POSIX Locale, Users
@section Installing Translations for Particular Programs
@cindex Translation Matrix
@cindex available translations

Languages are not equally well supported in all packages using GNU
@code{gettext}, and more translations are added over time.  Usually, you
use the translations that are shipped with the operating system
or with particular packages that you install afterwards.  But you can also
install newer localizations directly.  For doing this, you will need an
understanding where each localization file is stored on the file system.

@cindex @file{ABOUT-NLS} file
For programs that participate in the Translation Project, you can start
looking for translations here:
@url{http://translationproject.org/team/index.html}.
A snapshot of this information is also found in the @file{ABOUT-NLS} file
that is shipped with GNU gettext.

For programs that are part of the KDE project, the starting point is:
@url{http://i18n.kde.org/}.

For programs that are part of the GNOME project, the starting point is:
@url{http://www.gnome.org/i18n/}.

For other programs, you may check whether the program's source code package
contains some @file{@var{ll}.po} files; often they are kept together in a
directory called @file{po/}.  Each @file{@var{ll}.po} file contains the
message translations for the language whose abbreviation of @var{ll}.

@node PO Files, Sources, Users, Top
@chapter The Format of PO Files
@cindex PO files' format
@cindex file format, @file{.po}

The GNU @code{gettext} toolset helps programmers and translators
at producing, updating and using translation files, mainly those
PO files which are textual, editable files.  This chapter explains
the format of PO files.

A PO file is made up of many entries, each entry holding the relation
between an original untranslated string and its corresponding
translation.  All entries in a given PO file usually pertain
to a single project, and all translations are expressed in a single
target language.  One PO file @dfn{entry} has the following schematic
structure:

@example
@var{white-space}
#  @var{translator-comments}
#. @var{extracted-comments}
#: @var{reference}@dots{}
#, @var{flag}@dots{}
#| msgid @var{previous-untranslated-string}
msgid @var{untranslated-string}
msgstr @var{translated-string}
@end example

The general structure of a PO file should be well understood by
the translator.  When using PO mode, very little has to be known
about the format details, as PO mode takes care of them for her.

A simple entry can look like this:

@example
#: lib/error.c:116
msgid "Unknown system error"
msgstr "Error desconegut del sistema"
@end example

@cindex comments, translator
@cindex comments, automatic
@cindex comments, extracted
Entries begin with some optional white space.  Usually, when generated
through GNU @code{gettext} tools, there is exactly one blank line
between entries.  Then comments follow, on lines all starting with the
character @code{#}.  There are two kinds of comments: those which have
some white space immediately following the @code{#} - the @var{translator
comments} -, which comments are created and maintained exclusively by the
translator, and those which have some non-white character just after the
@code{#} - the @var{automatic comments} -, which comments are created and
maintained automatically by GNU @code{gettext} tools.  Comment lines
starting with @code{#.} contain comments given by the programmer, directed
at the translator; these comments are called @var{extracted comments}
because the @code{xgettext} program extracts them from the program's
source code.  Comment lines starting with @code{#:} contain references to
the program's source code.  Comment lines starting with @code{#,} contain
flags; more about these below.  Comment lines starting with @code{#|}
contain the previous untranslated string for which the translator gave
a translation.

All comments, of either kind, are optional.

@kwindex msgid
@kwindex msgstr
After white space and comments, entries show two strings, namely
first the untranslated string as it appears in the original program
sources, and then, the translation of this string.  The original
string is introduced by the keyword @code{msgid}, and the translation,
by @code{msgstr}.  The two strings, untranslated and translated,
are quoted in various ways in the PO file, using @code{"}
delimiters and @code{\} escapes, but the translator does not really
have to pay attention to the precise quoting format, as PO mode fully
takes care of quoting for her.

The @code{msgid} strings, as well as automatic comments, are produced
and managed by other GNU @code{gettext} tools, and PO mode does not
provide means for the translator to alter these.  The most she can
do is merely deleting them, and only by deleting the whole entry.
On the other hand, the @code{msgstr} string, as well as translator
comments, are really meant for the translator, and PO mode gives her
the full control she needs.

The comment lines beginning with @code{#,} are special because they are
not completely ignored by the programs as comments generally are.  The
comma separated list of @var{flag}s is used by the @code{msgfmt}
program to give the user some better diagnostic messages.  Currently
there are two forms of flags defined:

@table @code
@item fuzzy
@kwindex fuzzy@r{ flag}
This flag can be generated by the @code{msgmerge} program or it can be
inserted by the translator herself.  It shows that the @code{msgstr}
string might not be a correct translation (anymore).  Only the translator
can judge if the translation requires further modification, or is
acceptable as is.  Once satisfied with the translation, she then removes
this @code{fuzzy} attribute.  The @code{msgmerge} program inserts this
when it combined the @code{msgid} and @code{msgstr} entries after fuzzy
search only.  @xref{Fuzzy Entries}.

@item c-format
@kwindex c-format@r{ flag}
@itemx no-c-format
@kwindex no-c-format@r{ flag}
These flags should not be added by a human.  Instead only the
@code{xgettext} program adds them.  In an automated PO file processing
system as proposed here, the user's changes would be thrown away again as
soon as the @code{xgettext} program generates a new template file.

The @code{c-format} flag indicates that the untranslated string and the
translation are supposed to be C format strings.  The @code{no-c-format}
flag indicates that they are not C format strings, even though the untranslated
string happens to look like a C format string (with @samp{%} directives).

When the @code{c-format} flag is given for a string the @code{msgfmt}
program does some more tests to check the validity of the translation.
@xref{msgfmt Invocation}, @ref{c-format Flag} and @ref{c-format}.

@item objc-format
@kwindex objc-format@r{ flag}
@itemx no-objc-format
@kwindex no-objc-format@r{ flag}
Likewise for Objective C, see @ref{objc-format}.

@item sh-format
@kwindex sh-format@r{ flag}
@itemx no-sh-format
@kwindex no-sh-format@r{ flag}
Likewise for Shell, see @ref{sh-format}.

@item python-format
@kwindex python-format@r{ flag}
@itemx no-python-format
@kwindex no-python-format@r{ flag}
Likewise for Python, see @ref{python-format}.

@item python-brace-format
@kwindex python-brace-format@r{ flag}
@itemx no-python-brace-format
@kwindex no-python-brace-format@r{ flag}
Likewise for Python brace, see @ref{python-format}.

@item lisp-format
@kwindex lisp-format@r{ flag}
@itemx no-lisp-format
@kwindex no-lisp-format@r{ flag}
Likewise for Lisp, see @ref{lisp-format}.

@item elisp-format
@kwindex elisp-format@r{ flag}
@itemx no-elisp-format
@kwindex no-elisp-format@r{ flag}
Likewise for Emacs Lisp, see @ref{elisp-format}.

@item librep-format
@kwindex librep-format@r{ flag}
@itemx no-librep-format
@kwindex no-librep-format@r{ flag}
Likewise for librep, see @ref{librep-format}.

@item scheme-format
@kwindex scheme-format@r{ flag}
@itemx no-scheme-format
@kwindex no-scheme-format@r{ flag}
Likewise for Scheme, see @ref{scheme-format}.

@item smalltalk-format
@kwindex smalltalk-format@r{ flag}
@itemx no-smalltalk-format
@kwindex no-smalltalk-format@r{ flag}
Likewise for Smalltalk, see @ref{smalltalk-format}.

@item java-format
@kwindex java-format@r{ flag}
@itemx no-java-format
@kwindex no-java-format@r{ flag}
Likewise for Java, see @ref{java-format}.

@item csharp-format
@kwindex csharp-format@r{ flag}
@itemx no-csharp-format
@kwindex no-csharp-format@r{ flag}
Likewise for C#, see @ref{csharp-format}.

@item awk-format
@kwindex awk-format@r{ flag}
@itemx no-awk-format
@kwindex no-awk-format@r{ flag}
Likewise for awk, see @ref{awk-format}.

@item object-pascal-format
@kwindex object-pascal-format@r{ flag}
@itemx no-object-pascal-format
@kwindex no-object-pascal-format@r{ flag}
Likewise for Object Pascal, see @ref{object-pascal-format}.

@item ycp-format
@kwindex ycp-format@r{ flag}
@itemx no-ycp-format
@kwindex no-ycp-format@r{ flag}
Likewise for YCP, see @ref{ycp-format}.

@item tcl-format
@kwindex tcl-format@r{ flag}
@itemx no-tcl-format
@kwindex no-tcl-format@r{ flag}
Likewise for Tcl, see @ref{tcl-format}.

@item perl-format
@kwindex perl-format@r{ flag}
@itemx no-perl-format
@kwindex no-perl-format@r{ flag}
Likewise for Perl, see @ref{perl-format}.

@item perl-brace-format
@kwindex perl-brace-format@r{ flag}
@itemx no-perl-brace-format
@kwindex no-perl-brace-format@r{ flag}
Likewise for Perl brace, see @ref{perl-format}.

@item php-format
@kwindex php-format@r{ flag}
@itemx no-php-format
@kwindex no-php-format@r{ flag}
Likewise for PHP, see @ref{php-format}.

@item gcc-internal-format
@kwindex gcc-internal-format@r{ flag}
@itemx no-gcc-internal-format
@kwindex no-gcc-internal-format@r{ flag}
Likewise for the GCC sources, see @ref{gcc-internal-format}.

@item gfc-internal-format
@kwindex gfc-internal-format@r{ flag}
@itemx no-gfc-internal-format
@kwindex no-gfc-internal-format@r{ flag}
Likewise for the GNU Fortran Compiler sources, see @ref{gfc-internal-format}.

@item qt-format
@kwindex qt-format@r{ flag}
@itemx no-qt-format
@kwindex no-qt-format@r{ flag}
Likewise for Qt, see @ref{qt-format}.

@item qt-plural-format
@kwindex qt-plural-format@r{ flag}
@itemx no-qt-plural-format
@kwindex no-qt-plural-format@r{ flag}
Likewise for Qt plural forms, see @ref{qt-plural-format}.

@item kde-format
@kwindex kde-format@r{ flag}
@itemx no-kde-format
@kwindex no-kde-format@r{ flag}
Likewise for KDE, see @ref{kde-format}.

@item boost-format
@kwindex boost-format@r{ flag}
@itemx no-boost-format
@kwindex no-boost-format@r{ flag}
Likewise for Boost, see @ref{boost-format}.

@item lua-format
@kwindex lua-format@r{ flag}
@itemx no-lua-format
@kwindex no-lua-format@r{ flag}
Likewise for Lua, see @ref{lua-format}.

@item javascript-format
@kwindex javascript-format@r{ flag}
@itemx no-javascript-format
@kwindex no-javascript-format@r{ flag}
Likewise for JavaScript, see @ref{javascript-format}.

@end table

@kwindex msgctxt
@cindex context, in PO files
It is also possible to have entries with a context specifier. They look like
this:

@example
@var{white-space}
#  @var{translator-comments}
#. @var{extracted-comments}
#: @var{reference}@dots{}
#, @var{flag}@dots{}
#| msgctxt @var{previous-context}
#| msgid @var{previous-untranslated-string}
msgctxt @var{context}
msgid @var{untranslated-string}
msgstr @var{translated-string}
@end example

The context serves to disambiguate messages with the same
@var{untranslated-string}.  It is possible to have several entries with
the same @var{untranslated-string} in a PO file, provided that they each
have a different @var{context}.  Note that an empty @var{context} string
and an absent @code{msgctxt} line do not mean the same thing.

@kwindex msgid_plural
@cindex plural forms, in PO files
A different kind of entries is used for translations which involve
plural forms.

@example
@var{white-space}
#  @var{translator-comments}
#. @var{extracted-comments}
#: @var{reference}@dots{}
#, @var{flag}@dots{}
#| msgid @var{previous-untranslated-string-singular}
#| msgid_plural @var{previous-untranslated-string-plural}
msgid @var{untranslated-string-singular}
msgid_plural @var{untranslated-string-plural}
msgstr[0] @var{translated-string-case-0}
...
msgstr[N] @var{translated-string-case-n}
@end example

Such an entry can look like this:

@example
#: src/msgcmp.c:338 src/po-lex.c:699
#, c-format
msgid "found %d fatal error"
msgid_plural "found %d fatal errors"
msgstr[0] "s'ha trobat %d error fatal"
msgstr[1] "s'han trobat %d errors fatals"
@end example

Here also, a @code{msgctxt} context can be specified before @code{msgid},
like above.

Here, additional kinds of flags can be used:

@table @code
@item range:
@kwindex range:@r{ flag}
This flag is followed by a range of non-negative numbers, using the syntax
@code{range: @var{minimum-value}..@var{maximum-value}}.  It designates the
possible values that the numeric parameter of the message can take.  In some
languages, translators may produce slightly better translations if they know
that the value can only take on values between 0 and 10, for example.
@end table

The @var{previous-untranslated-string} is optionally inserted by the
@code{msgmerge} program, at the same time when it marks a message fuzzy.
It helps the translator to see which changes were done by the developers
on the @var{untranslated-string}.

It happens that some lines, usually whitespace or comments, follow the
very last entry of a PO file.  Such lines are not part of any entry,
and will be dropped when the PO file is processed by the tools, or may
disturb some PO file editors.

The remainder of this section may be safely skipped by those using
a PO file editor, yet it may be interesting for everybody to have a better
idea of the precise format of a PO file.  On the other hand, those
wishing to modify PO files by hand should carefully continue reading on.

An empty @var{untranslated-string} is reserved to contain the header
entry with the meta information (@pxref{Header Entry}).  This header
entry should be the first entry of the file.  The empty
@var{untranslated-string} is reserved for this purpose and must
not be used anywhere else.

Each of @var{untranslated-string} and @var{translated-string} respects
the C syntax for a character string, including the surrounding quotes
and embedded backslashed escape sequences.  When the time comes
to write multi-line strings, one should not use escaped newlines.
Instead, a closing quote should follow the last character on the
line to be continued, and an opening quote should resume the string
at the beginning of the following PO file line.  For example:

@example
msgid ""
"Here is an example of how one might continue a very long string\n"
"for the common case the string represents multi-line output.\n"
@end example

@noindent
In this example, the empty string is used on the first line, to
allow better alignment of the @code{H} from the word @samp{Here}
over the @code{f} from the word @samp{for}.  In this example, the
@code{msgid} keyword is followed by three strings, which are meant
to be concatenated.  Concatenating the empty string does not change
the resulting overall string, but it is a way for us to comply with
the necessity of @code{msgid} to be followed by a string on the same
line, while keeping the multi-line presentation left-justified, as
we find this to be a cleaner disposition.  The empty string could have
been omitted, but only if the string starting with @samp{Here} was
promoted on the first line, right after @code{msgid}.@footnote{This
limitation is not imposed by GNU @code{gettext}, but is for compatibility
with the @code{msgfmt} implementation on Solaris.} It was not really necessary
either to switch between the two last quoted strings immediately after
the newline @samp{\n}, the switch could have occurred after @emph{any}
other character, we just did it this way because it is neater.

@cindex newlines in PO files
One should carefully distinguish between end of lines marked as
@samp{\n} @emph{inside} quotes, which are part of the represented
string, and end of lines in the PO file itself, outside string quotes,
which have no incidence on the represented string.

@cindex comments in PO files
Outside strings, white lines and comments may be used freely.
Comments start at the beginning of a line with @samp{#} and extend
until the end of the PO file line.  Comments written by translators
should have the initial @samp{#} immediately followed by some white
space.  If the @samp{#} is not immediately followed by white space,
this comment is most likely generated and managed by specialized GNU
tools, and might disappear or be replaced unexpectedly when the PO
file is given to @code{msgmerge}.

@node Sources, Template, PO Files, Top
@chapter Preparing Program Sources
@cindex preparing programs for translation

@c FIXME: Rewrite (the whole chapter).

For the programmer, changes to the C source code fall into three
categories.  First, you have to make the localization functions
known to all modules needing message translation.  Second, you should
properly trigger the operation of GNU @code{gettext} when the program
initializes, usually from the @code{main} function.  Last, you should
identify, adjust and mark all constant strings in your program
needing translation.

@menu
* Importing::                   Importing the @code{gettext} declaration
* Triggering::                  Triggering @code{gettext} Operations
* Preparing Strings::           Preparing Translatable Strings
* Mark Keywords::               How Marks Appear in Sources
* Marking::                     Marking Translatable Strings
* c-format Flag::               Telling something about the following string
* Special cases::               Special Cases of Translatable Strings
* Bug Report Address::          Letting Users Report Translation Bugs
* Names::                       Marking Proper Names for Translation
* Libraries::                   Preparing Library Sources
@end menu

@node Importing, Triggering, Sources, Sources
@section Importing the @code{gettext} declaration

Presuming that your set of programs, or package, has been adjusted
so all needed GNU @code{gettext} files are available, and your
@file{Makefile} files are adjusted (@pxref{Maintainers}), each C module
having translated C strings should contain the line:

@cindex include file @file{libintl.h}
@example
#include <libintl.h>
@end example

Similarly, each C module containing @code{printf()}/@code{fprintf()}/...
calls with a format string that could be a translated C string (even if
the C string comes from a different C module) should contain the line:

@example
#include <libintl.h>
@end example

@node Triggering, Preparing Strings, Importing, Sources
@section Triggering @code{gettext} Operations

@cindex initialization
The initialization of locale data should be done with more or less
the same code in every program, as demonstrated below:

@example
@group
int
main (int argc, char *argv[])
@{
  @dots{}
  setlocale (LC_ALL, "");
  bindtextdomain (PACKAGE, LOCALEDIR);
  textdomain (PACKAGE);
  @dots{}
@}
@end group
@end example

@var{PACKAGE} and @var{LOCALEDIR} should be provided either by
@file{config.h} or by the Makefile.  For now consult the @code{gettext}
or @code{hello} sources for more information.

@cindex locale category, LC_ALL
@cindex locale category, LC_CTYPE
The use of @code{LC_ALL} might not be appropriate for you.
@code{LC_ALL} includes all locale categories and especially
@code{LC_CTYPE}.  This latter category is responsible for determining
character classes with the @code{isalnum} etc. functions from
@file{ctype.h} which could especially for programs, which process some
kind of input language, be wrong.  For example this would mean that a
source code using the @,{c} (c-cedilla character) is runnable in
France but not in the U.S.

Some systems also have problems with parsing numbers using the
@code{scanf} functions if an other but the @code{LC_ALL} locale category is
used.  The standards say that additional formats but the one known in the
@code{"C"} locale might be recognized.  But some systems seem to reject
numbers in the @code{"C"} locale format.  In some situation, it might
also be a problem with the notation itself which makes it impossible to
recognize whether the number is in the @code{"C"} locale or the local
format.  This can happen if thousands separator characters are used.
Some locales define this character according to the national
conventions to @code{'.'} which is the same character used in the
@code{"C"} locale to denote the decimal point.

So it is sometimes necessary to replace the @code{LC_ALL} line in the
code above by a sequence of @code{setlocale} lines

@example
@group
@{
  @dots{}
  setlocale (LC_CTYPE, "");
  setlocale (LC_MESSAGES, "");
  @dots{}
@}
@end group
@end example

@cindex locale category, LC_CTYPE
@cindex locale category, LC_COLLATE
@cindex locale category, LC_MONETARY
@cindex locale category, LC_NUMERIC
@cindex locale category, LC_TIME
@cindex locale category, LC_MESSAGES
@cindex locale category, LC_RESPONSES
@noindent
On all POSIX conformant systems the locale categories @code{LC_CTYPE},
@code{LC_MESSAGES}, @code{LC_COLLATE}, @code{LC_MONETARY},
@code{LC_NUMERIC}, and @code{LC_TIME} are available.  On some systems
which are only ISO C compliant, @code{LC_MESSAGES} is missing, but
a substitute for it is defined in GNU gettext's @code{<libintl.h>} and
in GNU gnulib's @code{<locale.h>}.

Note that changing the @code{LC_CTYPE} also affects the functions
declared in the @code{<ctype.h>} standard header and some functions
declared in the @code{<string.h>} and @code{<stdlib.h>} standard headers.
If this is not
desirable in your application (for example in a compiler's parser),
you can use a set of substitute functions which hardwire the C locale,
such as found in the modules @samp{c-ctype}, @samp{c-strcase},
@samp{c-strcasestr}, @samp{c-strtod}, @samp{c-strtold} in the GNU gnulib
source distribution.

It is also possible to switch the locale forth and back between the
environment dependent locale and the C locale, but this approach is
normally avoided because a @code{setlocale} call is expensive,
because it is tedious to determine the places where a locale switch
is needed in a large program's source, and because switching a locale
is not multithread-safe.

@node Preparing Strings, Mark Keywords, Triggering, Sources
@section Preparing Translatable Strings

@cindex marking strings, preparations
Before strings can be marked for translations, they sometimes need to
be adjusted.  Usually preparing a string for translation is done right
before marking it, during the marking phase which is described in the
next sections.  What you have to keep in mind while doing that is the
following.

@itemize @bullet
@item
Decent English style.

@item
Entire sentences.

@item
Split at paragraphs.

@item
Use format strings instead of string concatenation.

@item
Avoid unusual markup and unusual control characters.
@end itemize

@noindent
Let's look at some examples of these guidelines.

@cindex style
Translatable strings should be in good English style.  If slang language
with abbreviations and shortcuts is used, often translators will not
understand the message and will produce very inappropriate translations.

@example
"%s: is parameter\n"
@end example

@noindent
This is nearly untranslatable: Is the displayed item @emph{a} parameter or
@emph{the} parameter?

@example
"No match"
@end example

@noindent
The ambiguity in this message makes it unintelligible: Is the program
attempting to set something on fire? Does it mean "The given object does
not match the template"? Does it mean "The template does not fit for any
of the objects"?

@cindex ambiguities
In both cases, adding more words to the message will help both the
translator and the English speaking user.

@cindex sentences
Translatable strings should be entire sentences.  It is often not possible
to translate single verbs or adjectives in a substitutable way.

@example
printf ("File %s is %s protected", filename, rw ? "write" : "read");
@end example

@noindent
Most translators will not look at the source and will thus only see the
string @code{"File %s is %s protected"}, which is unintelligible.  Change
this to

@example
printf (rw ? "File %s is write protected" : "File %s is read protected",
        filename);
@end example

@noindent
This way the translator will not only understand the message, she will
also be able to find the appropriate grammatical construction.  A French
translator for example translates "write protected" like "protected
against writing".

Entire sentences are also important because in many languages, the
declination of some word in a sentence depends on the gender or the
number (singular/plural) of another part of the sentence.  There are
usually more interdependencies between words than in English.  The
consequence is that asking a translator to translate two half-sentences
and then combining these two half-sentences through dumb string concatenation
will not work, for many languages, even though it would work for English.
That's why translators need to handle entire sentences.

Often sentences don't fit into a single line.  If a sentence is output
using two subsequent @code{printf} statements, like this

@example
printf ("Locale charset \"%s\" is different from\n", lcharset);
printf ("input file charset \"%s\".\n", fcharset);
@end example

@noindent
the translator would have to translate two half sentences, but nothing
in the POT file would tell her that the two half sentences belong together.
It is necessary to merge the two @code{printf} statements so that the
translator can handle the entire sentence at once and decide at which
place to insert a line break in the translation (if at all):

@example
printf ("Locale charset \"%s\" is different from\n\
input file charset \"%s\".\n", lcharset, fcharset);
@end example

You may now ask: how about two or more adjacent sentences? Like in this case:

@example
puts ("Apollo 13 scenario: Stack overflow handling failed.");
puts ("On the next stack overflow we will crash!!!");
@end example

@noindent
Should these two statements merged into a single one? I would recommend to
merge them if the two sentences are related to each other, because then it
makes it easier for the translator to understand and translate both.  On
the other hand, if one of the two messages is a stereotypic one, occurring
in other places as well, you will do a favour to the translator by not
merging the two.  (Identical messages occurring in several places are
combined by xgettext, so the translator has to handle them once only.)

@cindex paragraphs
Translatable strings should be limited to one paragraph; don't let a
single message be longer than ten lines.  The reason is that when the
translatable string changes, the translator is faced with the task of
updating the entire translated string.  Maybe only a single word will
have changed in the English string, but the translator doesn't see that
(with the current translation tools), therefore she has to proofread
the entire message.

@cindex help option
Many GNU programs have a @samp{--help} output that extends over several
screen pages.  It is a courtesy towards the translators to split such a
message into several ones of five to ten lines each.  While doing that,
you can also attempt to split the documented options into groups,
such as the input options, the output options, and the informative
output options.  This will help every user to find the option he is
looking for.

@cindex string concatenation
@cindex concatenation of strings
Hardcoded string concatenation is sometimes used to construct English
strings:

@example
strcpy (s, "Replace ");
strcat (s, object1);
strcat (s, " with ");
strcat (s, object2);
strcat (s, "?");
@end example

@noindent
In order to present to the translator only entire sentences, and also
because in some languages the translator might want to swap the order
of @code{object1} and @code{object2}, it is necessary to change this
to use a format string:

@example
sprintf (s, "Replace %s with %s?", object1, object2);
@end example

@cindex @code{inttypes.h}
A similar case is compile time concatenation of strings.  The ISO C 99
include file @code{<inttypes.h>} contains a macro @code{PRId64} that
can be used as a formatting directive for outputting an @samp{int64_t}
integer through @code{printf}.  It expands to a constant string, usually
"d" or "ld" or "lld" or something like this, depending on the platform.
Assume you have code like

@example
printf ("The amount is %0" PRId64 "\n", number);
@end example

@noindent
The @code{gettext} tools and library have special support for these
@code{<inttypes.h>} macros.  You can therefore simply write

@example
printf (gettext ("The amount is %0" PRId64 "\n"), number);
@end example

@noindent
The PO file will contain the string "The amount is %0<PRId64>\n".
The translators will provide a translation containing "%0<PRId64>"
as well, and at runtime the @code{gettext} function's result will
contain the appropriate constant string, "d" or "ld" or "lld".

This works only for the predefined @code{<inttypes.h>} macros.  If
you have defined your own similar macros, let's say @samp{MYPRId64},
that are not known to @code{xgettext}, the solution for this problem
is to change the code like this:

@example
char buf1[100];
sprintf (buf1, "%0" MYPRId64, number);
printf (gettext ("The amount is %s\n"), buf1);
@end example

This means, you put the platform dependent code in one statement, and the
internationalization code in a different statement.  Note that a buffer length
of 100 is safe, because all available hardware integer types are limited to
128 bits, and to print a 128 bit integer one needs at most 54 characters,
regardless whether in decimal, octal or hexadecimal.

@cindex Java, string concatenation
@cindex C#, string concatenation
All this applies to other programming languages as well.  For example, in
Java and C#, string concatenation is very frequently used, because it is a
compiler built-in operator.  Like in C, in Java, you would change

@example
System.out.println("Replace "+object1+" with "+object2+"?");
@end example

@noindent
into a statement involving a format string:

@example
System.out.println(
    MessageFormat.format("Replace @{0@} with @{1@}?",
                         new Object[] @{ object1, object2 @}));
@end example

@noindent
Similarly, in C#, you would change

@example
Console.WriteLine("Replace "+object1+" with "+object2+"?");
@end example

@noindent
into a statement involving a format string:

@example
Console.WriteLine(
    String.Format("Replace @{0@} with @{1@}?", object1, object2));
@end example

@cindex markup
@cindex control characters
Unusual markup or control characters should not be used in translatable
strings.  Translators will likely not understand the particular meaning
of the markup or control characters.

For example, if you have a convention that @samp{|} delimits the
left-hand and right-hand part of some GUI elements, translators will
often not understand it without specific comments.  It might be
better to have the translator translate the left-hand and right-hand
part separately.

Another example is the @samp{argp} convention to use a single @samp{\v}
(vertical tab) control character to delimit two sections inside a
string.  This is flawed.  Some translators may convert it to a simple
newline, some to blank lines.  With some PO file editors it may not be
easy to even enter a vertical tab control character.  So, you cannot
be sure that the translation will contain a @samp{\v} character, at the
corresponding position.  The solution is, again, to let the translator
translate two separate strings and combine at run-time the two translated
strings with the @samp{\v} required by the convention.

HTML markup, however, is common enough that it's probably ok to use in
translatable strings.  But please bear in mind that the GNU gettext tools
don't verify that the translations are well-formed HTML.

@node Mark Keywords, Marking, Preparing Strings, Sources
@section How Marks Appear in Sources
@cindex marking strings that require translation

All strings requiring translation should be marked in the C sources.  Marking
is done in such a way that each translatable string appears to be
the sole argument of some function or preprocessor macro.  There are
only a few such possible functions or macros meant for translation,
and their names are said to be marking keywords.  The marking is
attached to strings themselves, rather than to what we do with them.
This approach has more uses.  A blatant example is an error message
produced by formatting.  The format string needs translation, as
well as some strings inserted through some @samp{%s} specification
in the format, while the result from @code{sprintf} may have so many
different instances that it is impractical to list them all in some
@samp{error_string_out()} routine, say.

This marking operation has two goals.  The first goal of marking
is for triggering the retrieval of the translation, at run time.
The keyword is possibly resolved into a routine able to dynamically
return the proper translation, as far as possible or wanted, for the
argument string.  Most localizable strings are found in executable
positions, that is, attached to variables or given as parameters to
functions.  But this is not universal usage, and some translatable
strings appear in structured initializations.  @xref{Special cases}.

The second goal of the marking operation is to help @code{xgettext}
at properly extracting all translatable strings when it scans a set
of program sources and produces PO file templates.

The canonical keyword for marking translatable strings is
@samp{gettext}, it gave its name to the whole GNU @code{gettext}
package.  For packages making only light use of the @samp{gettext}
keyword, macro or function, it is easily used @emph{as is}.  However,
for packages using the @code{gettext} interface more heavily, it
is usually more convenient to give the main keyword a shorter, less
obtrusive name.  Indeed, the keyword might appear on a lot of strings
all over the package, and programmers usually do not want nor need
their program sources to remind them forcefully, all the time, that they
are internationalized.  Further, a long keyword has the disadvantage
of using more horizontal space, forcing more indentation work on
sources for those trying to keep them within 79 or 80 columns.

@cindex @code{_}, a macro to mark strings for translation
Many packages use @samp{_} (a simple underline) as a keyword,
and write @samp{_("Translatable string")} instead of @samp{gettext
("Translatable string")}.  Further, the coding rule, from GNU standards,
wanting that there is a space between the keyword and the opening
parenthesis is relaxed, in practice, for this particular usage.
So, the textual overhead per translatable string is reduced to
only three characters: the underline and the two parentheses.
However, even if GNU @code{gettext} uses this convention internally,
it does not offer it officially.  The real, genuine keyword is truly
@samp{gettext} indeed.  It is fairly easy for those wanting to use
@samp{_} instead of @samp{gettext} to declare:

@example
#include <libintl.h>
#define _(String) gettext (String)
@end example

@noindent
instead of merely using @samp{#include <libintl.h>}.

The marking keywords @samp{gettext} and @samp{_} take the translatable
string as sole argument.  It is also possible to define marking functions
that take it at another argument position.  It is even possible to make
the marked argument position depend on the total number of arguments of
the function call; this is useful in C++.  All this is achieved using
@code{xgettext}'s @samp{--keyword} option.  How to pass such an option
to @code{xgettext}, assuming that @code{gettextize} is used, is described
in @ref{po/Makevars} and @ref{AM_XGETTEXT_OPTION}.

Note also that long strings can be split across lines, into multiple
adjacent string tokens.  Automatic string concatenation is performed
at compile time according to ISO C and ISO C++; @code{xgettext} also
supports this syntax.

Later on, the maintenance is relatively easy.  If, as a programmer,
you add or modify a string, you will have to ask yourself if the
new or altered string requires translation, and include it within
@samp{_()} if you think it should be translated.  For example, @samp{"%s"}
is an example of string @emph{not} requiring translation.  But
@samp{"%s: %d"} @emph{does} require translation, because in French, unlike
in English, it's customary to put a space before a colon.

@node Marking, c-format Flag, Mark Keywords, Sources
@section Marking Translatable Strings
@emindex marking strings for translation

In PO mode, one set of features is meant more for the programmer than
for the translator, and allows him to interactively mark which strings,
in a set of program sources, are translatable, and which are not.
Even if it is a fairly easy job for a programmer to find and mark
such strings by other means, using any editor of his choice, PO mode
makes this work more comfortable.  Further, this gives translators
who feel a little like programmers, or programmers who feel a little
like translators, a tool letting them work at marking translatable
strings in the program sources, while simultaneously producing a set of
translation in some language, for the package being internationalized.

@emindex @code{etags}, using for marking strings
The set of program sources, targeted by the PO mode commands describe
here, should have an Emacs tags table constructed for your project,
prior to using these PO file commands.  This is easy to do.  In any
shell window, change the directory to the root of your project, then
execute a command resembling:

@example
etags src/*.[hc] lib/*.[hc]
@end example

@noindent
presuming here you want to process all @file{.h} and @file{.c} files
from the @file{src/} and @file{lib/} directories.  This command will
explore all said files and create a @file{TAGS} file in your root
directory, somewhat summarizing the contents using a special file
format Emacs can understand.

@emindex @file{TAGS}, and marking translatable strings
For packages following the GNU coding standards, there is
a make goal @code{tags} or @code{TAGS} which constructs the tag files in
all directories and for all files containing source code.

Once your @file{TAGS} file is ready, the following commands assist
the programmer at marking translatable strings in his set of sources.
But these commands are necessarily driven from within a PO file
window, and it is likely that you do not even have such a PO file yet.
This is not a problem at all, as you may safely open a new, empty PO
file, mainly for using these commands.  This empty PO file will slowly
fill in while you mark strings as translatable in your program sources.

@table @kbd
@item ,
@efindex ,@r{, PO Mode command}
Search through program sources for a string which looks like a
candidate for translation (@code{po-tags-search}).

@item M-,
@efindex M-,@r{, PO Mode command}
Mark the last string found with @samp{_()} (@code{po-mark-translatable}).

@item M-.
@efindex M-.@r{, PO Mode command}
Mark the last string found with a keyword taken from a set of possible
keywords.  This command with a prefix allows some management of these
keywords (@code{po-select-mark-and-mark}).

@end table

@efindex po-tags-search@r{, PO Mode command}
The @kbd{,} (@code{po-tags-search}) command searches for the next
occurrence of a string which looks like a possible candidate for
translation, and displays the program source in another Emacs window,
positioned in such a way that the string is near the top of this other
window.  If the string is too big to fit whole in this window, it is
positioned so only its end is shown.  In any case, the cursor
is left in the PO file window.  If the shown string would be better
presented differently in different native languages, you may mark it
using @kbd{M-,} or @kbd{M-.}.  Otherwise, you might rather ignore it
and skip to the next string by merely repeating the @kbd{,} command.

A string is a good candidate for translation if it contains a sequence
of three or more letters.  A string containing at most two letters in
a row will be considered as a candidate if it has more letters than
non-letters.  The command disregards strings containing no letters,
or isolated letters only.  It also disregards strings within comments,
or strings already marked with some keyword PO mode knows (see below).

If you have never told Emacs about some @file{TAGS} file to use, the
command will request that you specify one from the minibuffer, the
first time you use the command.  You may later change your @file{TAGS}
file by using the regular Emacs command @w{@kbd{M-x visit-tags-table}},
which will ask you to name the precise @file{TAGS} file you want
to use.  @xref{Tags, , Tag Tables, emacs, The Emacs Editor}.

Each time you use the @kbd{,} command, the search resumes from where it was
left by the previous search, and goes through all program sources,
obeying the @file{TAGS} file, until all sources have been processed.
However, by giving a prefix argument to the command @w{(@kbd{C-u
,})}, you may request that the search be restarted all over again
from the first program source; but in this case, strings that you
recently marked as translatable will be automatically skipped.

Using this @kbd{,} command does not prevent using of other regular
Emacs tags commands.  For example, regular @code{tags-search} or
@code{tags-query-replace} commands may be used without disrupting the
independent @kbd{,} search sequence.  However, as implemented, the
@emph{initial} @kbd{,} command (or the @kbd{,} command is used with a
prefix) might also reinitialize the regular Emacs tags searching to the
first tags file, this reinitialization might be considered spurious.

@efindex po-mark-translatable@r{, PO Mode command}
@efindex po-select-mark-and-mark@r{, PO Mode command}
The @kbd{M-,} (@code{po-mark-translatable}) command will mark the
recently found string with the @samp{_} keyword.  The @kbd{M-.}
(@code{po-select-mark-and-mark}) command will request that you type
one keyword from the minibuffer and use that keyword for marking
the string.  Both commands will automatically create a new PO file
untranslated entry for the string being marked, and make it the
current entry (making it easy for you to immediately proceed to its
translation, if you feel like doing it right away).  It is possible
that the modifications made to the program source by @kbd{M-,} or
@kbd{M-.} render some source line longer than 80 columns, forcing you
to break and re-indent this line differently.  You may use the @kbd{O}
command from PO mode, or any other window changing command from
Emacs, to break out into the program source window, and do any
needed adjustments.  You will have to use some regular Emacs command
to return the cursor to the PO file window, if you want command
@kbd{,} for the next string, say.

The @kbd{M-.} command has a few built-in speedups, so you do not
have to explicitly type all keywords all the time.  The first such
speedup is that you are presented with a @emph{preferred} keyword,
which you may accept by merely typing @kbd{@key{RET}} at the prompt.
The second speedup is that you may type any non-ambiguous prefix of the
keyword you really mean, and the command will complete it automatically
for you.  This also means that PO mode has to @emph{know} all
your possible keywords, and that it will not accept mistyped keywords.

If you reply @kbd{?} to the keyword request, the command gives a
list of all known keywords, from which you may choose.  When the
command is prefixed by an argument @w{(@kbd{C-u M-.})}, it inhibits
updating any program source or PO file buffer, and does some simple
keyword management instead.  In this case, the command asks for a
keyword, written in full, which becomes a new allowed keyword for
later @kbd{M-.} commands.  Moreover, this new keyword automatically
becomes the @emph{preferred} keyword for later commands.  By typing
an already known keyword in response to @w{@kbd{C-u M-.}}, one merely
changes the @emph{preferred} keyword and does nothing more.

All keywords known for @kbd{M-.} are recognized by the @kbd{,} command
when scanning for strings, and strings already marked by any of those
known keywords are automatically skipped.  If many PO files are opened
simultaneously, each one has its own independent set of known keywords.
There is no provision in PO mode, currently, for deleting a known
keyword, you have to quit the file (maybe using @kbd{q}) and reopen
it afresh.  When a PO file is newly brought up in an Emacs window, only
@samp{gettext} and @samp{_} are known as keywords, and @samp{gettext}
is preferred for the @kbd{M-.} command.  In fact, this is not useful to
prefer @samp{_}, as this one is already built in the @kbd{M-,} command.

@node c-format Flag, Special cases, Marking, Sources
@section Special Comments preceding Keywords

@c FIXME document c-format and no-c-format.

@cindex format strings
In C programs strings are often used within calls of functions from the
@code{printf} family.  The special thing about these format strings is
that they can contain format specifiers introduced with @kbd{%}.  Assume
we have the code

@example
printf (gettext ("String `%s' has %d characters\n"), s, strlen (s));
@end example

@noindent
A possible German translation for the above string might be:

@example
"%d Zeichen lang ist die Zeichenkette `%s'"
@end example

A C programmer, even if he cannot speak German, will recognize that
there is something wrong here.  The order of the two format specifiers
is changed but of course the arguments in the @code{printf} don't have.
This will most probably lead to problems because now the length of the
string is regarded as the address.

To prevent errors at runtime caused by translations, the @code{msgfmt}
tool can check statically whether the arguments in the original and the
translation string match in type and number.  If this is not the case
and the @samp{-c} option has been passed to @code{msgfmt}, @code{msgfmt}
will give an error and refuse to produce a MO file.  Thus consistent
use of @samp{msgfmt -c} will catch the error, so that it cannot cause
problems at runtime.

@noindent
If the word order in the above German translation would be correct one
would have to write

@example
"%2$d Zeichen lang ist die Zeichenkette `%1$s'"
@end example

@noindent
The routines in @code{msgfmt} know about this special notation.

Because not all strings in a program will be format strings, it is not
useful for @code{msgfmt} to test all the strings in the @file{.po} file.
This might cause problems because the string might contain what looks
like a format specifier, but the string is not used in @code{printf}.

Therefore @code{xgettext} adds a special tag to those messages it
thinks might be a format string.  There is no absolute rule for this,
only a heuristic.  In the @file{.po} file the entry is marked using the
@code{c-format} flag in the @code{#,} comment line (@pxref{PO Files}).

@kwindex c-format@r{, and @code{xgettext}}
@kwindex no-c-format@r{, and @code{xgettext}}
The careful reader now might say that this again can cause problems.
The heuristic might guess it wrong.  This is true and therefore
@code{xgettext} knows about a special kind of comment which lets
the programmer take over the decision.  If in the same line as or
the immediately preceding line to the @code{gettext} keyword
the @code{xgettext} program finds a comment containing the words
@code{xgettext:c-format}, it will mark the string in any case with
the @code{c-format} flag.  This kind of comment should be used when
@code{xgettext} does not recognize the string as a format string but
it really is one and it should be tested.  Please note that when the
comment is in the same line as the @code{gettext} keyword, it must be
before the string to be translated.

This situation happens quite often.  The @code{printf} function is often
called with strings which do not contain a format specifier.  Of course
one would normally use @code{fputs} but it does happen.  In this case
@code{xgettext} does not recognize this as a format string but what
happens if the translation introduces a valid format specifier?  The
@code{printf} function will try to access one of the parameters but none
exists because the original code does not pass any parameters.

@code{xgettext} of course could make a wrong decision the other way
round, i.e.@: a string marked as a format string actually is not a format
string.  In this case the @code{msgfmt} might give too many warnings and
would prevent translating the @file{.po} file.  The method to prevent
this wrong decision is similar to the one used above, only the comment
to use must contain the string @code{xgettext:no-c-format}.

If a string is marked with @code{c-format} and this is not correct the
user can find out who is responsible for the decision.  See
@ref{xgettext Invocation} to see how the @code{--debug} option can be
used for solving this problem.

@node Special cases, Bug Report Address, c-format Flag, Sources
@section Special Cases of Translatable Strings

@cindex marking string initializers
The attentive reader might now point out that it is not always possible
to mark translatable string with @code{gettext} or something like this.
Consider the following case:

@example
@group
@{
  static const char *messages[] = @{
    "some very meaningful message",
    "and another one"
  @};
  const char *string;
  @dots{}
  string
    = index > 1 ? "a default message" : messages[index];

  fputs (string);
  @dots{}
@}
@end group
@end example

While it is no problem to mark the string @code{"a default message"} it
is not possible to mark the string initializers for @code{messages}.
What is to be done?  We have to fulfill two tasks.  First we have to mark the
strings so that the @code{xgettext} program (@pxref{xgettext Invocation})
can find them, and second we have to translate the string at runtime
before printing them.

The first task can be fulfilled by creating a new keyword, which names a
no-op.  For the second we have to mark all access points to a string
from the array.  So one solution can look like this:

@example
@group
#define gettext_noop(String) String

@{
  static const char *messages[] = @{
    gettext_noop ("some very meaningful message"),
    gettext_noop ("and another one")
  @};
  const char *string;
  @dots{}
  string
    = index > 1 ? gettext ("a default message") : gettext (messages[index]);

  fputs (string);
  @dots{}
@}
@end group
@end example

Please convince yourself that the string which is written by
@code{fputs} is translated in any case.  How to get @code{xgettext} know
the additional keyword @code{gettext_noop} is explained in @ref{xgettext
Invocation}.

The above is of course not the only solution.  You could also come along
with the following one:

@example
@group
#define gettext_noop(String) String

@{
  static const char *messages[] = @{
    gettext_noop ("some very meaningful message"),
    gettext_noop ("and another one")
  @};
  const char *string;
  @dots{}
  string
    = index > 1 ? gettext_noop ("a default message") : messages[index];

  fputs (gettext (string));
  @dots{}
@}
@end group
@end example

But this has a drawback.  The programmer has to take care that
he uses @code{gettext_noop} for the string @code{"a default message"}.
A use of @code{gettext} could have in rare cases unpredictable results.

One advantage is that you need not make control flow analysis to make
sure the output is really translated in any case.  But this analysis is
generally not very difficult.  If it should be in any situation you can
use this second method in this situation.

@node Bug Report Address, Names, Special cases, Sources
@section Letting Users Report Translation Bugs

Code sometimes has bugs, but translations sometimes have bugs too.  The
users need to be able to report them.  Reporting translation bugs to the
programmer or maintainer of a package is not very useful, since the
maintainer must never change a translation, except on behalf of the
translator.  Hence the translation bugs must be reported to the
translators.

Here is a way to organize this so that the maintainer does not need to
forward translation bug reports, nor even keep a list of the addresses of
the translators or their translation teams.

Every program has a place where is shows the bug report address.  For
GNU programs, it is the code which handles the ``--help'' option,
typically in a function called ``usage''.  In this place, instruct the
translator to add her own bug reporting address.  For example, if that
code has a statement

@example
@group
printf (_("Report bugs to <%s>.\n"), PACKAGE_BUGREPORT);
@end group
@end example

you can add some translator instructions like this:

@example
@group
/* TRANSLATORS: The placeholder indicates the bug-reporting address
   for this package.  Please add _another line_ saying
   "Report translation bugs to <...>\n" with the address for translation
   bugs (typically your translation team's web or email address).  */
printf (_("Report bugs to <%s>.\n"), PACKAGE_BUGREPORT);
@end group
@end example

These will be extracted by @samp{xgettext}, leading to a .pot file that
contains this:

@example
@group
#. TRANSLATORS: The placeholder indicates the bug-reporting address
#. for this package.  Please add _another line_ saying
#. "Report translation bugs to <...>\n" with the address for translation
#. bugs (typically your translation team's web or email address).
#: src/hello.c:178
#, c-format
msgid "Report bugs to <%s>.\n"
msgstr ""
@end group
@end example

@node Names, Libraries, Bug Report Address, Sources
@section Marking Proper Names for Translation

Should names of persons, cities, locations etc. be marked for translation
or not?  People who only know languages that can be written with Latin
letters (English, Spanish, French, German, etc.) are tempted to say ``no'',
because names usually do not change when transported between these languages.
However, in general when translating from one script to another, names
are translated too, usually phonetically or by transliteration.  For
example, Russian or Greek names are converted to the Latin alphabet when
being translated to English, and English or French names are converted
to the Katakana script when being translated to Japanese.  This is
necessary because the speakers of the target language in general cannot
read the script the name is originally written in.

As a programmer, you should therefore make sure that names are marked
for translation, with a special comment telling the translators that it
is a proper name and how to pronounce it.  In its simple form, it looks
like this:

@example
@group
printf (_("Written by %s.\n"),
        /* TRANSLATORS: This is a proper name.  See the gettext
           manual, section Names.  Note this is actually a non-ASCII
           name: The first name is (with Unicode escapes)
           "Fran\u00e7ois" or (with HTML entities) "Fran&ccedil;ois".
           Pronunciation is like "fraa-swa pee-nar".  */
        _("Francois Pinard"));
@end group
@end example

@noindent
The GNU gnulib library offers a module @samp{propername}
(@url{http://www.gnu.org/software/gnulib/MODULES.html#module=propername})
which takes care to automatically append the original name, in parentheses,
to the translated name.  For names that cannot be written in ASCII, it
also frees the translator from the task of entering the appropriate non-ASCII
characters if no script change is needed.  In this more comfortable form,
it looks like this:

@example
@group
printf (_("Written by %s and %s.\n"),
        proper_name ("Ulrich Drepper"),
        /* TRANSLATORS: This is a proper name.  See the gettext
           manual, section Names.  Note this is actually a non-ASCII
           name: The first name is (with Unicode escapes)
           "Fran\u00e7ois" or (with HTML entities) "Fran&ccedil;ois".
           Pronunciation is like "fraa-swa pee-nar".  */
        proper_name_utf8 ("Francois Pinard", "Fran\303\247ois Pinard"));
@end group
@end example

@noindent
You can also write the original name directly in Unicode (rather than with
Unicode escapes or HTML entities) and denote the pronunciation using the
International Phonetic Alphabet (see
@url{http://www.wikipedia.org/wiki/International_Phonetic_Alphabet}).

As a translator, you should use some care when translating names, because
it is frustrating if people see their names mutilated or distorted.

If your language uses the Latin script, all you need to do is to reproduce
the name as perfectly as you can within the usual character set of your
language.  In this particular case, this means to provide a translation
containing the c-cedilla character.  If your language uses a different
script and the people speaking it don't usually read Latin words, it means
transliteration.  If the programmer used the simple case, you should still
give, in parentheses, the original writing of the name -- for the sake of
the people that do read the Latin script.  If the programmer used the
@samp{propername} module mentioned above, you don't need to give the original
writing of the name in parentheses, because the program will already do so.
Here is an example, using Greek as the target script:

@example
@group
#. This is a proper name.  See the gettext
#. manual, section Names.  Note this is actually a non-ASCII
#. name: The first name is (with Unicode escapes)
#. "Fran\u00e7ois" or (with HTML entities) "Fran&ccedil;ois".
#. Pronunciation is like "fraa-swa pee-nar".
msgid "Francois Pinard"
msgstr "\phi\rho\alpha\sigma\omicron\alpha \pi\iota\nu\alpha\rho"
       " (Francois Pinard)"
@end group
@end example

Because translation of names is such a sensitive domain, it is a good
idea to test your translation before submitting it.

@node Libraries,  , Names, Sources
@section Preparing Library Sources

When you are preparing a library, not a program, for the use of
@code{gettext}, only a few details are different.  Here we assume that
the library has a translation domain and a POT file of its own.  (If
it uses the translation domain and POT file of the main program, then
the previous sections apply without changes.)

@enumerate
@item
The library code doesn't call @code{setlocale (LC_ALL, "")}.  It's the
responsibility of the main program to set the locale.  The library's
documentation should mention this fact, so that developers of programs
using the library are aware of it.

@item
The library code doesn't call @code{textdomain (PACKAGE)}, because it
would interfere with the text domain set by the main program.

@item
The initialization code for a program was

@smallexample
  setlocale (LC_ALL, "");
  bindtextdomain (PACKAGE, LOCALEDIR);
  textdomain (PACKAGE);
@end smallexample

@noindent
For a library it is reduced to

@smallexample
  bindtextdomain (PACKAGE, LOCALEDIR);
@end smallexample

@noindent
If your library's API doesn't already have an initialization function,
you need to create one, containing at least the @code{bindtextdomain}
invocation.  However, you usually don't need to export and document this
initialization function: It is sufficient that all entry points of the
library call the initialization function if it hasn't been called before.
The typical idiom used to achieve this is a static boolean variable that
indicates whether the initialization function has been called. Like this:

@example
@group
static bool libfoo_initialized;

static void
libfoo_initialize (void)
@{
  bindtextdomain (PACKAGE, LOCALEDIR);
  libfoo_initialized = true;
@}

/* This function is part of the exported API.  */
struct foo *
create_foo (...)
@{
  /* Must ensure the initialization is performed.  */
  if (!libfoo_initialized)
    libfoo_initialize ();
  ...
@}

/* This function is part of the exported API.  The argument must be
   non-NULL and have been created through create_foo().  */
int
foo_refcount (struct foo *argument)
@{
  /* No need to invoke the initialization function here, because
     create_foo() must already have been called before.  */
  ...
@}
@end group
@end example

@item
The usual declaration of the @samp{_} macro in each source file was

@smallexample
#include <libintl.h>
#define _(String) gettext (String)
@end smallexample

@noindent
for a program.  For a library, which has its own translation domain,
it reads like this:

@smallexample
#include <libintl.h>
#define _(String) dgettext (PACKAGE, String)
@end smallexample

In other words, @code{dgettext} is used instead of @code{gettext}.
Similarly, the @code{dngettext} function should be used in place of the
@code{ngettext} function.
@end enumerate

@node Template, Creating, Sources, Top
@chapter Making the PO Template File
@cindex PO template file

After preparing the sources, the programmer creates a PO template file.
This section explains how to use @code{xgettext} for this purpose.

@code{xgettext} creates a file named @file{@var{domainname}.po}.  You
should then rename it to @file{@var{domainname}.pot}.  (Why doesn't
@code{xgettext} create it under the name @file{@var{domainname}.pot}
right away?  The answer is: for historical reasons.  When @code{xgettext}
was specified, the distinction between a PO file and PO file template
was fuzzy, and the suffix @samp{.pot} wasn't in use at that time.)

@c FIXME: Rewrite.

@menu
* xgettext Invocation::         Invoking the @code{xgettext} Program
@end menu

@node xgettext Invocation,  , Template, Template
@section Invoking the @code{xgettext} Program

@include xgettext.texi

@node Creating, Updating, Template, Top
@chapter Creating a New PO File
@cindex creating a new PO file

When starting a new translation, the translator creates a file called
@file{@var{LANG}.po}, as a copy of the @file{@var{package}.pot} template
file with modifications in the initial comments (at the beginning of the file)
and in the header entry (the first entry, near the beginning of the file).

The easiest way to do so is by use of the @samp{msginit} program.
For example:

@example
$ cd @var{PACKAGE}-@var{VERSION}
$ cd po
$ msginit
@end example

The alternative way is to do the copy and modifications by hand.
To do so, the translator copies @file{@var{package}.pot} to
@file{@var{LANG}.po}.  Then she modifies the initial comments and
the header entry of this file.

@menu
* msginit Invocation::          Invoking the @code{msginit} Program
* Header Entry::                Filling in the Header Entry
@end menu

@node msginit Invocation, Header Entry, Creating, Creating
@section Invoking the @code{msginit} Program

@include msginit.texi

@node Header Entry,  , msginit Invocation, Creating
@section Filling in the Header Entry
@cindex header entry of a PO file

The initial comments "SOME DESCRIPTIVE TITLE", "YEAR" and
"FIRST AUTHOR <EMAIL@@ADDRESS>, YEAR" ought to be replaced by sensible
information.  This can be done in any text editor; if Emacs is used
and it switched to PO mode automatically (because it has recognized
the file's suffix), you can disable it by typing @kbd{M-x fundamental-mode}.

Modifying the header entry can already be done using PO mode: in Emacs,
type @kbd{M-x po-mode RET} and then @kbd{RET} again to start editing the
entry.  You should fill in the following fields.

@table @asis
@item Project-Id-Version
This is the name and version of the package.  Fill it in if it has not
already been filled in by @code{xgettext}.

@item Report-Msgid-Bugs-To
This has already been filled in by @code{xgettext}.  It contains an email
address or URL where you can report bugs in the untranslated strings:

@itemize -
@item Strings which are not entire sentences, see the maintainer guidelines
in @ref{Preparing Strings}.
@item Strings which use unclear terms or require additional context to be
understood.
@item Strings which make invalid assumptions about notation of date, time or
money.
@item Pluralisation problems.
@item Incorrect English spelling.
@item Incorrect formatting.
@end itemize

@item POT-Creation-Date
This has already been filled in by @code{xgettext}.

@item PO-Revision-Date
You don't need to fill this in.  It will be filled by the PO file editor
when you save the file.

@item Last-Translator
Fill in your name and email address (without double quotes).

@item Language-Team
Fill in the English name of the language, and the email address or
homepage URL of the language team you are part of.

Before starting a translation, it is a good idea to get in touch with
your translation team, not only to make sure you don't do duplicated work,
but also to coordinate difficult linguistic issues.

@cindex list of translation teams, where to find
In the Free Translation Project, each translation team has its own mailing
list.  The up-to-date list of teams can be found at the Free Translation
Project's homepage, @uref{http://translationproject.org/}, in the "Teams"
area.

@item Language
@c The purpose of this field is to make it possible to automatically
@c - convert PO files to translation memory,
@c - initialize a spell checker based on the PO file,
@c - perform language specific checks.
Fill in the language code of the language.  This can be in one of three
forms:

@itemize -
@item
@samp{@var{ll}}, an @w{ISO 639} two-letter language code (lowercase).
See @ref{Language Codes} for the list of codes.

@item
@samp{@var{ll}_@var{CC}}, where @samp{@var{ll}} is an @w{ISO 639} two-letter
language code (lowercase) and @samp{@var{CC}} is an @w{ISO 3166} two-letter
country code (uppercase).  The country code specification is not redundant:
Some languages have dialects in different countries.  For example,
@samp{de_AT} is used for Austria, and @samp{pt_BR} for Brazil.  The country
code serves to distinguish the dialects. See @ref{Language Codes} and
@ref{Country Codes} for the lists of codes.

@item
@samp{@var{ll}_@var{CC}@@@var{variant}}, where @samp{@var{ll}} is an
@w{ISO 639} two-letter language code (lowercase), @samp{@var{CC}} is an
@w{ISO 3166} two-letter country code (uppercase), and @samp{@var{variant}} is
a variant designator. The variant designator (lowercase) can be a script
designator, such as @samp{latin} or @samp{cyrillic}.
@end itemize

The naming convention @samp{@var{ll}_@var{CC}} is also the way locales are
named on systems based on GNU libc.  But there are three important differences:

@itemize @bullet
@item
In this PO file field, but not in locale names, @samp{@var{ll}_@var{CC}}
combinations denoting a language's main dialect are abbreviated as
@samp{@var{ll}}.  For example, @samp{de} is equivalent to @samp{de_DE}
(German as spoken in Germany), and @samp{pt} to @samp{pt_PT} (Portuguese as
spoken in Portugal) in this context.

@item
In this PO file field, suffixes like @samp{.@var{encoding}} are not used.

@item
In this PO file field, variant designators that are not relevant to message
translation, such as @samp{@@euro}, are not used.
@end itemize

So, if your locale name is @samp{de_DE.UTF-8}, the language specification in
PO files is just @samp{de}.

@item Content-Type
@cindex encoding of PO files
@cindex charset of PO files
Replace @samp{CHARSET} with the character encoding used for your language,
in your locale, or UTF-8.  This field is needed for correct operation of the
@code{msgmerge} and @code{msgfmt} programs, as well as for users whose
locale's character encoding differs from yours (see @ref{Charset conversion}).

@cindex @code{locale} program
You get the character encoding of your locale by running the shell command
@samp{locale charmap}.  If the result is @samp{C} or @samp{ANSI_X3.4-1968},
which is equivalent to @samp{ASCII} (= @samp{US-ASCII}), it means that your
locale is not correctly configured.  In this case, ask your translation
team which charset to use.  @samp{ASCII} is not usable for any language
except Latin.

@cindex encoding list
Because the PO files must be portable to operating systems with less advanced
internationalization facilities, the character encodings that can be used
are limited to those supported by both GNU @code{libc} and GNU
@code{libiconv}.  These are:
@code{ASCII}, @code{ISO-8859-1}, @code{ISO-8859-2}, @code{ISO-8859-3},
@code{ISO-8859-4}, @code{ISO-8859-5}, @code{ISO-8859-6}, @code{ISO-8859-7},
@code{ISO-8859-8}, @code{ISO-8859-9}, @code{ISO-8859-13}, @code{ISO-8859-14},
@code{ISO-8859-15},
@code{KOI8-R}, @code{KOI8-U}, @code{KOI8-T},
@code{CP850}, @code{CP866}, @code{CP874},
@code{CP932}, @code{CP949}, @code{CP950}, @code{CP1250}, @code{CP1251},
@code{CP1252}, @code{CP1253}, @code{CP1254}, @code{CP1255}, @code{CP1256},
@code{CP1257}, @code{GB2312}, @code{EUC-JP}, @code{EUC-KR}, @code{EUC-TW},
@code{BIG5}, @code{BIG5-HKSCS}, @code{GBK}, @code{GB18030}, @code{SHIFT_JIS},
@code{JOHAB}, @code{TIS-620}, @code{VISCII}, @code{GEORGIAN-PS}, @code{UTF-8}.

@c This data is taken from glibc/localedata/SUPPORTED.
@cindex Linux
In the GNU system, the following encodings are frequently used for the
corresponding languages.

@cindex encoding for your language
@itemize
@item @code{ISO-8859-1} for
Afrikaans, Albanian, Basque, Breton, Catalan, Cornish, Danish, Dutch,
English, Estonian, Faroese, Finnish, French, Galician, German,
Greenlandic, Icelandic, Indonesian, Irish, Italian, Malay, Manx,
Norwegian, Occitan, Portuguese, Spanish, Swedish, Tagalog, Uzbek,
Walloon,
@item @code{ISO-8859-2} for
Bosnian, Croatian, Czech, Hungarian, Polish, Romanian, Serbian, Slovak,
Slovenian,
@item @code{ISO-8859-3} for Maltese,
@item @code{ISO-8859-5} for Macedonian, Serbian,
@item @code{ISO-8859-6} for Arabic,
@item @code{ISO-8859-7} for Greek,
@item @code{ISO-8859-8} for Hebrew,
@item @code{ISO-8859-9} for Turkish,
@item @code{ISO-8859-13} for Latvian, Lithuanian, Maori,
@item @code{ISO-8859-14} for Welsh,
@item @code{ISO-8859-15} for
Basque, Catalan, Dutch, English, Finnish, French, Galician, German, Irish,
Italian, Portuguese, Spanish, Swedish, Walloon,
@item @code{KOI8-R} for Russian,
@item @code{KOI8-U} for Ukrainian,
@item @code{KOI8-T} for Tajik,
@item @code{CP1251} for Bulgarian, Belarusian,
@item @code{GB2312}, @code{GBK}, @code{GB18030}
for simplified writing of Chinese,
@item @code{BIG5}, @code{BIG5-HKSCS}
for traditional writing of Chinese,
@item @code{EUC-JP} for Japanese,
@item @code{EUC-KR} for Korean,
@item @code{TIS-620} for Thai,
@item @code{GEORGIAN-PS} for Georgian,
@item @code{UTF-8} for any language, including those listed above.
@end itemize

@cindex quote characters, use in PO files
@cindex quotation marks
When single quote characters or double quote characters are used in
translations for your language, and your locale's encoding is one of the
ISO-8859-* charsets, it is best if you create your PO files in UTF-8
encoding, instead of your locale's encoding.  This is because in UTF-8
the real quote characters can be represented (single quote characters:
U+2018, U+2019, double quote characters: U+201C, U+201D), whereas none of
ISO-8859-* charsets has them all.  Users in UTF-8 locales will see the
real quote characters, whereas users in ISO-8859-* locales will see the
vertical apostrophe and the vertical double quote instead (because that's
what the character set conversion will transliterate them to).

@cindex @code{xmodmap} program, and typing quotation marks
To enter such quote characters under X11, you can change your keyboard
mapping using the @code{xmodmap} program.  The X11 names of the quote
characters are "leftsinglequotemark", "rightsinglequotemark",
"leftdoublequotemark", "rightdoublequotemark", "singlelowquotemark",
"doublelowquotemark".

Note that only recent versions of GNU Emacs support the UTF-8 encoding:
Emacs 20 with Mule-UCS, and Emacs 21.  As of January 2001, XEmacs doesn't
support the UTF-8 encoding.

The character encoding name can be written in either upper or lower case.
Usually upper case is preferred.

@item Content-Transfer-Encoding
Set this to @code{8bit}.

@item Plural-Forms
This field is optional.  It is only needed if the PO file has plural forms.
You can find them by searching for the @samp{msgid_plural} keyword.  The
format of the plural forms field is described in @ref{Plural forms} and
@ref{Translating plural forms}.
@end table

@node Updating, Editing, Creating, Top
@chapter Updating Existing PO Files

@menu
* msgmerge Invocation::         Invoking the @code{msgmerge} Program
@end menu

@node msgmerge Invocation,  , Updating, Updating
@section Invoking the @code{msgmerge} Program

@include msgmerge.texi

@node Editing, Manipulating, Updating, Top
@chapter Editing PO Files
@cindex Editing PO Files

@menu
* KBabel::                      KDE's PO File Editor
* Gtranslator::                 GNOME's PO File Editor
* PO Mode::                     Emacs's PO File Editor
* Compendium::                  Using Translation Compendia
@end menu

@node KBabel, Gtranslator, Editing, Editing
@section KDE's PO File Editor
@cindex KDE PO file editor

@node Gtranslator, PO Mode, KBabel, Editing
@section GNOME's PO File Editor
@cindex GNOME PO file editor

@node PO Mode, Compendium, Gtranslator, Editing
@section Emacs's PO File Editor
@cindex Emacs PO Mode

@c FIXME: Rewrite.

For those of you being
the lucky users of Emacs, PO mode has been specifically created
for providing a cozy environment for editing or modifying PO files.
While editing a PO file, PO mode allows for the easy browsing of
auxiliary and compendium PO files, as well as for following references into
the set of C program sources from which PO files have been derived.
It has a few special features, among which are the interactive marking
of program strings as translatable, and the validation of PO files
with easy repositioning to PO file lines showing errors.

For the beginning, besides main PO mode commands
(@pxref{Main PO Commands}), you should know how to move between entries
(@pxref{Entry Positioning}), and how to handle untranslated entries
(@pxref{Untranslated Entries}).

@menu
* Installation::                Completing GNU @code{gettext} Installation
* Main PO Commands::            Main Commands
* Entry Positioning::           Entry Positioning
* Normalizing::                 Normalizing Strings in Entries
* Translated Entries::          Translated Entries
* Fuzzy Entries::               Fuzzy Entries
* Untranslated Entries::        Untranslated Entries
* Obsolete Entries::            Obsolete Entries
* Modifying Translations::      Modifying Translations
* Modifying Comments::          Modifying Comments
* Subedit::                     Mode for Editing Translations
* C Sources Context::           C Sources Context
* Auxiliary::                   Consulting Auxiliary PO Files
@end menu

@node Installation, Main PO Commands, PO Mode, PO Mode
@subsection Completing GNU @code{gettext} Installation

@cindex installing @code{gettext}
@cindex @code{gettext} installation
Once you have received, unpacked, configured and compiled the GNU
@code{gettext} distribution, the @samp{make install} command puts in
place the programs @code{xgettext}, @code{msgfmt}, @code{gettext}, and
@code{msgmerge}, as well as their available message catalogs.  To
top off a comfortable installation, you might also want to make the
PO mode available to your Emacs users.

@emindex @file{.emacs} customizations
@emindex installing PO mode
During the installation of the PO mode, you might want to modify your
file @file{.emacs}, once and for all, so it contains a few lines looking
like:

@example
(setq auto-mode-alist
      (cons '("\\.po\\'\\|\\.po\\." . po-mode) auto-mode-alist))
(autoload 'po-mode "po-mode" "Major mode for translators to edit PO files" t)
@end example

Later, whenever you edit some @file{.po}
file, or any file having the string @samp{.po.} within its name,
Emacs loads @file{po-mode.elc} (or @file{po-mode.el}) as needed, and
automatically activates PO mode commands for the associated buffer.
The string @emph{PO} appears in the mode line for any buffer for
which PO mode is active.  Many PO files may be active at once in a
single Emacs session.

If you are using Emacs version 20 or newer, and have already installed
the appropriate international fonts on your system, you may also tell
Emacs how to determine automatically the coding system of every PO file.
This will often (but not always) cause the necessary fonts to be loaded
and used for displaying the translations on your Emacs screen.  For this
to happen, add the lines:

@example
(modify-coding-system-alist 'file "\\.po\\'\\|\\.po\\."
                            'po-find-file-coding-system)
(autoload 'po-find-file-coding-system "po-mode")
@end example

@noindent
to your @file{.emacs} file.  If, with this, you still see boxes instead
of international characters, try a different font set (via Shift Mouse
button 1).

@node Main PO Commands, Entry Positioning, Installation, PO Mode
@subsection Main PO mode Commands

@cindex PO mode (Emacs) commands
@emindex commands
After setting up Emacs with something similar to the lines in
@ref{Installation}, PO mode is activated for a window when Emacs finds a
PO file in that window.  This puts the window read-only and establishes a
po-mode-map, which is a genuine Emacs mode, in a way that is not derived
from text mode in any way.  Functions found on @code{po-mode-hook},
if any, will be executed.

When PO mode is active in a window, the letters @samp{PO} appear
in the mode line for that window.  The mode line also displays how
many entries of each kind are held in the PO file.  For example,
the string @samp{132t+3f+10u+2o} would tell the translator that the
PO mode contains 132 translated entries (@pxref{Translated Entries},
3 fuzzy entries (@pxref{Fuzzy Entries}), 10 untranslated entries
(@pxref{Untranslated Entries}) and 2 obsolete entries (@pxref{Obsolete
Entries}).  Zero-coefficients items are not shown.  So, in this example, if
the fuzzy entries were unfuzzied, the untranslated entries were translated
and the obsolete entries were deleted, the mode line would merely display
@samp{145t} for the counters.

The main PO commands are those which do not fit into the other categories of
subsequent sections.  These allow for quitting PO mode or for managing windows
in special ways.

@table @kbd
@item _
@efindex _@r{, PO Mode command}
Undo last modification to the PO file (@code{po-undo}).

@item Q
@efindex Q@r{, PO Mode command}
Quit processing and save the PO file (@code{po-quit}).

@item q
@efindex q@r{, PO Mode command}
Quit processing, possibly after confirmation (@code{po-confirm-and-quit}).

@item 0
@efindex 0@r{, PO Mode command}
Temporary leave the PO file window (@code{po-other-window}).

@item ?
@itemx h
@efindex ?@r{, PO Mode command}
@efindex h@r{, PO Mode command}
Show help about PO mode (@code{po-help}).

@item =
@efindex =@r{, PO Mode command}
Give some PO file statistics (@code{po-statistics}).

@item V
@efindex V@r{, PO Mode command}
Batch validate the format of the whole PO file (@code{po-validate}).

@end table

@efindex _@r{, PO Mode command}
@efindex po-undo@r{, PO Mode command}
The command @kbd{_} (@code{po-undo}) interfaces to the Emacs
@emph{undo} facility.  @xref{Undo, , Undoing Changes, emacs, The Emacs
Editor}.  Each time @kbd{_} is typed, modifications which the translator
did to the PO file are undone a little more.  For the purpose of
undoing, each PO mode command is atomic.  This is especially true for
the @kbd{@key{RET}} command: the whole edition made by using a single
use of this command is undone at once, even if the edition itself
implied several actions.  However, while in the editing window, one
can undo the edition work quite parsimoniously.

@efindex Q@r{, PO Mode command}
@efindex q@r{, PO Mode command}
@efindex po-quit@r{, PO Mode command}
@efindex po-confirm-and-quit@r{, PO Mode command}
The commands @kbd{Q} (@code{po-quit}) and @kbd{q}
(@code{po-confirm-and-quit}) are used when the translator is done with the
PO file.  The former is a bit less verbose than the latter.  If the file
has been modified, it is saved to disk first.  In both cases, and prior to
all this, the commands check if any untranslated messages remain in the
PO file and, if so, the translator is asked if she really wants to leave
off working with this PO file.  This is the preferred way of getting rid
of an Emacs PO file buffer.  Merely killing it through the usual command
@w{@kbd{C-x k}} (@code{kill-buffer}) is not the tidiest way to proceed.

@efindex 0@r{, PO Mode command}
@efindex po-other-window@r{, PO Mode command}
The command @kbd{0} (@code{po-other-window}) is another, softer way,
to leave PO mode, temporarily.  It just moves the cursor to some other
Emacs window, and pops one if necessary.  For example, if the translator
just got PO mode to show some source context in some other, she might
discover some apparent bug in the program source that needs correction.
This command allows the translator to change sex, become a programmer,
and have the cursor right into the window containing the program she
(or rather @emph{he}) wants to modify.  By later getting the cursor back
in the PO file window, or by asking Emacs to edit this file once again,
PO mode is then recovered.

@efindex ?@r{, PO Mode command}
@efindex h@r{, PO Mode command}
@efindex po-help@r{, PO Mode command}
The command @kbd{h} (@code{po-help}) displays a summary of all available PO
mode commands.  The translator should then type any character to resume
normal PO mode operations.  The command @kbd{?} has the same effect
as @kbd{h}.

@efindex =@r{, PO Mode command}
@efindex po-statistics@r{, PO Mode command}
The command @kbd{=} (@code{po-statistics}) computes the total number of
entries in the PO file, the ordinal of the current entry (counted from
1), the number of untranslated entries, the number of obsolete entries,
and displays all these numbers.

@efindex V@r{, PO Mode command}
@efindex po-validate@r{, PO Mode command}
The command @kbd{V} (@code{po-validate}) launches @code{msgfmt} in
checking and verbose
mode over the current PO file.  This command first offers to save the
current PO file on disk.  The @code{msgfmt} tool, from GNU @code{gettext},
has the purpose of creating a MO file out of a PO file, and PO mode uses
the features of this program for checking the overall format of a PO file,
as well as all individual entries.

@efindex next-error@r{, stepping through PO file validation results}
The program @code{msgfmt} runs asynchronously with Emacs, so the
translator regains control immediately while her PO file is being studied.
Error output is collected in the Emacs @samp{*compilation*} buffer,
displayed in another window.  The regular Emacs command @kbd{C-x`}
(@code{next-error}), as well as other usual compile commands, allow the
translator to reposition quickly to the offending parts of the PO file.
Once the cursor is on the line in error, the translator may decide on
any PO mode action which would help correcting the error.

@node Entry Positioning, Normalizing, Main PO Commands, PO Mode
@subsection Entry Positioning

@emindex current entry of a PO file
The cursor in a PO file window is almost always part of
an entry.  The only exceptions are the special case when the cursor
is after the last entry in the file, or when the PO file is
empty.  The entry where the cursor is found to be is said to be the
current entry.  Many PO mode commands operate on the current entry,
so moving the cursor does more than allowing the translator to browse
the PO file, this also selects on which entry commands operate.

@emindex moving through a PO file
Some PO mode commands alter the position of the cursor in a specialized
way.  A few of those special purpose positioning are described here,
the others are described in following sections (for a complete list try
@kbd{C-h m}):

@table @kbd

@item .
@efindex .@r{, PO Mode command}
Redisplay the current entry (@code{po-current-entry}).

@item n
@efindex n@r{, PO Mode command}
Select the entry after the current one (@code{po-next-entry}).

@item p
@efindex p@r{, PO Mode command}
Select the entry before the current one (@code{po-previous-entry}).

@item <
@efindex <@r{, PO Mode command}
Select the first entry in the PO file (@code{po-first-entry}).

@item >
@efindex >@r{, PO Mode command}
Select the last entry in the PO file (@code{po-last-entry}).

@item m
@efindex m@r{, PO Mode command}
Record the location of the current entry for later use
(@code{po-push-location}).

@item r
@efindex r@r{, PO Mode command}
Return to a previously saved entry location (@code{po-pop-location}).

@item x
@efindex x@r{, PO Mode command}
Exchange the current entry location with the previously saved one
(@code{po-exchange-location}).

@end table

@efindex .@r{, PO Mode command}
@efindex po-current-entry@r{, PO Mode command}
Any Emacs command able to reposition the cursor may be used
to select the current entry in PO mode, including commands which
move by characters, lines, paragraphs, screens or pages, and search
commands.  However, there is a kind of standard way to display the
current entry in PO mode, which usual Emacs commands moving
the cursor do not especially try to enforce.  The command @kbd{.}
(@code{po-current-entry}) has the sole purpose of redisplaying the
current entry properly, after the current entry has been changed by
means external to PO mode, or the Emacs screen otherwise altered.

It is yet to be decided if PO mode helps the translator, or otherwise
irritates her, by forcing a rigid window disposition while she
is doing her work.  We originally had quite precise ideas about
how windows should behave, but on the other hand, anyone used to
Emacs is often happy to keep full control.  Maybe a fixed window
disposition might be offered as a PO mode option that the translator
might activate or deactivate at will, so it could be offered on an
experimental basis.  If nobody feels a real need for using it, or
a compulsion for writing it, we should drop this whole idea.
The incentive for doing it should come from translators rather than
programmers, as opinions from an experienced translator are surely
more worth to me than opinions from programmers @emph{thinking} about
how @emph{others} should do translation.

@efindex n@r{, PO Mode command}
@efindex po-next-entry@r{, PO Mode command}
@efindex p@r{, PO Mode command}
@efindex po-previous-entry@r{, PO Mode command}
The commands @kbd{n} (@code{po-next-entry}) and @kbd{p}
(@code{po-previous-entry}) move the cursor the entry following,
or preceding, the current one.  If @kbd{n} is given while the
cursor is on the last entry of the PO file, or if @kbd{p}
is given while the cursor is on the first entry, no move is done.

@efindex <@r{, PO Mode command}
@efindex po-first-entry@r{, PO Mode command}
@efindex >@r{, PO Mode command}
@efindex po-last-entry@r{, PO Mode command}
The commands @kbd{<} (@code{po-first-entry}) and @kbd{>}
(@code{po-last-entry}) move the cursor to the first entry, or last
entry, of the PO file.  When the cursor is located past the last
entry in a PO file, most PO mode commands will return an error saying
@samp{After last entry}.  Moreover, the commands @kbd{<} and @kbd{>}
have the special property of being able to work even when the cursor
is not into some PO file entry, and one may use them for nicely
correcting this situation.  But even these commands will fail on a
truly empty PO file.  There are development plans for the PO mode for it
to interactively fill an empty PO file from sources.  @xref{Marking}.

The translator may decide, before working at the translation of
a particular entry, that she needs to browse the remainder of the
PO file, maybe for finding the terminology or phraseology used
in related entries.  She can of course use the standard Emacs idioms
for saving the current cursor location in some register, and use that
register for getting back, or else, use the location ring.

@efindex m@r{, PO Mode command}
@efindex po-push-location@r{, PO Mode command}
@efindex r@r{, PO Mode command}
@efindex po-pop-location@r{, PO Mode command}
PO mode offers another approach, by which cursor locations may be saved
onto a special stack.  The command @kbd{m} (@code{po-push-location})
merely adds the location of current entry to the stack, pushing
the already saved locations under the new one.  The command
@kbd{r} (@code{po-pop-location}) consumes the top stack element and
repositions the cursor to the entry associated with that top element.
This position is then lost, for the next @kbd{r} will move the cursor
to the previously saved location, and so on until no locations remain
on the stack.

If the translator wants the position to be kept on the location stack,
maybe for taking a look at the entry associated with the top
element, then go elsewhere with the intent of getting back later, she
ought to use @kbd{m} immediately after @kbd{r}.

@efindex x@r{, PO Mode command}
@efindex po-exchange-location@r{, PO Mode command}
The command @kbd{x} (@code{po-exchange-location}) simultaneously
repositions the cursor to the entry associated with the top element of
the stack of saved locations, and replaces that top element with the
location of the current entry before the move.  Consequently, repeating
the @kbd{x} command toggles alternatively between two entries.
For achieving this, the translator will position the cursor on the
first entry, use @kbd{m}, then position to the second entry, and
merely use @kbd{x} for making the switch.

@node Normalizing, Translated Entries, Entry Positioning, PO Mode
@subsection Normalizing Strings in Entries
@cindex string normalization in entries

There are many different ways for encoding a particular string into a
PO file entry, because there are so many different ways to split and
quote multi-line strings, and even, to represent special characters
by backslashed escaped sequences.  Some features of PO mode rely on
the ability for PO mode to scan an already existing PO file for a
particular string encoded into the @code{msgid} field of some entry.
Even if PO mode has internally all the built-in machinery for
implementing this recognition easily, doing it fast is technically
difficult.  To facilitate a solution to this efficiency problem,
we decided on a canonical representation for strings.

A conventional representation of strings in a PO file is currently
under discussion, and PO mode experiments with a canonical representation.
Having both @code{xgettext} and PO mode converging towards a uniform
way of representing equivalent strings would be useful, as the internal
normalization needed by PO mode could be automatically satisfied
when using @code{xgettext} from GNU @code{gettext}.  An explicit
PO mode normalization should then be only necessary for PO files
imported from elsewhere, or for when the convention itself evolves.

So, for achieving normalization of at least the strings of a given
PO file needing a canonical representation, the following PO mode
command is available:

@emindex string normalization in entries
@table @kbd
@item M-x po-normalize
@efindex po-normalize@r{, PO Mode command}
Tidy the whole PO file by making entries more uniform.

@end table

The special command @kbd{M-x po-normalize}, which has no associated
keys, revises all entries, ensuring that strings of both original
and translated entries use uniform internal quoting in the PO file.
It also removes any crumb after the last entry.  This command may be
useful for PO files freshly imported from elsewhere, or if we ever
improve on the canonical quoting format we use.  This canonical format
is not only meant for getting cleaner PO files, but also for greatly
speeding up @code{msgid} string lookup for some other PO mode commands.

@kbd{M-x po-normalize} presently makes three passes over the entries.
The first implements heuristics for converting PO files for GNU
@code{gettext} 0.6 and earlier, in which @code{msgid} and @code{msgstr}
fields were using K&R style C string syntax for multi-line strings.
These heuristics may fail for comments not related to obsolete
entries and ending with a backslash; they also depend on subsequent
passes for finalizing the proper commenting of continued lines for
obsolete entries.  This first pass might disappear once all oldish PO
files would have been adjusted.  The second and third pass normalize
all @code{msgid} and @code{msgstr} strings respectively.  They also
clean out those trailing backslashes used by XView's @code{msgfmt}
for continued lines.

@cindex importing PO files
Having such an explicit normalizing command allows for importing PO
files from other sources, but also eases the evolution of the current
convention, evolution driven mostly by aesthetic concerns, as of now.
It is easy to make suggested adjustments at a later time, as the
normalizing command and eventually, other GNU @code{gettext} tools
should greatly automate conformance.  A description of the canonical
string format is given below, for the particular benefit of those not
having Emacs handy, and who would nevertheless want to handcraft
their PO files in nice ways.

@cindex multi-line strings
Right now, in PO mode, strings are single line or multi-line.  A string
goes multi-line if and only if it has @emph{embedded} newlines, that
is, if it matches @samp{[^\n]\n+[^\n]}.  So, we would have:

@example
msgstr "\n\nHello, world!\n\n\n"
@end example

but, replacing the space by a newline, this becomes:

@example
msgstr ""
"\n"
"\n"
"Hello,\n"
"world!\n"
"\n"
"\n"
@end example

We are deliberately using a caricatural example, here, to make the
point clearer.  Usually, multi-lines are not that bad looking.
It is probable that we will implement the following suggestion.
We might lump together all initial newlines into the empty string,
and also all newlines introducing empty lines (that is, for @w{@var{n}
> 1}, the @var{n}-1'th last newlines would go together on a separate
string), so making the previous example appear:

@example
msgstr "\n\n"
"Hello,\n"
"world!\n"
"\n\n"
@end example

There are a few yet undecided little points about string normalization,
to be documented in this manual, once these questions settle.

@node Translated Entries, Fuzzy Entries, Normalizing, PO Mode
@subsection Translated Entries
@cindex translated entries

Each PO file entry for which the @code{msgstr} field has been filled with
a translation, and which is not marked as fuzzy (@pxref{Fuzzy Entries}),
is said to be a @dfn{translated} entry.  Only translated entries will
later be compiled by GNU @code{msgfmt} and become usable in programs.
Other entry types will be excluded; translation will not occur for them.

@emindex moving by translated entries
Some commands are more specifically related to translated entry processing.

@table @kbd
@item t
@efindex t@r{, PO Mode command}
Find the next translated entry (@code{po-next-translated-entry}).

@item T
@efindex T@r{, PO Mode command}
Find the previous translated entry (@code{po-previous-translated-entry}).

@end table

@efindex t@r{, PO Mode command}
@efindex po-next-translated-entry@r{, PO Mode command}
@efindex T@r{, PO Mode command}
@efindex po-previous-translated-entry@r{, PO Mode command}
The commands @kbd{t} (@code{po-next-translated-entry}) and @kbd{T}
(@code{po-previous-translated-entry}) move forwards or backwards, chasing
for an translated entry.  If none is found, the search is extended and
wraps around in the PO file buffer.

@evindex po-auto-fuzzy-on-edit@r{, PO Mode variable}
Translated entries usually result from the translator having edited in
a translation for them, @ref{Modifying Translations}.  However, if the
variable @code{po-auto-fuzzy-on-edit} is not @code{nil}, the entry having
received a new translation first becomes a fuzzy entry, which ought to
be later unfuzzied before becoming an official, genuine translated entry.
@xref{Fuzzy Entries}.

@node Fuzzy Entries, Untranslated Entries, Translated Entries, PO Mode
@subsection Fuzzy Entries
@cindex fuzzy entries

@cindex attributes of a PO file entry
@cindex attribute, fuzzy
Each PO file entry may have a set of @dfn{attributes}, which are
qualities given a name and explicitly associated with the translation,
using a special system comment.  One of these attributes
has the name @code{fuzzy}, and entries having this attribute are said
to have a fuzzy translation.  They are called fuzzy entries, for short.

Fuzzy entries, even if they account for translated entries for
most other purposes, usually call for revision by the translator.
Those may be produced by applying the program @code{msgmerge} to
update an older translated PO files according to a new PO template
file, when this tool hypothesises that some new @code{msgid} has
been modified only slightly out of an older one, and chooses to pair
what it thinks to be the old translation for the new modified entry.
The slight alteration in the original string (the @code{msgid} string)
should often be reflected in the translated string, and this requires
the intervention of the translator.  For this reason, @code{msgmerge}
might mark some entries as being fuzzy.

@emindex moving by fuzzy entries
Also, the translator may decide herself to mark an entry as fuzzy
for her own convenience, when she wants to remember that the entry
has to be later revisited.  So, some commands are more specifically
related to fuzzy entry processing.

@table @kbd
@item f
@efindex f@r{, PO Mode command}
@c better append "-entry" all the time. -ke-
Find the next fuzzy entry (@code{po-next-fuzzy-entry}).

@item F
@efindex F@r{, PO Mode command}
Find the previous fuzzy entry (@code{po-previous-fuzzy-entry}).

@item @key{TAB}
@efindex TAB@r{, PO Mode command}
Remove the fuzzy attribute of the current entry (@code{po-unfuzzy}).

@end table

@efindex f@r{, PO Mode command}
@efindex po-next-fuzzy-entry@r{, PO Mode command}
@efindex F@r{, PO Mode command}
@efindex po-previous-fuzzy-entry@r{, PO Mode command}
The commands @kbd{f} (@code{po-next-fuzzy-entry}) and @kbd{F}
(@code{po-previous-fuzzy-entry}) move forwards or backwards, chasing for
a fuzzy entry.  If none is found, the search is extended and wraps
around in the PO file buffer.

@efindex TAB@r{, PO Mode command}
@efindex po-unfuzzy@r{, PO Mode command}
@evindex po-auto-select-on-unfuzzy@r{, PO Mode variable}
The command @kbd{@key{TAB}} (@code{po-unfuzzy}) removes the fuzzy
attribute associated with an entry, usually leaving it translated.
Further, if the variable @code{po-auto-select-on-unfuzzy} has not
the @code{nil} value, the @kbd{@key{TAB}} command will automatically chase
for another interesting entry to work on.  The initial value of
@code{po-auto-select-on-unfuzzy} is @code{nil}.

The initial value of @code{po-auto-fuzzy-on-edit} is @code{nil}.  However,
if the variable @code{po-auto-fuzzy-on-edit} is set to @code{t}, any entry
edited through the @kbd{@key{RET}} command is marked fuzzy, as a way to
ensure some kind of double check, later.  In this case, the usual paradigm
is that an entry becomes fuzzy (if not already) whenever the translator
modifies it.  If she is satisfied with the translation, she then uses
@kbd{@key{TAB}} to pick another entry to work on, clearing the fuzzy attribute
on the same blow.  If she is not satisfied yet, she merely uses @kbd{@key{SPC}}
to chase another entry, leaving the entry fuzzy.

@efindex DEL@r{, PO Mode command}
@efindex po-fade-out-entry@r{, PO Mode command}
The translator may also use the @kbd{@key{DEL}} command
(@code{po-fade-out-entry}) over any translated entry to mark it as being
fuzzy, when she wants to easily leave a trace she wants to later return
working at this entry.

Also, when time comes to quit working on a PO file buffer with the @kbd{q}
command, the translator is asked for confirmation, if fuzzy string
still exists.

@node Untranslated Entries, Obsolete Entries, Fuzzy Entries, PO Mode
@subsection Untranslated Entries
@cindex untranslated entries

When @code{xgettext} originally creates a PO file, unless told
otherwise, it initializes the @code{msgid} field with the untranslated
string, and leaves the @code{msgstr} string to be empty.  Such entries,
having an empty translation, are said to be @dfn{untranslated} entries.
Later, when the programmer slightly modifies some string right in
the program, this change is later reflected in the PO file
by the appearance of a new untranslated entry for the modified string.

The usual commands moving from entry to entry consider untranslated
entries on the same level as active entries.  Untranslated entries
are easily recognizable by the fact they end with @w{@samp{msgstr ""}}.

@emindex moving by untranslated entries
The work of the translator might be (quite naively) seen as the process
of seeking for an untranslated entry, editing a translation for
it, and repeating these actions until no untranslated entries remain.
Some commands are more specifically related to untranslated entry
processing.

@table @kbd
@item u
@efindex u@r{, PO Mode command}
Find the next untranslated entry (@code{po-next-untranslated-entry}).

@item U
@efindex U@r{, PO Mode command}
Find the previous untranslated entry (@code{po-previous-untransted-entry}).

@item k
@efindex k@r{, PO Mode command}
Turn the current entry into an untranslated one (@code{po-kill-msgstr}).

@end table

@efindex u@r{, PO Mode command}
@efindex po-next-untranslated-entry@r{, PO Mode command}
@efindex U@r{, PO Mode command}
@efindex po-previous-untransted-entry@r{, PO Mode command}
The commands @kbd{u} (@code{po-next-untranslated-entry}) and @kbd{U}
(@code{po-previous-untransted-entry}) move forwards or backwards,
chasing for an untranslated entry.  If none is found, the search is
extended and wraps around in the PO file buffer.

@efindex k@r{, PO Mode command}
@efindex po-kill-msgstr@r{, PO Mode command}
An entry can be turned back into an untranslated entry by
merely emptying its translation, using the command @kbd{k}
(@code{po-kill-msgstr}).  @xref{Modifying Translations}.

Also, when time comes to quit working on a PO file buffer
with the @kbd{q} command, the translator is asked for confirmation,
if some untranslated string still exists.

@node Obsolete Entries, Modifying Translations, Untranslated Entries, PO Mode
@subsection Obsolete Entries
@cindex obsolete entries

By @dfn{obsolete} PO file entries, we mean those entries which are
commented out, usually by @code{msgmerge} when it found that the
translation is not needed anymore by the package being localized.

The usual commands moving from entry to entry consider obsolete
entries on the same level as active entries.  Obsolete entries are
easily recognizable by the fact that all their lines start with
@code{#}, even those lines containing @code{msgid} or @code{msgstr}.

Commands exist for emptying the translation or reinitializing it
to the original untranslated string.  Commands interfacing with the
kill ring may force some previously saved text into the translation.
The user may interactively edit the translation.  All these commands
may apply to obsolete entries, carefully leaving the entry obsolete
after the fact.

@emindex moving by obsolete entries
Moreover, some commands are more specifically related to obsolete
entry processing.

@table @kbd
@item o
@efindex o@r{, PO Mode command}
Find the next obsolete entry (@code{po-next-obsolete-entry}).

@item O
@efindex O@r{, PO Mode command}
Find the previous obsolete entry (@code{po-previous-obsolete-entry}).

@item @key{DEL}
@efindex DEL@r{, PO Mode command}
Make an active entry obsolete, or zap out an obsolete entry
(@code{po-fade-out-entry}).

@end table

@efindex o@r{, PO Mode command}
@efindex po-next-obsolete-entry@r{, PO Mode command}
@efindex O@r{, PO Mode command}
@efindex po-previous-obsolete-entry@r{, PO Mode command}
The commands @kbd{o} (@code{po-next-obsolete-entry}) and @kbd{O}
(@code{po-previous-obsolete-entry}) move forwards or backwards,
chasing for an obsolete entry.  If none is found, the search is
extended and wraps around in the PO file buffer.

PO mode does not provide ways for un-commenting an obsolete entry
and making it active, because this would reintroduce an original
untranslated string which does not correspond to any marked string
in the program sources.  This goes with the philosophy of never
introducing useless @code{msgid} values.

@efindex DEL@r{, PO Mode command}
@efindex po-fade-out-entry@r{, PO Mode command}
@emindex obsolete active entry
@emindex comment out PO file entry
However, it is possible to comment out an active entry, so making
it obsolete.  GNU @code{gettext} utilities will later react to the
disappearance of a translation by using the untranslated string.
The command @kbd{@key{DEL}} (@code{po-fade-out-entry}) pushes the current entry
a little further towards annihilation.  If the entry is active (it is a
translated entry), then it is first made fuzzy.  If it is already fuzzy,
then the entry is merely commented out, with confirmation.  If the entry
is already obsolete, then it is completely deleted from the PO file.
It is easy to recycle the translation so deleted into some other PO file
entry, usually one which is untranslated.  @xref{Modifying Translations}.

Here is a quite interesting problem to solve for later development of
PO mode, for those nights you are not sleepy.  The idea would be that
PO mode might become bright enough, one of these days, to make good
guesses at retrieving the most probable candidate, among all obsolete
entries, for initializing the translation of a newly appeared string.
I think it might be a quite hard problem to do this algorithmically, as
we have to develop good and efficient measures of string similarity.
Right now, PO mode completely lets the decision to the translator,
when the time comes to find the adequate obsolete translation, it
merely tries to provide handy tools for helping her to do so.

@node Modifying Translations, Modifying Comments, Obsolete Entries, PO Mode
@subsection Modifying Translations
@cindex editing translations
@emindex editing translations

PO mode prevents direct modification of the PO file, by the usual
means Emacs gives for altering a buffer's contents.  By doing so,
it pretends helping the translator to avoid little clerical errors
about the overall file format, or the proper quoting of strings,
as those errors would be easily made.  Other kinds of errors are
still possible, but some may be caught and diagnosed by the batch
validation process, which the translator may always trigger by the
@kbd{V} command.  For all other errors, the translator has to rely on
her own judgment, and also on the linguistic reports submitted to her
by the users of the translated package, having the same mother tongue.

When the time comes to create a translation, correct an error diagnosed
mechanically or reported by a user, the translators have to resort to
using the following commands for modifying the translations.

@table @kbd
@item @key{RET}
@efindex RET@r{, PO Mode command}
Interactively edit the translation (@code{po-edit-msgstr}).

@item @key{LFD}
@itemx C-j
@efindex LFD@r{, PO Mode command}
@efindex C-j@r{, PO Mode command}
Reinitialize the translation with the original, untranslated string
(@code{po-msgid-to-msgstr}).

@item k
@efindex k@r{, PO Mode command}
Save the translation on the kill ring, and delete it (@code{po-kill-msgstr}).

@item w
@efindex w@r{, PO Mode command}
Save the translation on the kill ring, without deleting it
(@code{po-kill-ring-save-msgstr}).

@item y
@efindex y@r{, PO Mode command}
Replace the translation, taking the new from the kill ring
(@code{po-yank-msgstr}).

@end table

@efindex RET@r{, PO Mode command}
@efindex po-edit-msgstr@r{, PO Mode command}
The command @kbd{@key{RET}} (@code{po-edit-msgstr}) opens a new Emacs
window meant to edit in a new translation, or to modify an already existing
translation.  The new window contains a copy of the translation taken from
the current PO file entry, all ready for edition, expunged of all quoting
marks, fully modifiable and with the complete extent of Emacs modifying
commands.  When the translator is done with her modifications, she may use
@w{@kbd{C-c C-c}} to close the subedit window with the automatically requoted
results, or @w{@kbd{C-c C-k}} to abort her modifications.  @xref{Subedit},
for more information.

@efindex LFD@r{, PO Mode command}
@efindex C-j@r{, PO Mode command}
@efindex po-msgid-to-msgstr@r{, PO Mode command}
The command @kbd{@key{LFD}} (@code{po-msgid-to-msgstr}) initializes, or
reinitializes the translation with the original string.  This command is
normally used when the translator wants to redo a fresh translation of
the original string, disregarding any previous work.

@evindex po-auto-edit-with-msgid@r{, PO Mode variable}
It is possible to arrange so, whenever editing an untranslated
entry, the @kbd{@key{LFD}} command be automatically executed.  If you set
@code{po-auto-edit-with-msgid} to @code{t}, the translation gets
initialised with the original string, in case none exists already.
The default value for @code{po-auto-edit-with-msgid} is @code{nil}.

@emindex starting a string translation
In fact, whether it is best to start a translation with an empty
string, or rather with a copy of the original string, is a matter of
taste or habit.  Sometimes, the source language and the
target language are so different that is simply best to start writing
on an empty page.  At other times, the source and target languages
are so close that it would be a waste to retype a number of words
already being written in the original string.  A translator may also
like having the original string right under her eyes, as she will
progressively overwrite the original text with the translation, even
if this requires some extra editing work to get rid of the original.

@emindex cut and paste for translated strings
@efindex k@r{, PO Mode command}
@efindex po-kill-msgstr@r{, PO Mode command}
@efindex w@r{, PO Mode command}
@efindex po-kill-ring-save-msgstr@r{, PO Mode command}
The command @kbd{k} (@code{po-kill-msgstr}) merely empties the
translation string, so turning the entry into an untranslated
one.  But while doing so, its previous contents is put apart in
a special place, known as the kill ring.  The command @kbd{w}
(@code{po-kill-ring-save-msgstr}) has also the effect of taking a
copy of the translation onto the kill ring, but it otherwise leaves
the entry alone, and does @emph{not} remove the translation from the
entry.  Both commands use exactly the Emacs kill ring, which is shared
between buffers, and which is well known already to Emacs lovers.

The translator may use @kbd{k} or @kbd{w} many times in the course
of her work, as the kill ring may hold several saved translations.
From the kill ring, strings may later be reinserted in various
Emacs buffers.  In particular, the kill ring may be used for moving
translation strings between different entries of a single PO file
buffer, or if the translator is handling many such buffers at once,
even between PO files.

To facilitate exchanges with buffers which are not in PO mode, the
translation string put on the kill ring by the @kbd{k} command is fully
unquoted before being saved: external quotes are removed, multi-line
strings are concatenated, and backslash escaped sequences are turned
into their corresponding characters.  In the special case of obsolete
entries, the translation is also uncommented prior to saving.

@efindex y@r{, PO Mode command}
@efindex po-yank-msgstr@r{, PO Mode command}
The command @kbd{y} (@code{po-yank-msgstr}) completely replaces the
translation of the current entry by a string taken from the kill ring.
Following Emacs terminology, we then say that the replacement
string is @dfn{yanked} into the PO file buffer.
@xref{Yanking, , , emacs, The Emacs Editor}.
The first time @kbd{y} is used, the translation receives the value of
the most recent addition to the kill ring.  If @kbd{y} is typed once
again, immediately, without intervening keystrokes, the translation
just inserted is taken away and replaced by the second most recent
addition to the kill ring.  By repeating @kbd{y} many times in a row,
the translator may travel along the kill ring for saved strings,
until she finds the string she really wanted.

When a string is yanked into a PO file entry, it is fully and
automatically requoted for complying with the format PO files should
have.  Further, if the entry is obsolete, PO mode then appropriately
push the inserted string inside comments.  Once again, translators
should not burden themselves with quoting considerations besides, of
course, the necessity of the translated string itself respective to
the program using it.

Note that @kbd{k} or @kbd{w} are not the only commands pushing strings
on the kill ring, as almost any PO mode command replacing translation
strings (or the translator comments) automatically saves the old string
on the kill ring.  The main exceptions to this general rule are the
yanking commands themselves.

@emindex using obsolete translations to make new entries
To better illustrate the operation of killing and yanking, let's
use an actual example, taken from a common situation.  When the
programmer slightly modifies some string right in the program, his
change is later reflected in the PO file by the appearance
of a new untranslated entry for the modified string, and the fact
that the entry translating the original or unmodified string becomes
obsolete.  In many cases, the translator might spare herself some work
by retrieving the unmodified translation from the obsolete entry,
then initializing the untranslated entry @code{msgstr} field with
this retrieved translation.  Once this done, the obsolete entry is
not wanted anymore, and may be safely deleted.

When the translator finds an untranslated entry and suspects that a
slight variant of the translation exists, she immediately uses @kbd{m}
to mark the current entry location, then starts chasing obsolete
entries with @kbd{o}, hoping to find some translation corresponding
to the unmodified string.  Once found, she uses the @kbd{@key{DEL}} command
for deleting the obsolete entry, knowing that @kbd{@key{DEL}} also @emph{kills}
the translation, that is, pushes the translation on the kill ring.
Then, @kbd{r} returns to the initial untranslated entry, and @kbd{y}
then @emph{yanks} the saved translation right into the @code{msgstr}
field.  The translator is then free to use @kbd{@key{RET}} for fine
tuning the translation contents, and maybe to later use @kbd{u},
then @kbd{m} again, for going on with the next untranslated string.

When some sequence of keys has to be typed over and over again, the
translator may find it useful to become better acquainted with the Emacs
capability of learning these sequences and playing them back under request.
@xref{Keyboard Macros, , , emacs, The Emacs Editor}.

@node Modifying Comments, Subedit, Modifying Translations, PO Mode
@subsection Modifying Comments
@cindex editing comments in PO files
@emindex editing comments

Any translation work done seriously will raise many linguistic
difficulties, for which decisions have to be made, and the choices
further documented.  These documents may be saved within the
PO file in form of translator comments, which the translator
is free to create, delete, or modify at will.  These comments may
be useful to herself when she returns to this PO file after a while.

Comments not having whitespace after the initial @samp{#}, for example,
those beginning with @samp{#.} or @samp{#:}, are @emph{not} translator
comments, they are exclusively created by other @code{gettext} tools.
So, the commands below will never alter such system added comments,
they are not meant for the translator to modify.  @xref{PO Files}.

The following commands are somewhat similar to those modifying translations,
so the general indications given for those apply here.  @xref{Modifying
Translations}.

@table @kbd

@item #
@efindex #@r{, PO Mode command}
Interactively edit the translator comments (@code{po-edit-comment}).

@item K
@efindex K@r{, PO Mode command}
Save the translator comments on the kill ring, and delete it
(@code{po-kill-comment}).

@item W
@efindex W@r{, PO Mode command}
Save the translator comments on the kill ring, without deleting it
(@code{po-kill-ring-save-comment}).

@item Y
@efindex Y@r{, PO Mode command}
Replace the translator comments, taking the new from the kill ring
(@code{po-yank-comment}).

@end table

These commands parallel PO mode commands for modifying the translation
strings, and behave much the same way as they do, except that they handle
this part of PO file comments meant for translator usage, rather
than the translation strings.  So, if the descriptions given below are
slightly succinct, it is because the full details have already been given.
@xref{Modifying Translations}.

@efindex #@r{, PO Mode command}
@efindex po-edit-comment@r{, PO Mode command}
The command @kbd{#} (@code{po-edit-comment}) opens a new Emacs window
containing a copy of the translator comments on the current PO file entry.
If there are no such comments, PO mode understands that the translator wants
to add a comment to the entry, and she is presented with an empty screen.
Comment marks (@code{#}) and the space following them are automatically
removed before edition, and reinstated after.  For translator comments
pertaining to obsolete entries, the uncommenting and recommenting operations
are done twice.  Once in the editing window, the keys @w{@kbd{C-c C-c}}
allow the translator to tell she is finished with editing the comment.
@xref{Subedit}, for further details.

@evindex po-subedit-mode-hook@r{, PO Mode variable}
Functions found on @code{po-subedit-mode-hook}, if any, are executed after
the string has been inserted in the edit buffer.

@efindex K@r{, PO Mode command}
@efindex po-kill-comment@r{, PO Mode command}
@efindex W@r{, PO Mode command}
@efindex po-kill-ring-save-comment@r{, PO Mode command}
@efindex Y@r{, PO Mode command}
@efindex po-yank-comment@r{, PO Mode command}
The command @kbd{K} (@code{po-kill-comment}) gets rid of all
translator comments, while saving those comments on the kill ring.
The command @kbd{W} (@code{po-kill-ring-save-comment}) takes
a copy of the translator comments on the kill ring, but leaves
them undisturbed in the current entry.  The command @kbd{Y}
(@code{po-yank-comment}) completely replaces the translator comments
by a string taken at the front of the kill ring.  When this command
is immediately repeated, the comments just inserted are withdrawn,
and replaced by other strings taken along the kill ring.

On the kill ring, all strings have the same nature.  There is no
distinction between @emph{translation} strings and @emph{translator
comments} strings.  So, for example, let's presume the translator
has just finished editing a translation, and wants to create a new
translator comment to document why the previous translation was
not good, just to remember what was the problem.  Foreseeing that she
will do that in her documentation, the translator may want to quote
the previous translation in her translator comments.  To do so, she
may initialize the translator comments with the previous translation,
still at the head of the kill ring.  Because editing already pushed the
previous translation on the kill ring, she merely has to type @kbd{M-w}
prior to @kbd{#}, and the previous translation will be right there,
all ready for being introduced by some explanatory text.

On the other hand, presume there are some translator comments already
and that the translator wants to add to those comments, instead
of wholly replacing them.  Then, she should edit the comment right
away with @kbd{#}.  Once inside the editing window, she can use the
regular Emacs commands @kbd{C-y} (@code{yank}) and @kbd{M-y}
(@code{yank-pop}) to get the previous translation where she likes.

@node Subedit, C Sources Context, Modifying Comments, PO Mode
@subsection Details of Sub Edition
@emindex subedit minor mode

The PO subedit minor mode has a few peculiarities worth being described
in fuller detail.  It installs a few commands over the usual editing set
of Emacs, which are described below.

@table @kbd
@item C-c C-c
@efindex C-c C-c@r{, PO Mode command}
Complete edition (@code{po-subedit-exit}).

@item C-c C-k
@efindex C-c C-k@r{, PO Mode command}
Abort edition (@code{po-subedit-abort}).

@item C-c C-a
@efindex C-c C-a@r{, PO Mode command}
Consult auxiliary PO files (@code{po-subedit-cycle-auxiliary}).

@end table

@emindex exiting PO subedit
@efindex C-c C-c@r{, PO Mode command}
@efindex po-subedit-exit@r{, PO Mode command}
The window's contents represents a translation for a given message,
or a translator comment.  The translator may modify this window to
her heart's content.  Once this is done, the command @w{@kbd{C-c C-c}}
(@code{po-subedit-exit}) may be used to return the edited translation into
the PO file, replacing the original translation, even if it moved out of
sight or if buffers were switched.

@efindex C-c C-k@r{, PO Mode command}
@efindex po-subedit-abort@r{, PO Mode command}
If the translator becomes unsatisfied with her translation or comment,
to the extent she prefers keeping what was existent prior to the
@kbd{@key{RET}} or @kbd{#} command, she may use the command @w{@kbd{C-c C-k}}
(@code{po-subedit-abort}) to merely get rid of edition, while preserving
the original translation or comment.  Another way would be for her to exit
normally with @w{@kbd{C-c C-c}}, then type @code{U} once for undoing the
whole effect of last edition.

@efindex C-c C-a@r{, PO Mode command}
@efindex po-subedit-cycle-auxiliary@r{, PO Mode command}
The command @w{@kbd{C-c C-a}} (@code{po-subedit-cycle-auxiliary})
allows for glancing through translations
already achieved in other languages, directly while editing the current
translation.  This may be quite convenient when the translator is fluent
at many languages, but of course, only makes sense when such completed
auxiliary PO files are already available to her (@pxref{Auxiliary}).

Functions found on @code{po-subedit-mode-hook}, if any, are executed after
the string has been inserted in the edit buffer.

While editing her translation, the translator should pay attention to not
inserting unwanted @kbd{@key{RET}} (newline) characters at the end of
the translated string if those are not meant to be there, or to removing
such characters when they are required.  Since these characters are not
visible in the editing buffer, they are easily introduced by mistake.
To help her, @kbd{@key{RET}} automatically puts the character @code{<}
at the end of the string being edited, but this @code{<} is not really
part of the string.  On exiting the editing window with @w{@kbd{C-c C-c}},
PO mode automatically removes such @kbd{<} and all whitespace added after
it.  If the translator adds characters after the terminating @code{<}, it
looses its delimiting property and integrally becomes part of the string.
If she removes the delimiting @code{<}, then the edited string is taken
@emph{as is}, with all trailing newlines, even if invisible.  Also, if
the translated string ought to end itself with a genuine @code{<}, then
the delimiting @code{<} may not be removed; so the string should appear,
in the editing window, as ending with two @code{<} in a row.

@emindex editing multiple entries
When a translation (or a comment) is being edited, the translator may move
the cursor back into the PO file buffer and freely move to other entries,
browsing at will.  If, with an edition pending, the translator wanders in the
PO file buffer, she may decide to start modifying another entry.  Each entry
being edited has its own subedit buffer.  It is possible to simultaneously
edit the translation @emph{and} the comment of a single entry, or to
edit entries in different PO files, all at once.  Typing @kbd{@key{RET}}
on a field already being edited merely resumes that particular edit.  Yet,
the translator should better be comfortable at handling many Emacs windows!

@emindex pending subedits
Pending subedits may be completed or aborted in any order, regardless
of how or when they were started.  When many subedits are pending and the
translator asks for quitting the PO file (with the @kbd{q} command), subedits
are automatically resumed one at a time, so she may decide for each of them.

@node C Sources Context, Auxiliary, Subedit, PO Mode
@subsection C Sources Context
@emindex consulting program sources
@emindex looking at the source to aid translation
@emindex use the source, Luke

PO mode is particularly powerful when used with PO files
created through GNU @code{gettext} utilities, as those utilities
insert special comments in the PO files they generate.
Some of these special comments relate the PO file entry to
exactly where the untranslated string appears in the program sources.

When the translator gets to an untranslated entry, she is fairly
often faced with an original string which is not as informative as
it normally should be, being succinct, cryptic, or otherwise ambiguous.
Before choosing how to translate the string, she needs to understand
better what the string really means and how tight the translation has
to be.  Most of the time, when problems arise, the only way left to make
her judgment is looking at the true program sources from where this
string originated, searching for surrounding comments the programmer
might have put in there, and looking around for helping clues of
@emph{any} kind.

Surely, when looking at program sources, the translator will receive
more help if she is a fluent programmer.  However, even if she is
not versed in programming and feels a little lost in C code, the
translator should not be shy at taking a look, once in a while.
It is most probable that she will still be able to find some of the
hints she needs.  She will learn quickly to not feel uncomfortable
in program code, paying more attention to programmer's comments,
variable and function names (if he dared choosing them well), and
overall organization, than to the program code itself.

@emindex find source fragment for a PO file entry
The following commands are meant to help the translator at getting
program source context for a PO file entry.

@table @kbd
@item s
@efindex s@r{, PO Mode command}
Resume the display of a program source context, or cycle through them
(@code{po-cycle-source-reference}).

@item M-s
@efindex M-s@r{, PO Mode command}
Display of a program source context selected by menu
(@code{po-select-source-reference}).

@item S
@efindex S@r{, PO Mode command}
Add a directory to the search path for source files
(@code{po-consider-source-path}).

@item M-S
@efindex M-S@r{, PO Mode command}
Delete a directory from the search path for source files
(@code{po-ignore-source-path}).

@end table

@efindex s@r{, PO Mode command}
@efindex po-cycle-source-reference@r{, PO Mode command}
@efindex M-s@r{, PO Mode command}
@efindex po-select-source-reference@r{, PO Mode command}
The commands @kbd{s} (@code{po-cycle-source-reference}) and @kbd{M-s}
(@code{po-select-source-reference}) both open another window displaying
some source program file, and already positioned in such a way that
it shows an actual use of the string to be translated.  By doing
so, the command gives source program context for the string.  But if
the entry has no source context references, or if all references
are unresolved along the search path for program sources, then the
command diagnoses this as an error.

Even if @kbd{s} (or @kbd{M-s}) opens a new window, the cursor stays
in the PO file window.  If the translator really wants to
get into the program source window, she ought to do it explicitly,
maybe by using command @kbd{O}.

When @kbd{s} is typed for the first time, or for a PO file entry which
is different of the last one used for getting source context, then the
command reacts by giving the first context available for this entry,
if any.  If some context has already been recently displayed for the
current PO file entry, and the translator wandered off to do other
things, typing @kbd{s} again will merely resume, in another window,
the context last displayed.  In particular, if the translator moved
the cursor away from the context in the source file, the command will
bring the cursor back to the context.  By using @kbd{s} many times
in a row, with no other commands intervening, PO mode will cycle to
the next available contexts for this particular entry, getting back
to the first context once the last has been shown.

The command @kbd{M-s} behaves differently.  Instead of cycling through
references, it lets the translator choose a particular reference among
many, and displays that reference.  It is best used with completion,
if the translator types @kbd{@key{TAB}} immediately after @kbd{M-s}, in
response to the question, she will be offered a menu of all possible
references, as a reminder of which are the acceptable answers.
This command is useful only where there are really many contexts
available for a single string to translate.

@efindex S@r{, PO Mode command}
@efindex po-consider-source-path@r{, PO Mode command}
@efindex M-S@r{, PO Mode command}
@efindex po-ignore-source-path@r{, PO Mode command}
Program source files are usually found relative to where the PO
file stands.  As a special provision, when this fails, the file is
also looked for, but relative to the directory immediately above it.
Those two cases take proper care of most PO files.  However, it might
happen that a PO file has been moved, or is edited in a different
place than its normal location.  When this happens, the translator
should tell PO mode in which directory normally sits the genuine PO
file.  Many such directories may be specified, and all together, they
constitute what is called the @dfn{search path} for program sources.
The command @kbd{S} (@code{po-consider-source-path}) is used to interactively
enter a new directory at the front of the search path, and the command
@kbd{M-S} (@code{po-ignore-source-path}) is used to select, with completion,
one of the directories she does not want anymore on the search path.

@node Auxiliary,  , C Sources Context, PO Mode
@subsection Consulting Auxiliary PO Files
@emindex consulting translations to other languages

PO mode is able to help the knowledgeable translator, being fluent in
many languages, at taking advantage of translations already achieved
in other languages she just happens to know.  It provides these other
language translations as additional context for her own work.  Moreover,
it has features to ease the production of translations for many languages
at once, for translators preferring to work in this way.

@cindex auxiliary PO file
@emindex auxiliary PO file
An @dfn{auxiliary} PO file is an existing PO file meant for the same
package the translator is working on, but targeted to a different mother
tongue language.  Commands exist for declaring and handling auxiliary
PO files, and also for showing contexts for the entry under work.

Here are the auxiliary file commands available in PO mode.

@table @kbd
@item a
@efindex a@r{, PO Mode command}
Seek auxiliary files for another translation for the same entry
(@code{po-cycle-auxiliary}).

@item C-c C-a
@efindex C-c C-a@r{, PO Mode command}
Switch to a particular auxiliary file (@code{po-select-auxiliary}).

@item A
@efindex A@r{, PO Mode command}
Declare this PO file as an auxiliary file (@code{po-consider-as-auxiliary}).

@item M-A
@efindex M-A@r{, PO Mode command}
Remove this PO file from the list of auxiliary files
(@code{po-ignore-as-auxiliary}).

@end table

@efindex A@r{, PO Mode command}
@efindex po-consider-as-auxiliary@r{, PO Mode command}
@efindex M-A@r{, PO Mode command}
@efindex po-ignore-as-auxiliary@r{, PO Mode command}
Command @kbd{A} (@code{po-consider-as-auxiliary}) adds the current
PO file to the list of auxiliary files, while command @kbd{M-A}
(@code{po-ignore-as-auxiliary} just removes it.

@efindex a@r{, PO Mode command}
@efindex po-cycle-auxiliary@r{, PO Mode command}
The command @kbd{a} (@code{po-cycle-auxiliary}) seeks all auxiliary PO
files, round-robin, searching for a translated entry in some other language
having an @code{msgid} field identical as the one for the current entry.
The found PO file, if any, takes the place of the current PO file in
the display (its window gets on top).  Before doing so, the current PO
file is also made into an auxiliary file, if not already.  So, @kbd{a}
in this newly displayed PO file will seek another PO file, and so on,
so repeating @kbd{a} will eventually yield back the original PO file.

@efindex C-c C-a@r{, PO Mode command}
@efindex po-select-auxiliary@r{, PO Mode command}
The command @kbd{C-c C-a} (@code{po-select-auxiliary}) asks the translator
for her choice of a particular auxiliary file, with completion, and
then switches to that selected PO file.  The command also checks if
the selected file has an @code{msgid} field identical as the one for
the current entry, and if yes, this entry becomes current.  Otherwise,
the cursor of the selected file is left undisturbed.

For all this to work fully, auxiliary PO files will have to be normalized,
in that way that @code{msgid} fields should be written @emph{exactly}
the same way.  It is possible to write @code{msgid} fields in various
ways for representing the same string, different writing would break the
proper behaviour of the auxiliary file commands of PO mode.  This is not
expected to be much a problem in practice, as most existing PO files have
their @code{msgid} entries written by the same GNU @code{gettext} tools.

@efindex normalize@r{, PO Mode command}
However, PO files initially created by PO mode itself, while marking
strings in source files, are normalised differently.  So are PO
files resulting of the @samp{M-x normalize} command.  Until these
discrepancies between PO mode and other GNU @code{gettext} tools get
fully resolved, the translator should stay aware of normalisation issues.

@node Compendium,  , PO Mode, Editing
@section Using Translation Compendia
@emindex using translation compendia

@cindex compendium
A @dfn{compendium} is a special PO file containing a set of
translations recurring in many different packages.  The translator can
use gettext tools to build a new compendium, to add entries to her
compendium, and to initialize untranslated entries, or to update
already translated entries, from translations kept in the compendium.

@menu
* Creating Compendia::          Merging translations for later use
* Using Compendia::             Using older translations if they fit
@end menu

@node Creating Compendia, Using Compendia, Compendium, Compendium
@subsection Creating Compendia
@cindex creating compendia
@cindex compendium, creating

Basically every PO file consisting of translated entries only can be
declared as a valid compendium.  Often the translator wants to have
special compendia; let's consider two cases: @cite{concatenating PO
files} and @cite{extracting a message subset from a PO file}.

@subsubsection Concatenate PO Files

@cindex concatenating PO files into a compendium
@cindex accumulating translations
To concatenate several valid PO files into one compendium file you can
use @samp{msgcomm} or @samp{msgcat} (the latter preferred):

@example
msgcat -o compendium.po file1.po file2.po
@end example

By default, @code{msgcat} will accumulate divergent translations
for the same string.  Those occurrences will be marked as @code{fuzzy}
and highly visible decorated; calling @code{msgcat} on
@file{file1.po}:

@example
#: src/hello.c:200
#, c-format
msgid "Report bugs to <%s>.\n"
msgstr "Comunicar `bugs' a <%s>.\n"
@end example

@noindent
and @file{file2.po}:

@example
#: src/bye.c:100
#, c-format
msgid "Report bugs to <%s>.\n"
msgstr "Comunicar \"bugs\" a <%s>.\n"
@end example

@noindent
will result in:

@example
#: src/hello.c:200 src/bye.c:100
#, fuzzy, c-format
msgid "Report bugs to <%s>.\n"
msgstr ""
"#-#-#-#-#  file1.po  #-#-#-#-#\n"
"Comunicar `bugs' a <%s>.\n"
"#-#-#-#-#  file2.po  #-#-#-#-#\n"
"Comunicar \"bugs\" a <%s>.\n"
@end example

@noindent
The translator will have to resolve this ``conflict'' manually; she
has to decide whether the first or the second version is appropriate
(or provide a new translation), to delete the ``marker lines'', and
finally to remove the @code{fuzzy} mark.

If the translator knows in advance the first found translation of a
message is always the best translation she can make use to the
@samp{--use-first} switch:

@example
msgcat --use-first -o compendium.po file1.po file2.po
@end example

A good compendium file must not contain @code{fuzzy} or untranslated
entries.  If input files are ``dirty'' you must preprocess the input
files or postprocess the result using @samp{msgattrib --translated --no-fuzzy}.

@subsubsection Extract a Message Subset from a PO File
@cindex extracting parts of a PO file into a compendium

Nobody wants to translate the same messages again and again; thus you
may wish to have a compendium file containing @file{getopt.c} messages.

To extract a message subset (e.g., all @file{getopt.c} messages) from an
existing PO file into one compendium file you can use @samp{msggrep}:

@example
msggrep --location src/getopt.c -o compendium.po file.po
@end example

@node Using Compendia,  , Creating Compendia, Compendium
@subsection Using Compendia

You can use a compendium file to initialize a translation from scratch
or to update an already existing translation.

@subsubsection Initialize a New Translation File
@cindex initialize translations from a compendium

Since a PO file with translations does not exist the translator can
merely use @file{/dev/null} to fake the ``old'' translation file.

@example
msgmerge --compendium compendium.po -o file.po /dev/null file.pot
@end example

@subsubsection Update an Existing Translation File
@cindex update translations from a compendium

Concatenate the compendium file(s) and the existing PO, merge the
result with the POT file and remove the obsolete entries (optional,
here done using @samp{msgattrib}):

@example
msgcat --use-first -o update.po compendium1.po compendium2.po file.po
msgmerge update.po file.pot | msgattrib --no-obsolete > file.po
@end example

@node Manipulating, Binaries, Editing, Top
@chapter Manipulating PO Files
@cindex manipulating PO files

Sometimes it is necessary to manipulate PO files in a way that is better
performed automatically than by hand.  GNU @code{gettext} includes a
complete set of tools for this purpose.

@cindex merging two PO files
When merging two packages into a single package, the resulting POT file
will be the concatenation of the two packages' POT files.  Thus the
maintainer must concatenate the two existing package translations into
a single translation catalog, for each language.  This is best performed
using @samp{msgcat}.  It is then the translators' duty to deal with any
possible conflicts that arose during the merge.

@cindex encoding conversion
When a translator takes over the translation job from another translator,
but she uses a different character encoding in her locale, she will
convert the catalog to her character encoding.  This is best done through
the @samp{msgconv} program.

When a maintainer takes a source file with tagged messages from another
package, he should also take the existing translations for this source
file (and not let the translators do the same job twice).  One way to do
this is through @samp{msggrep}, another is to create a POT file for
that source file and use @samp{msgmerge}.

@cindex dialect
@cindex orthography
When a translator wants to adjust some translation catalog for a special
dialect or orthography --- for example, German as written in Switzerland
versus German as written in Germany --- she needs to apply some text
processing to every message in the catalog.  The tool for doing this is
@samp{msgfilter}.

Another use of @code{msgfilter} is to produce approximately the POT file for
which a given PO file was made.  This can be done through a filter command
like @samp{msgfilter sed -e d | sed -e '/^# /d'}.  Note that the original
POT file may have had different comments and different plural message counts,
that's why it's better to use the original POT file if available.

@cindex checking of translations
When a translator wants to check her translations, for example according
to orthography rules or using a non-interactive spell checker, she can do
so using the @samp{msgexec} program.

@cindex duplicate elimination
When third party tools create PO or POT files, sometimes duplicates cannot
be avoided.  But the GNU @code{gettext} tools give an error when they
encounter duplicate msgids in the same file and in the same domain.
To merge duplicates, the @samp{msguniq} program can be used.

@samp{msgcomm} is a more general tool for keeping or throwing away
duplicates, occurring in different files.

@samp{msgcmp} can be used to check whether a translation catalog is
completely translated.

@cindex attributes, manipulating
@samp{msgattrib} can be used to select and extract only the fuzzy
or untranslated messages of a translation catalog.

@samp{msgen} is useful as a first step for preparing English translation
catalogs.  It copies each message's msgid to its msgstr.

Finally, for those applications where all these various programs are not
sufficient, a library @samp{libgettextpo} is provided that can be used to
write other specialized programs that process PO files.

@menu
* msgcat Invocation::           Invoking the @code{msgcat} Program
* msgconv Invocation::          Invoking the @code{msgconv} Program
* msggrep Invocation::          Invoking the @code{msggrep} Program
* msgfilter Invocation::        Invoking the @code{msgfilter} Program
* msguniq Invocation::          Invoking the @code{msguniq} Program
* msgcomm Invocation::          Invoking the @code{msgcomm} Program
* msgcmp Invocation::           Invoking the @code{msgcmp} Program
* msgattrib Invocation::        Invoking the @code{msgattrib} Program
* msgen Invocation::            Invoking the @code{msgen} Program
* msgexec Invocation::          Invoking the @code{msgexec} Program
* Colorizing::                  Highlighting parts of PO files
* libgettextpo::                Writing your own programs that process PO files
@end menu

@node msgcat Invocation, msgconv Invocation, Manipulating, Manipulating
@section Invoking the @code{msgcat} Program

@include msgcat.texi

@node msgconv Invocation, msggrep Invocation, msgcat Invocation, Manipulating
@section Invoking the @code{msgconv} Program

@include msgconv.texi

@node msggrep Invocation, msgfilter Invocation, msgconv Invocation, Manipulating
@section Invoking the @code{msggrep} Program

@include msggrep.texi

@node msgfilter Invocation, msguniq Invocation, msggrep Invocation, Manipulating
@section Invoking the @code{msgfilter} Program

@include msgfilter.texi

@node msguniq Invocation, msgcomm Invocation, msgfilter Invocation, Manipulating
@section Invoking the @code{msguniq} Program

@include msguniq.texi

@node msgcomm Invocation, msgcmp Invocation, msguniq Invocation, Manipulating
@section Invoking the @code{msgcomm} Program

@include msgcomm.texi

@node msgcmp Invocation, msgattrib Invocation, msgcomm Invocation, Manipulating
@section Invoking the @code{msgcmp} Program

@include msgcmp.texi

@node msgattrib Invocation, msgen Invocation, msgcmp Invocation, Manipulating
@section Invoking the @code{msgattrib} Program

@include msgattrib.texi

@node msgen Invocation, msgexec Invocation, msgattrib Invocation, Manipulating
@section Invoking the @code{msgen} Program

@include msgen.texi

@node msgexec Invocation, Colorizing, msgen Invocation, Manipulating
@section Invoking the @code{msgexec} Program

@include msgexec.texi

@node Colorizing, libgettextpo, msgexec Invocation, Manipulating
@section Highlighting parts of PO files

Translators are usually only interested in seeing the untranslated and
fuzzy messages of a PO file.  Also, when a message is set fuzzy because
the msgid changed, they want to see the differences between the previous
msgid and the current one (especially if the msgid is long and only few
words in it have changed).  Finally, it's always welcome to highlight the
different sections of a message in a PO file (comments, msgid, msgstr, etc.).

Such highlighting is possible through the @code{msgcat} options
@samp{--color} and @samp{--style}.

@menu
* The --color option::          Triggering colorized output
* The TERM variable::           The environment variable @code{TERM}
* The --style option::          The @code{--style} option
* Style rules::                 Style rules for PO files
* Customizing less::            Customizing @code{less} for viewing PO files
@end menu

@node The --color option, The TERM variable, Colorizing, Colorizing
@subsection The @code{--color} option

@opindex --color@r{, @code{msgcat} option}
The @samp{--color=@var{when}} option specifies under which conditions
colorized output should be generated.  The @var{when} part can be one of
the following:

@table @code
@item always
@itemx yes
The output will be colorized.

@item never
@itemx no
The output will not be colorized.

@item auto
@itemx tty
The output will be colorized if the output device is a tty, i.e.@: when the
output goes directly to a text screen or terminal emulator window.

@item html
The output will be colorized and be in HTML format.
@end table

@noindent
@samp{--color} is equivalent to @samp{--color=yes}.  The default is
@samp{--color=auto}.

Thus, a command like @samp{msgcat vi.po} will produce colorized output
when called by itself in a command window.  Whereas in a pipe, such as
@samp{msgcat vi.po | less -R}, it will not produce colorized output.  To
get colorized output in this situation nevertheless, use the command
@samp{msgcat --color vi.po | less -R}.

The @samp{--color=html} option will produce output that can be viewed in
a browser.  This can be useful, for example, for Indic languages,
because the renderic of Indic scripts in browser is usually better than
in terminal emulators.

Note that the output produced with the @code{--color} option is @emph{not}
a valid PO file in itself.  It contains additional terminal-specific escape
sequences or HTML tags.  A PO file reader will give a syntax error when
confronted with such content.  Except for the @samp{--color=html} case,
you therefore normally don't need to save output produced with the
@code{--color} option in a file.

@node The TERM variable, The --style option, The --color option, Colorizing
@subsection The environment variable @code{TERM}

@vindex TERM@r{, environment variable}
The environment variable @code{TERM} contains a identifier for the text
window's capabilities.  You can get a detailed list of these cababilities
by using the @samp{infocmp} command, using @samp{man 5 terminfo} as a
reference.

When producing text with embedded color directives, @code{msgcat} looks
at the @code{TERM} variable.  Text windows today typically support at least
8 colors.  Often, however, the text window supports 16 or more colors,
even though the @code{TERM} variable is set to a identifier denoting only
8 supported colors.  It can be worth setting the @code{TERM} variable to
a different value in these cases:

@table @code
@item xterm
@code{xterm} is in most cases built with support for 16 colors.  It can also
be built with support for 88 or 256 colors (but not both).  You can try to
set @code{TERM} to either @code{xterm-16color}, @code{xterm-88color}, or
@code{xterm-256color}.

@item rxvt
@code{rxvt} is often built with support for 16 colors.  You can try to set
@code{TERM} to @code{rxvt-16color}.

@item konsole
@code{konsole} too is often built with support for 16 colors.  You can try to
set @code{TERM} to @code{konsole-16color} or @code{xterm-16color}.
@end table

After setting @code{TERM}, you can verify it by invoking
@samp{msgcat --color=test} and seeing whether the output looks like a
reasonable color map.

@node The --style option, Style rules, The TERM variable, Colorizing
@subsection The @code{--style} option

@opindex --style@r{, @code{msgcat} option}
The @samp{--style=@var{style_file}} option specifies the style file to use
when colorizing.  It has an effect only when the @code{--color} option is
effective.

@vindex PO_STYLE@r{, environment variable}
If the @code{--style} option is not specified, the environment variable
@code{PO_STYLE} is considered.  It is meant to point to the user's
preferred style for PO files.

The default style file is @file{$prefix/share/gettext/styles/po-default.css},
where @code{$prefix} is the installation location.

A few style files are predefined:
@table @file
@item po-vim.css
This style imitates the look used by vim 7.

@item po-emacs-x.css
This style imitates the look used by GNU Emacs 21 and 22 in an X11 window.

@item po-emacs-xterm.css
@itemx po-emacs-xterm16.css
@itemx po-emacs-xterm256.css
This style imitates the look used by GNU Emacs 22 in a terminal of type
@samp{xterm} (8 colors) or @samp{xterm-16color} (16 colors) or
@samp{xterm-256color} (256 colors), respectively.
@end table

@noindent
You can use these styles without specifying a directory.  They are actually
located in @file{$prefix/share/gettext/styles/}, where @code{$prefix} is the
installation location.

You can also design your own styles.  This is described in the next section.


@node Style rules, Customizing less, The --style option, Colorizing
@subsection Style rules for PO files

The same style file can be used for styling of a PO file, for terminal
output and for HTML output.  It is written in CSS (Cascading Style Sheet)
syntax.  See @url{http://www.w3.org/TR/css2/cover.html} for a formal
definition of CSS.  Many HTML authoring tutorials also contain explanations
of CSS.

In the case of HTML output, the style file is embedded in the HTML output.
In the case of text output, the style file is interpreted by the
@code{msgcat} program.  This means, in particular, that when
@code{@@import} is used with relative file names, the file names are

@itemize @minus
@item
relative to the resulting HTML file, in the case of HTML output,

@item
relative to the style sheet containing the @code{@@import}, in the case of
text output.  (Actually, @code{@@import}s are not yet supported in this case,
due to a limitation in @code{libcroco}.)
@end itemize

CSS rules are built up from selectors and declarations.  The declarations
specify graphical properties; the selectors specify specify when they apply.

In PO files, the following simple selectors (based on "CSS classes", see
the CSS2 spec, section 5.8.3) are supported.

@itemize @bullet
@item
Selectors that apply to entire messages:

@table @code
@item .header
This matches the header entry of a PO file.

@item .translated
This matches a translated message.

@item .untranslated
This matches an untranslated message (i.e.@: a message with empty translation).

@item .fuzzy
This matches a fuzzy message (i.e.@: a message which has a translation that
needs review by the translator).

@item .obsolete
This matches an obsolete message (i.e.@: a message that was translated but is
not needed by the current POT file any more).
@end table

@item
Selectors that apply to parts of a message in PO syntax.  Recall the general
structure of a message in PO syntax:

@example
@var{white-space}
#  @var{translator-comments}
#. @var{extracted-comments}
#: @var{reference}@dots{}
#, @var{flag}@dots{}
#| msgid @var{previous-untranslated-string}
msgid @var{untranslated-string}
msgstr @var{translated-string}
@end example

@table @code
@item .comment
This matches all comments (translator comments, extracted comments,
source file reference comments, flag comments, previous message comments,
as well as the entire obsolete messages).

@item .translator-comment
This matches the translator comments.

@item .extracted-comment
This matches the extracted comments, i.e.@: the comments placed by the
programmer at the attention of the translator.

@item .reference-comment
This matches the source file reference comments (entire lines).

@item .reference
This matches the individual source file references inside the source file
reference comment lines.

@item .flag-comment
This matches the flag comment lines (entire lines).

@item .flag
This matches the individual flags inside flag comment lines.

@item .fuzzy-flag
This matches the `fuzzy' flag inside flag comment lines.

@item .previous-comment
This matches the comments containing the previous untranslated string (entire
lines).

@item .previous
This matches the previous untranslated string including the string delimiters,
the associated keywords (@code{msgid} etc.) and the spaces between them.

@item .msgid
This matches the untranslated string including the string delimiters,
the associated keywords (@code{msgid} etc.) and the spaces between them.

@item .msgstr
This matches the translated string including the string delimiters,
the associated keywords (@code{msgstr} etc.) and the spaces between them.

@item .keyword
This matches the keywords (@code{msgid}, @code{msgstr}, etc.).

@item .string
This matches strings, including the string delimiters (double quotes).
@end table

@item
Selectors that apply to parts of strings:

@table @code
@item .text
This matches the entire contents of a string (excluding the string delimiters,
i.e.@: the double quotes).

@item .escape-sequence
This matches an escape sequence (starting with a backslash).

@item .format-directive
This matches a format string directive (starting with a @samp{%} sign in the
case of most programming languages, with a @samp{@{} in the case of
@code{java-format} and @code{csharp-format}, with a @samp{~} in the case of
@code{lisp-format} and @code{scheme-format}, or with @samp{$} in the case of
@code{sh-format}).

@item .invalid-format-directive
This matches an invalid format string directive.

@item .added
In an untranslated string, this matches a part of the string that was not
present in the previous untranslated string.  (Not yet implemented in this
release.)

@item .changed
In an untranslated string or in a previous untranslated string, this matches
a part of the string that is changed or replaced.  (Not yet implemented in
this release.)

@item .removed
In a previous untranslated string, this matches a part of the string that
is not present in the current untranslated string.  (Not yet implemented in
this release.)
@end table
@end itemize

These selectors can be combined to hierarchical selectors.  For example,

@smallexample
.msgstr .invalid-format-directive @{ color: red; @}
@end smallexample

@noindent
will highlight the invalid format directives in the translated strings.

In text mode, pseudo-classes (CSS2 spec, section 5.11) and pseudo-elements
(CSS2 spec, section 5.12) are not supported.

The declarations in HTML mode are not limited; any graphical attribute
supported by the browsers can be used.

The declarations in text mode are limited to the following properties.  Other
properties will be silently ignored.

@table @asis
@item @code{color} (CSS2 spec, section 14.1)
@itemx @code{background-color} (CSS2 spec, section 14.2.1)
These properties is supported.  Colors will be adjusted to match the terminal's
capabilities.  Note that many terminals support only 8 colors.

@item @code{font-weight} (CSS2 spec, section 15.2.3)
This property is supported, but most terminals can only render two different
weights: @code{normal} and @code{bold}.  Values >= 600 are rendered as
@code{bold}.

@item @code{font-style} (CSS2 spec, section 15.2.3)
This property is supported.  The values @code{italic} and @code{oblique} are
rendered the same way.

@item @code{text-decoration} (CSS2 spec, section 16.3.1)
This property is supported, limited to the values @code{none} and
@code{underline}.
@end table

@node Customizing less,  , Style rules, Colorizing
@subsection Customizing @code{less} for viewing PO files

The @samp{less} program is a popular text file browser for use in a text
screen or terminal emulator.  It also supports text with embedded escape
sequences for colors and text decorations.

You can use @code{less} to view a PO file like this (assuming an UTF-8
environment):

@smallexample
msgcat --to-code=UTF-8 --color xyz.po | less -R
@end smallexample

You can simplify this to this simple command:

@smallexample
less xyz.po
@end smallexample

@noindent
after these three preparations:

@enumerate
@item
Add the options @samp{-R} and @samp{-f} to the @code{LESS} environment
variable.  In sh shells:
@smallexample
$ LESS="$LESS -R -f"
$ export LESS
@end smallexample

@item
If your system does not already have the @file{lessopen.sh} and
@file{lessclose.sh} scripts, create them and set the @code{LESSOPEN} and
@code{LESSCLOSE} environment variables, as indicated in the manual page
(@samp{man less}).

@item
Add to @file{lessopen.sh} a piece of script that recognizes PO files
through their file extension and invokes @code{msgcat} on them, producing
a temporary file.  Like this:

@smallexample
case "$1" in
  *.po)
    tmpfile=`mktemp "$@{TMPDIR-/tmp@}/less.XXXXXX"`
    msgcat --to-code=UTF-8 --color "$1" > "$tmpfile"
    echo "$tmpfile"
    exit 0
    ;;
esac
@end smallexample
@end enumerate

@node libgettextpo,  , Colorizing, Manipulating
@section Writing your own programs that process PO files

For the tasks for which a combination of @samp{msgattrib}, @samp{msgcat} etc.
is not sufficient, a set of C functions is provided in a library, to make it
possible to process PO files in your own programs.  When you use this library,
you don't need to write routines to parse the PO file; instead, you retrieve
a pointer in memory to each of messages contained in the PO file.  Functions
for writing PO files are not provided at this time.

The functions are declared in the header file @samp{<gettext-po.h>}, and are
defined in a library called @samp{libgettextpo}.

@deftp {Data Type} po_file_t
This is a pointer type that refers to the contents of a PO file, after it has
been read into memory.
@end deftp

@deftp {Data Type} po_message_iterator_t
This is a pointer type that refers to an iterator that produces a sequence of
messages.
@end deftp

@deftp {Data Type} po_message_t
This is a pointer type that refers to a message of a PO file, including its
translation.
@end deftp

@deftypefun po_file_t po_file_read (const char *@var{filename})
The @code{po_file_read} function reads a PO file into memory.  The file name
is given as argument.  The return value is a handle to the PO file's contents,
valid until @code{po_file_free} is called on it.  In case of error, the return
value is @code{NULL}, and @code{errno} is set.
@end deftypefun

@deftypefun void po_file_free (po_file_t @var{file})
The @code{po_file_free} function frees a PO file's contents from memory,
including all messages that are only implicitly accessible through iterators.
@end deftypefun

@deftypefun {const char * const *} po_file_domains (po_file_t @var{file})
The @code{po_file_domains} function returns the domains for which the given
PO file has messages.  The return value is a @code{NULL} terminated array
which is valid as long as the @var{file} handle is valid.  For PO files which
contain no @samp{domain} directive, the return value contains only one domain,
namely the default domain @code{"messages"}.
@end deftypefun

@deftypefun po_message_iterator_t po_message_iterator (po_file_t @var{file}, const char *@var{domain})
The @code{po_message_iterator} returns an iterator that will produce the
messages of @var{file} that belong to the given @var{domain}.  If @var{domain}
is @code{NULL}, the default domain is used instead.  To list the messages,
use the function @code{po_next_message} repeatedly.
@end deftypefun

@deftypefun void po_message_iterator_free (po_message_iterator_t @var{iterator})
The @code{po_message_iterator_free} function frees an iterator previously
allocated through the @code{po_message_iterator} function.
@end deftypefun

@deftypefun po_message_t po_next_message (po_message_iterator_t @var{iterator})
The @code{po_next_message} function returns the next message from
@var{iterator} and advances the iterator.  It returns @code{NULL} when the
iterator has reached the end of its message list.
@end deftypefun

The following functions returns details of a @code{po_message_t}.  Recall
that the results are valid as long as the @var{file} handle is valid.

@deftypefun {const char *} po_message_msgid (po_message_t @var{message})
The @code{po_message_msgid} function returns the @code{msgid} (untranslated
English string) of a message.  This is guaranteed to be non-@code{NULL}.
@end deftypefun

@deftypefun {const char *} po_message_msgid_plural (po_message_t @var{message})
The @code{po_message_msgid_plural} function returns the @code{msgid_plural}
(untranslated English plural string) of a message with plurals, or @code{NULL}
for a message without plural.
@end deftypefun

@deftypefun {const char *} po_message_msgstr (po_message_t @var{message})
The @code{po_message_msgstr} function returns the @code{msgstr} (translation)
of a message.  For an untranslated message, the return value is an empty
string.
@end deftypefun

@deftypefun {const char *} po_message_msgstr_plural (po_message_t @var{message}, int @var{index})
The @code{po_message_msgstr_plural} function returns the
@code{msgstr[@var{index}]} of a message with plurals, or @code{NULL} when
the @var{index} is out of range or for a message without plural.
@end deftypefun

Here is an example code how these functions can be used.

@example
const char *filename = @dots{};
po_file_t file = po_file_read (filename);

if (file == NULL)
  error (EXIT_FAILURE, errno, "couldn't open the PO file %s", filename);
@{
  const char * const *domains = po_file_domains (file);
  const char * const *domainp;

  for (domainp = domains; *domainp; domainp++)
    @{
      const char *domain = *domainp;
      po_message_iterator_t iterator = po_message_iterator (file, domain);

      for (;;)
        @{
          po_message_t *message = po_next_message (iterator);

          if (message == NULL)
            break;
          @{
            const char *msgid = po_message_msgid (message);
            const char *msgstr = po_message_msgstr (message);

            @dots{}
          @}
        @}
      po_message_iterator_free (iterator);
    @}
@}
po_file_free (file);
@end example

@node Binaries, Programmers, Manipulating, Top
@chapter Producing Binary MO Files

@c FIXME: Rewrite.

@menu
* msgfmt Invocation::           Invoking the @code{msgfmt} Program
* msgunfmt Invocation::         Invoking the @code{msgunfmt} Program
* MO Files::                    The Format of GNU MO Files
@end menu

@node msgfmt Invocation, msgunfmt Invocation, Binaries, Binaries
@section Invoking the @code{msgfmt} Program

@include msgfmt.texi

@node msgunfmt Invocation, MO Files, msgfmt Invocation, Binaries
@section Invoking the @code{msgunfmt} Program

@include msgunfmt.texi

@node MO Files,  , msgunfmt Invocation, Binaries
@section The Format of GNU MO Files
@cindex MO file's format
@cindex file format, @file{.mo}

The format of the generated MO files is best described by a picture,
which appears below.

@cindex magic signature of MO files
The first two words serve the identification of the file.  The magic
number will always signal GNU MO files.  The number is stored in the
byte order of the generating machine, so the magic number really is
two numbers: @code{0x950412de} and @code{0xde120495}.

The second word describes the current revision of the file format,
composed of a major and a minor revision number.  The revision numbers
ensure that the readers of MO files can distinguish new formats from
old ones and handle their contents, as far as possible.  For now the
major revision is 0 or 1, and the minor revision is also 0 or 1.  More
revisions might be added in the future.  A program seeing an unexpected
major revision number should stop reading the MO file entirely; whereas
an unexpected minor revision number means that the file can be read but
will not reveal its full contents, when parsed by a program that
supports only smaller minor revision numbers.

The version is kept
separate from the magic number, instead of using different magic
numbers for different formats, mainly because @file{/etc/magic} is
not updated often.

Follow a number of pointers to later tables in the file, allowing
for the extension of the prefix part of MO files without having to
recompile programs reading them.  This might become useful for later
inserting a few flag bits, indication about the charset used, new
tables, or other things.

Then, at offset @var{O} and offset @var{T} in the picture, two tables
of string descriptors can be found.  In both tables, each string
descriptor uses two 32 bits integers, one for the string length,
another for the offset of the string in the MO file, counting in bytes
from the start of the file.  The first table contains descriptors
for the original strings, and is sorted so the original strings
are in increasing lexicographical order.  The second table contains
descriptors for the translated strings, and is parallel to the first
table: to find the corresponding translation one has to access the
array slot in the second array with the same index.

Having the original strings sorted enables the use of simple binary
search, for when the MO file does not contain an hashing table, or
for when it is not practical to use the hashing table provided in
the MO file.  This also has another advantage, as the empty string
in a PO file GNU @code{gettext} is usually @emph{translated} into
some system information attached to that particular MO file, and the
empty string necessarily becomes the first in both the original and
translated tables, making the system information very easy to find.

@cindex hash table, inside MO files
The size @var{S} of the hash table can be zero.  In this case, the
hash table itself is not contained in the MO file.  Some people might
prefer this because a precomputed hashing table takes disk space, and
does not win @emph{that} much speed.  The hash table contains indices
to the sorted array of strings in the MO file.  Conflict resolution is
done by double hashing.  The precise hashing algorithm used is fairly
dependent on GNU @code{gettext} code, and is not documented here.

As for the strings themselves, they follow the hash file, and each
is terminated with a @key{NUL}, and this @key{NUL} is not counted in
the length which appears in the string descriptor.  The @code{msgfmt}
program has an option selecting the alignment for MO file strings.
With this option, each string is separately aligned so it starts at
an offset which is a multiple of the alignment value.  On some RISC
machines, a correct alignment will speed things up.

@cindex context, in MO files
Contexts are stored by storing the concatenation of the context, a
@key{EOT} byte, and the original string, instead of the original string.

@cindex plural forms, in MO files
Plural forms are stored by letting the plural of the original string
follow the singular of the original string, separated through a
@key{NUL} byte.  The length which appears in the string descriptor
includes both.  However, only the singular of the original string
takes part in the hash table lookup.  The plural variants of the
translation are all stored consecutively, separated through a
@key{NUL} byte.  Here also, the length in the string descriptor
includes all of them.

Nothing prevents a MO file from having embedded @key{NUL}s in strings.
However, the program interface currently used already presumes
that strings are @key{NUL} terminated, so embedded @key{NUL}s are
somewhat useless.  But the MO file format is general enough so other
interfaces would be later possible, if for example, we ever want to
implement wide characters right in MO files, where @key{NUL} bytes may
accidentally appear.  (No, we don't want to have wide characters in MO
files.  They would make the file unnecessarily large, and the
@samp{wchar_t} type being platform dependent, MO files would be
platform dependent as well.)

This particular issue has been strongly debated in the GNU
@code{gettext} development forum, and it is expectable that MO file
format will evolve or change over time.  It is even possible that many
formats may later be supported concurrently.  But surely, we have to
start somewhere, and the MO file format described here is a good start.
Nothing is cast in concrete, and the format may later evolve fairly
easily, so we should feel comfortable with the current approach.

@example
@group
        byte
             +------------------------------------------+
          0  | magic number = 0x950412de                |
             |                                          |
          4  | file format revision = 0                 |
             |                                          |
          8  | number of strings                        |  == N
             |                                          |
         12  | offset of table with original strings    |  == O
             |                                          |
         16  | offset of table with translation strings |  == T
             |                                          |
         20  | size of hashing table                    |  == S
             |                                          |
         24  | offset of hashing table                  |  == H
             |                                          |
             .                                          .
             .    (possibly more entries later)         .
             .                                          .
             |                                          |
          O  | length & offset 0th string  ----------------.
      O + 8  | length & offset 1st string  ------------------.
              ...                                    ...   | |
O + ((N-1)*8)| length & offset (N-1)th string           |  | |
             |                                          |  | |
          T  | length & offset 0th translation  ---------------.
      T + 8  | length & offset 1st translation  -----------------.
              ...                                    ...   | | | |
T + ((N-1)*8)| length & offset (N-1)th translation      |  | | | |
             |                                          |  | | | |
          H  | start hash table                         |  | | | |
              ...                                    ...   | | | |
  H + S * 4  | end hash table                           |  | | | |
             |                                          |  | | | |
             | NUL terminated 0th string  <----------------' | | |
             |                                          |    | | |
             | NUL terminated 1st string  <------------------' | |
             |                                          |      | |
              ...                                    ...       | |
             |                                          |      | |
             | NUL terminated 0th translation  <---------------' |
             |                                          |        |
             | NUL terminated 1st translation  <-----------------'
             |                                          |
              ...                                    ...
             |                                          |
             +------------------------------------------+
@end group
@end example

@node Programmers, Translators, Binaries, Top
@chapter The Programmer's View

@c FIXME: Reorganize whole chapter.

One aim of the current message catalog implementation provided by
GNU @code{gettext} was to use the system's message catalog handling, if the
installer wishes to do so.  So we perhaps should first take a look at
the solutions we know about.  The people in the POSIX committee did not
manage to agree on one of the semi-official standards which we'll
describe below.  In fact they couldn't agree on anything, so they decided
only to include an example of an interface.  The major Unix vendors
are split in the usage of the two most important specifications: X/Open's
catgets vs. Uniforum's gettext interface.  We'll describe them both and
later explain our solution of this dilemma.

@menu
* catgets::                     About @code{catgets}
* gettext::                     About @code{gettext}
* Comparison::                  Comparing the two interfaces
* Using libintl.a::             Using libintl.a in own programs
* gettext grok::                Being a @code{gettext} grok
* Temp Programmers::            Temporary Notes for the Programmers Chapter
@end menu

@node catgets, gettext, Programmers, Programmers
@section About @code{catgets}
@cindex @code{catgets}, X/Open specification

The @code{catgets} implementation is defined in the X/Open Portability
Guide, Volume 3, XSI Supplementary Definitions, Chapter 5.  But the
process of creating this standard seemed to be too slow for some of
the Unix vendors so they created their implementations on preliminary
versions of the standard.  Of course this leads again to problems while
writing platform independent programs: even the usage of @code{catgets}
does not guarantee a unique interface.

Another, personal comment on this that only a bunch of committee members
could have made this interface.  They never really tried to program
using this interface.  It is a fast, memory-saving implementation, an
user can happily live with it.  But programmers hate it (at least I and
some others do@dots{})

But we must not forget one point: after all the trouble with transferring
the rights on Unix(tm) they at last came to X/Open, the very same who
published this specification.  This leads me to making the prediction
that this interface will be in future Unix standards (e.g.@: Spec1170) and
therefore part of all Unix implementation (implementations, which are
@emph{allowed} to wear this name).

@menu
* Interface to catgets::        The interface
* Problems with catgets::       Problems with the @code{catgets} interface?!
@end menu

@node Interface to catgets, Problems with catgets, catgets, catgets
@subsection The Interface
@cindex interface to @code{catgets}

The interface to the @code{catgets} implementation consists of three
functions which correspond to those used in file access: @code{catopen}
to open the catalog for using, @code{catgets} for accessing the message
tables, and @code{catclose} for closing after work is done.  Prototypes
for the functions and the needed definitions are in the
@code{<nl_types.h>} header file.

@cindex @code{catopen}, a @code{catgets} function
@code{catopen} is used like in this:

@example
nl_catd catd = catopen ("catalog_name", 0);
@end example

The function takes as the argument the name of the catalog.  This usual
refers to the name of the program or the package.  The second parameter
is not further specified in the standard.  I don't even know whether it
is implemented consistently among various systems.  So the common advice
is to use @code{0} as the value.  The return value is a handle to the
message catalog, equivalent to handles to file returned by @code{open}.

@cindex @code{catgets}, a @code{catgets} function
This handle is of course used in the @code{catgets} function which can
be used like this:

@example
char *translation = catgets (catd, set_no, msg_id, "original string");
@end example

The first parameter is this catalog descriptor.  The second parameter
specifies the set of messages in this catalog, in which the message
described by @code{msg_id} is obtained.  @code{catgets} therefore uses a
three-stage addressing:

@display
catalog name @result{} set number @result{} message ID @result{} translation
@end display

@c Anybody else loving Haskell??? :-) -- Uli

The fourth argument is not used to address the translation.  It is given
as a default value in case when one of the addressing stages fail.  One
important thing to remember is that although the return type of catgets
is @code{char *} the resulting string @emph{must not} be changed.  It
should better be @code{const char *}, but the standard is published in
1988, one year before ANSI C.

@noindent
@cindex @code{catclose}, a @code{catgets} function
The last of these functions is used and behaves as expected:

@example
catclose (catd);
@end example

After this no @code{catgets} call using the descriptor is legal anymore.

@node Problems with catgets,  , Interface to catgets, catgets
@subsection Problems with the @code{catgets} Interface?!
@cindex problems with @code{catgets} interface

Now that this description seemed to be really easy --- where are the
problems we speak of?  In fact the interface could be used in a
reasonable way, but constructing the message catalogs is a pain.  The
reason for this lies in the third argument of @code{catgets}: the unique
message ID.  This has to be a numeric value for all messages in a single
set.  Perhaps you could imagine the problems keeping such a list while
changing the source code.  Add a new message here, remove one there.  Of
course there have been developed a lot of tools helping to organize this
chaos but one as the other fails in one aspect or the other.  We don't
want to say that the other approach has no problems but they are far
more easy to manage.

@node gettext, Comparison, catgets, Programmers
@section About @code{gettext}
@cindex @code{gettext}, a programmer's view

The definition of the @code{gettext} interface comes from a Uniforum
proposal.  It was submitted there by Sun, who had implemented the
@code{gettext} function in SunOS 4, around 1990.  Nowadays, the
@code{gettext} interface is specified by the OpenI18N standard.

The main point about this solution is that it does not follow the
method of normal file handling (open-use-close) and that it does not
burden the programmer with so many tasks, especially the unique key handling.
Of course here also a unique key is needed, but this key is the message
itself (how long or short it is).  See @ref{Comparison} for a more
detailed comparison of the two methods.

The following section contains a rather detailed description of the
interface.  We make it that detailed because this is the interface
we chose for the GNU @code{gettext} Library.  Programmers interested
in using this library will be interested in this description.

@menu
* Interface to gettext::        The interface
* Ambiguities::                 Solving ambiguities
* Locating Catalogs::           Locating message catalog files
* Charset conversion::          How to request conversion to Unicode
* Contexts::                    Solving ambiguities in GUI programs
* Plural forms::                Additional functions for handling plurals
* Optimized gettext::           Optimization of the *gettext functions
@end menu

@node Interface to gettext, Ambiguities, gettext, gettext
@subsection The Interface
@cindex @code{gettext} interface

The minimal functionality an interface must have is a) to select a
domain the strings are coming from (a single domain for all programs is
not reasonable because its construction and maintenance is difficult,
perhaps impossible) and b) to access a string in a selected domain.

This is principally the description of the @code{gettext} interface.  It
has a global domain which unqualified usages reference.  Of course this
domain is selectable by the user.

@example
char *textdomain (const char *domain_name);
@end example

This provides the possibility to change or query the current status of
the current global domain of the @code{LC_MESSAGE} category.  The
argument is a null-terminated string, whose characters must be legal in
the use in filenames.  If the @var{domain_name} argument is @code{NULL},
the function returns the current value.  If no value has been set
before, the name of the default domain is returned: @emph{messages}.
Please note that although the return value of @code{textdomain} is of
type @code{char *} no changing is allowed.  It is also important to know
that no checks of the availability are made.  If the name is not
available you will see this by the fact that no translations are provided.

@noindent
To use a domain set by @code{textdomain} the function

@example
char *gettext (const char *msgid);
@end example

@noindent
is to be used.  This is the simplest reasonable form one can imagine.
The translation of the string @var{msgid} is returned if it is available
in the current domain.  If it is not available, the argument itself is
returned.  If the argument is @code{NULL} the result is undefined.

One thing which should come into mind is that no explicit dependency to
the used domain is given.  The current value of the domain is used.
If this changes between two
executions of the same @code{gettext} call in the program, both calls
reference a different message catalog.

For the easiest case, which is normally used in internationalized
packages, once at the beginning of execution a call to @code{textdomain}
is issued, setting the domain to a unique name, normally the package
name.  In the following code all strings which have to be translated are
filtered through the gettext function.  That's all, the package speaks
your language.

@node Ambiguities, Locating Catalogs, Interface to gettext, gettext
@subsection Solving Ambiguities
@cindex several domains
@cindex domain ambiguities
@cindex large package

While this single name domain works well for most applications there
might be the need to get translations from more than one domain.  Of
course one could switch between different domains with calls to
@code{textdomain}, but this is really not convenient nor is it fast.  A
possible situation could be one case subject to discussion during this
writing:  all
error messages of functions in the set of common used functions should
go into a separate domain @code{error}.  By this mean we would only need
to translate them once.
Another case are messages from a library, as these @emph{have} to be
independent of the current domain set by the application.

@noindent
For this reasons there are two more functions to retrieve strings:

@example
char *dgettext (const char *domain_name, const char *msgid);
char *dcgettext (const char *domain_name, const char *msgid,
                 int category);
@end example

Both take an additional argument at the first place, which corresponds
to the argument of @code{textdomain}.  The third argument of
@code{dcgettext} allows to use another locale category but @code{LC_MESSAGES}.
But I really don't know where this can be useful.  If the
@var{domain_name} is @code{NULL} or @var{category} has an value beside
the known ones, the result is undefined.  It should also be noted that
this function is not part of the second known implementation of this
function family, the one found in Solaris.

A second ambiguity can arise by the fact, that perhaps more than one
domain has the same name.  This can be solved by specifying where the
needed message catalog files can be found.

@example
char *bindtextdomain (const char *domain_name,
                      const char *dir_name);
@end example

Calling this function binds the given domain to a file in the specified
directory (how this file is determined follows below).  Especially a
file in the systems default place is not favored against the specified
file anymore (as it would be by solely using @code{textdomain}).  A
@code{NULL} pointer for the @var{dir_name} parameter returns the binding
associated with @var{domain_name}.  If @var{domain_name} itself is
@code{NULL} nothing happens and a @code{NULL} pointer is returned.  Here
again as for all the other functions is true that none of the return
value must be changed!

It is important to remember that relative path names for the
@var{dir_name} parameter can be trouble.  Since the path is always
computed relative to the current directory different results will be
achieved when the program executes a @code{chdir} command.  Relative
paths should always be avoided to avoid dependencies and
unreliabilities.

@node Locating Catalogs, Charset conversion, Ambiguities, gettext
@subsection Locating Message Catalog Files
@cindex message catalog files location

Because many different languages for many different packages have to be
stored we need some way to add these information to file message catalog
files.  The way usually used in Unix environments is have this encoding
in the file name.  This is also done here.  The directory name given in
@code{bindtextdomain}s second argument (or the default directory),
followed by the name of the locale, the locale category, and the domain name
are concatenated:

@example
@var{dir_name}/@var{locale}/LC_@var{category}/@var{domain_name}.mo
@end example

The default value for @var{dir_name} is system specific.  For the GNU
library, and for packages adhering to its conventions, it's:
@example
/usr/local/share/locale
@end example

@noindent
@var{locale} is the name of the locale category which is designated by
@code{LC_@var{category}}.  For @code{gettext} and @code{dgettext} this
@code{LC_@var{category}} is always @code{LC_MESSAGES}.@footnote{Some
system, e.g.@: mingw, don't have @code{LC_MESSAGES}.  Here we use a more or
less arbitrary value for it, namely 1729, the smallest positive integer
which can be represented in two different ways as the sum of two cubes.}
The name of the locale category is determined through
@code{setlocale (LC_@var{category}, NULL)}.
@footnote{When the system does not support @code{setlocale} its behavior
in setting the locale values is simulated by looking at the environment
variables.}
When using the function @code{dcgettext}, you can specify the locale category
through the third argument.

@node Charset conversion, Contexts, Locating Catalogs, gettext
@subsection How to specify the output character set @code{gettext} uses
@cindex charset conversion at runtime
@cindex encoding conversion at runtime

@code{gettext} not only looks up a translation in a message catalog.  It
also converts the translation on the fly to the desired output character
set.  This is useful if the user is working in a different character set
than the translator who created the message catalog, because it avoids
distributing variants of message catalogs which differ only in the
character set.

The output character set is, by default, the value of @code{nl_langinfo
(CODESET)}, which depends on the @code{LC_CTYPE} part of the current
locale.  But programs which store strings in a locale independent way
(e.g.@: UTF-8) can request that @code{gettext} and related functions
return the translations in that encoding, by use of the
@code{bind_textdomain_codeset} function.

Note that the @var{msgid} argument to @code{gettext} is not subject to
character set conversion.  Also, when @code{gettext} does not find a
translation for @var{msgid}, it returns @var{msgid} unchanged --
independently of the current output character set.  It is therefore
recommended that all @var{msgid}s be US-ASCII strings.

@deftypefun {char *} bind_textdomain_codeset (const char *@var{domainname}, const char *@var{codeset})
The @code{bind_textdomain_codeset} function can be used to specify the
output character set for message catalogs for domain @var{domainname}.
The @var{codeset} argument must be a valid codeset name which can be used
for the @code{iconv_open} function, or a null pointer.

If the @var{codeset} parameter is the null pointer,
@code{bind_textdomain_codeset} returns the currently selected codeset
for the domain with the name @var{domainname}.  It returns @code{NULL} if
no codeset has yet been selected.

The @code{bind_textdomain_codeset} function can be used several times. 
If used multiple times with the same @var{domainname} argument, the
later call overrides the settings made by the earlier one.

The @code{bind_textdomain_codeset} function returns a pointer to a
string containing the name of the selected codeset.  The string is
allocated internally in the function and must not be changed by the
user.  If the system went out of core during the execution of
@code{bind_textdomain_codeset}, the return value is @code{NULL} and the
global variable @var{errno} is set accordingly.
@end deftypefun

@node Contexts, Plural forms, Charset conversion, gettext
@subsection Using contexts for solving ambiguities
@cindex context
@cindex GUI programs
@cindex translating menu entries
@cindex menu entries

One place where the @code{gettext} functions, if used normally, have big
problems is within programs with graphical user interfaces (GUIs).  The
problem is that many of the strings which have to be translated are very
short.  They have to appear in pull-down menus which restricts the
length.  But strings which are not containing entire sentences or at
least large fragments of a sentence may appear in more than one
situation in the program but might have different translations.  This is
especially true for the one-word strings which are frequently used in
GUI programs.

As a consequence many people say that the @code{gettext} approach is
wrong and instead @code{catgets} should be used which indeed does not
have this problem.  But there is a very simple and powerful method to
handle this kind of problems with the @code{gettext} functions.

Contexts can be added to strings to be translated.  A context dependent
translation lookup is when a translation for a given string is searched,
that is limited to a given context.  The translation for the same string
in a different context can be different.  The different translations of
the same string in different contexts can be stored in the in the same
MO file, and can be edited by the translator in the same PO file.

The @file{gettext.h} include file contains the lookup macros for strings
with contexts.  They are implemented as thin macros and inline functions
over the functions from @code{<libintl.h>}.

@findex pgettext
@example
const char *pgettext (const char *msgctxt, const char *msgid);
@end example

In a call of this macro, @var{msgctxt} and @var{msgid} must be string
literals.  The macro returns the translation of @var{msgid}, restricted
to the context given by @var{msgctxt}.

The @var{msgctxt} string is visible in the PO file to the translator.
You should try to make it somehow canonical and never changing.  Because
every time you change an @var{msgctxt}, the translator will have to review
the translation of @var{msgid}.

Finding a canonical @var{msgctxt} string that doesn't change over time can
be hard.  But you shouldn't use the file name or class name containing the
@code{pgettext} call -- because it is a common development task to rename
a file or a class, and it shouldn't cause translator work.  Also you shouldn't
use a comment in the form of a complete English sentence as @var{msgctxt} --
because orthography or grammar changes are often applied to such sentences,
and again, it shouldn't force the translator to do a review.

The @samp{p} in @samp{pgettext} stands for ``particular'': @code{pgettext}
fetches a particular translation of the @var{msgid}.

@findex dpgettext
@findex dcpgettext
@example
const char *dpgettext (const char *domain_name,
                       const char *msgctxt, const char *msgid);
const char *dcpgettext (const char *domain_name,
                        const char *msgctxt, const char *msgid,
                        int category);
@end example

These are generalizations of @code{pgettext}.  They behave similarly to
@code{dgettext} and @code{dcgettext}, respectively.  The @var{domain_name}
argument defines the translation domain.  The @var{category} argument
allows to use another locale category than @code{LC_MESSAGES}.

As as example consider the following fictional situation.  A GUI program
has a menu bar with the following entries:

@smallexample
+------------+------------+--------------------------------------+
| File       | Printer    |                                      |
+------------+------------+--------------------------------------+
| Open     | | Select   |
| New      | | Open     |
+----------+ | Connect  |
             +----------+
@end smallexample

To have the strings @code{File}, @code{Printer}, @code{Open},
@code{New}, @code{Select}, and @code{Connect} translated there has to be
at some point in the code a call to a function of the @code{gettext}
family.  But in two places the string passed into the function would be
@code{Open}.  The translations might not be the same and therefore we
are in the dilemma described above.

What distinguishes the two places is the menu path from the menu root to
the particular menu entries:

@smallexample
Menu|File
Menu|Printer
Menu|File|Open
Menu|File|New
Menu|Printer|Select
Menu|Printer|Open
Menu|Printer|Connect
@end smallexample

The context is thus the menu path without its last part.  So, the calls
look like this:

@smallexample
pgettext ("Menu|", "File")
pgettext ("Menu|", "Printer")
pgettext ("Menu|File|", "Open")
pgettext ("Menu|File|", "New")
pgettext ("Menu|Printer|", "Select")
pgettext ("Menu|Printer|", "Open")
pgettext ("Menu|Printer|", "Connect")
@end smallexample

Whether or not to use the @samp{|} character at the end of the context is a
matter of style.

For more complex cases, where the @var{msgctxt} or @var{msgid} are not
string literals, more general macros are available:

@findex pgettext_expr
@findex dpgettext_expr
@findex dcpgettext_expr
@example
const char *pgettext_expr (const char *msgctxt, const char *msgid);
const char *dpgettext_expr (const char *domain_name,
                            const char *msgctxt, const char *msgid);
const char *dcpgettext_expr (const char *domain_name,
                             const char *msgctxt, const char *msgid,
                             int category);
@end example

Here @var{msgctxt} and @var{msgid} can be arbitrary string-valued expressions.
These macros are more general.  But in the case that both argument expressions
are string literals, the macros without the @samp{_expr} suffix are more
efficient.

@node Plural forms, Optimized gettext, Contexts, gettext
@subsection Additional functions for plural forms
@cindex plural forms

The functions of the @code{gettext} family described so far (and all the
@code{catgets} functions as well) have one problem in the real world
which have been neglected completely in all existing approaches.  What
is meant here is the handling of plural forms.

Looking through Unix source code before the time anybody thought about
internationalization (and, sadly, even afterwards) one can often find
code similar to the following:

@smallexample
   printf ("%d file%s deleted", n, n == 1 ? "" : "s");
@end smallexample

@noindent
After the first complaints from people internationalizing the code people
either completely avoided formulations like this or used strings like
@code{"file(s)"}.  Both look unnatural and should be avoided.  First
tries to solve the problem correctly looked like this:

@smallexample
   if (n == 1)
     printf ("%d file deleted", n);
   else
     printf ("%d files deleted", n);
@end smallexample

But this does not solve the problem.  It helps languages where the
plural form of a noun is not simply constructed by adding an
@ifhtml
‘s’
@end ifhtml
@ifnothtml
`s'
@end ifnothtml
but that is all.  Once again people fell into the trap of believing the
rules their language is using are universal.  But the handling of plural
forms differs widely between the language families.  For example,
Rafal Maszkowski @code{<rzm@@mat.uni.torun.pl>} reports:

@quotation
In Polish we use e.g.@: plik (file) this way:
@example
1 plik
2,3,4 pliki
5-21 pliko'w
22-24 pliki
25-31 pliko'w
@end example
and so on (o' means 8859-2 oacute which should be rather okreska,
similar to aogonek).
@end quotation

There are two things which can differ between languages (and even inside
language families);

@itemize @bullet
@item
The form how plural forms are built differs.  This is a problem with
languages which have many irregularities.  German, for instance, is a
drastic case.  Though English and German are part of the same language
family (Germanic), the almost regular forming of plural noun forms
(appending an
@ifhtml
‘s’)
@end ifhtml
@ifnothtml
`s')
@end ifnothtml
is hardly found in German.

@item
The number of plural forms differ.  This is somewhat surprising for
those who only have experiences with Romanic and Germanic languages
since here the number is the same (there are two).

But other language families have only one form or many forms.  More
information on this in an extra section.
@end itemize

The consequence of this is that application writers should not try to
solve the problem in their code.  This would be localization since it is
only usable for certain, hardcoded language environments.  Instead the
extended @code{gettext} interface should be used.

These extra functions are taking instead of the one key string two
strings and a numerical argument.  The idea behind this is that using
the numerical argument and the first string as a key, the implementation
can select using rules specified by the translator the right plural
form.  The two string arguments then will be used to provide a return
value in case no message catalog is found (similar to the normal
@code{gettext} behavior).  In this case the rules for Germanic language
is used and it is assumed that the first string argument is the singular
form, the second the plural form.

This has the consequence that programs without language catalogs can
display the correct strings only if the program itself is written using
a Germanic language.  This is a limitation but since the GNU C library
(as well as the GNU @code{gettext} package) are written as part of the
GNU package and the coding standards for the GNU project require program
being written in English, this solution nevertheless fulfills its
purpose.

@deftypefun {char *} ngettext (const char *@var{msgid1}, const char *@var{msgid2}, unsigned long int @var{n})
The @code{ngettext} function is similar to the @code{gettext} function
as it finds the message catalogs in the same way.  But it takes two
extra arguments.  The @var{msgid1} parameter must contain the singular
form of the string to be converted.  It is also used as the key for the
search in the catalog.  The @var{msgid2} parameter is the plural form.
The parameter @var{n} is used to determine the plural form.  If no
message catalog is found @var{msgid1} is returned if @code{n == 1},
otherwise @code{msgid2}.

An example for the use of this function is:

@smallexample
printf (ngettext ("%d file removed", "%d files removed", n), n);
@end smallexample

Please note that the numeric value @var{n} has to be passed to the
@code{printf} function as well.  It is not sufficient to pass it only to
@code{ngettext}.

In the English singular case, the number -- always 1 -- can be replaced with
"one":

@smallexample
printf (ngettext ("One file removed", "%d files removed", n), n);
@end smallexample

@noindent
This works because the @samp{printf} function discards excess arguments that
are not consumed by the format string.

If this function is meant to yield a format string that takes two or more
arguments, you can not use it like this:

@smallexample
printf (ngettext ("%d file removed from directory %s",
                  "%d files removed from directory %s",
                  n),
        n, dir);
@end smallexample

@noindent
because in many languages the translators want to replace the @samp{%d}
with an explicit word in the singular case, just like ``one'' in English,
and C format strings cannot consume the second argument but skip the first
argument.  Instead, you have to reorder the arguments so that @samp{n}
comes last:

@smallexample
printf (ngettext ("%$2d file removed from directory %$1s",
                  "%$2d files removed from directory %$1s",
                  n),
        dir, n);
@end smallexample

@noindent
See @ref{c-format} for details about this argument reordering syntax.

When you know that the value of @code{n} is within a given range, you can
specify it as a comment directed to the @code{xgettext} tool.  This
information may help translators to use more adequate translations.  Like
this:

@smallexample
if (days > 7 && days < 14)
  /* xgettext: range: 1..6 */
  printf (ngettext ("one week and one day", "one week and %d days",
                    days - 7),
          days - 7);
@end smallexample

It is also possible to use this function when the strings don't contain a
cardinal number:

@smallexample
puts (ngettext ("Delete the selected file?",
                "Delete the selected files?",
                n));
@end smallexample

In this case the number @var{n} is only used to choose the plural form.
@end deftypefun

@deftypefun {char *} dngettext (const char *@var{domain}, const char *@var{msgid1}, const char *@var{msgid2}, unsigned long int @var{n})
The @code{dngettext} is similar to the @code{dgettext} function in the
way the message catalog is selected.  The difference is that it takes
two extra parameter to provide the correct plural form.  These two
parameters are handled in the same way @code{ngettext} handles them.
@end deftypefun

@deftypefun {char *} dcngettext (const char *@var{domain}, const char *@var{msgid1}, const char *@var{msgid2}, unsigned long int @var{n}, int @var{category})
The @code{dcngettext} is similar to the @code{dcgettext} function in the
way the message catalog is selected.  The difference is that it takes
two extra parameter to provide the correct plural form.  These two
parameters are handled in the same way @code{ngettext} handles them.
@end deftypefun

Now, how do these functions solve the problem of the plural forms?
Without the input of linguists (which was not available) it was not
possible to determine whether there are only a few different forms in
which plural forms are formed or whether the number can increase with
every new supported language.

Therefore the solution implemented is to allow the translator to specify
the rules of how to select the plural form.  Since the formula varies
with every language this is the only viable solution except for
hardcoding the information in the code (which still would require the
possibility of extensions to not prevent the use of new languages).

@cindex specifying plural form in a PO file
@kwindex nplurals@r{, in a PO file header}
@kwindex plural@r{, in a PO file header}
The information about the plural form selection has to be stored in the
header entry of the PO file (the one with the empty @code{msgid} string).
The plural form information looks like this:

@smallexample
Plural-Forms: nplurals=2; plural=n == 1 ? 0 : 1;
@end smallexample

The @code{nplurals} value must be a decimal number which specifies how
many different plural forms exist for this language.  The string
following @code{plural} is an expression which is using the C language
syntax.  Exceptions are that no negative numbers are allowed, numbers
must be decimal, and the only variable allowed is @code{n}.  Spaces are
allowed in the expression, but backslash-newlines are not; in the
examples below the backslash-newlines are present for formatting purposes
only.  This expression will be evaluated whenever one of the functions
@code{ngettext}, @code{dngettext}, or @code{dcngettext} is called.  The
numeric value passed to these functions is then substituted for all uses
of the variable @code{n} in the expression.  The resulting value then
must be greater or equal to zero and smaller than the value given as the
value of @code{nplurals}.

@noindent
@cindex plural form formulas
The following rules are known at this point.  The language with families
are listed.  But this does not necessarily mean the information can be
generalized for the whole family (as can be easily seen in the table
below).@footnote{Additions are welcome.  Send appropriate information to
@email{bug-gnu-gettext@@gnu.org} and @email{bug-glibc-manual@@gnu.org}.
The Unicode CLDR Project (@uref{http://cldr.unicode.org}) provides a
comprehensive set of plural forms in a different format.  The
@code{msginit} program has preliminary support for the format so you can
use it as a baseline (@pxref{msginit Invocation}).}

@table @asis
@item Only one form:
Some languages only require one single form.  There is no distinction
between the singular and plural form.  An appropriate header entry
would look like this:

@smallexample
Plural-Forms: nplurals=1; plural=0;
@end smallexample

@noindent
Languages with this property include:

@table @asis
@item Asian family
Japanese, @c   122.1 million speakers
Vietnamese, @c  68.6 million speakers
Korean @c       66.3 million speakers
@item Tai-Kadai family
Thai @c         20.4 million speakers
@end table

@item Two forms, singular used for one only
This is the form used in most existing programs since it is what English
is using.  A header entry would look like this:

@smallexample
Plural-Forms: nplurals=2; plural=n != 1;
@end smallexample

(Note: this uses the feature of C expressions that boolean expressions
have to value zero or one.)

@noindent
Languages with this property include:

@table @asis
@item Germanic family
English, @c    328.0 million speakers
German, @c      96.9 million speakers
Dutch, @c       21.7 million speakers
Swedish, @c      8.3 million speakers
Danish, @c       5.6 million speakers
Norwegian, @c    4.6 million speakers
Faroese @c       0.05 million speakers
@item Romanic family
Spanish, @c    328.5 million speakers
Portuguese, @c 178.0 million speakers - 163 million Brazilian Portuguese
Italian, @c     61.7 million speakers
Bulgarian @c     9.1 million speakers
@item Latin/Greek family
Greek @c        13.1 million speakers
@item Finno-Ugric family
Finnish, @c      5.0 million speakers
Estonian @c      1.0 million speakers
@item Semitic family
Hebrew @c        5.3 million speakers
@item Austronesian family
Bahasa Indonesian @c 23.2 million speakers
@item Artificial
Esperanto @c       2 million speakers
@end table

@noindent
Other languages using the same header entry are:

@table @asis
@item Finno-Ugric family
Hungarian @c   12.5 million speakers
@item Turkic/Altaic family
Turkish @c     50.8 million speakers
@end table

Hungarian does not appear to have a plural if you look at sentences involving
cardinal numbers.  For example, ``1 apple'' is ``1 alma'', and ``123 apples'' is
``123 alma''.  But when the number is not explicit, the distinction between
singular and plural exists: ``the apple'' is ``az alma'', and ``the apples'' is
``az alm@'{a}k''.  Since @code{ngettext} has to support both types of sentences,
it is classified here, under ``two forms''.

The same holds for Turkish: ``1 apple'' is ``1 elma'', and ``123 apples'' is
``123 elma''.  But when the number is omitted, the distinction between singular
and plural exists: ``the apple'' is ``elma'', and ``the apples'' is
``elmalar''.

@item Two forms, singular used for zero and one
Exceptional case in the language family.  The header entry would be:

@smallexample
Plural-Forms: nplurals=2; plural=n>1;
@end smallexample

@noindent
Languages with this property include:

@table @asis
@item Romanic family
Brazilian Portuguese, @c 163 million speakers
French @c                 67.8 million speakers
@end table

@item Three forms, special case for zero
The header entry would be:

@smallexample
Plural-Forms: nplurals=3; plural=n%10==1 && n%100!=11 ? 0 : n != 0 ? 1 : 2;
@end smallexample

@noindent
Languages with this property include:

@table @asis
@item Baltic family
Latvian @c     1.5 million speakers
@end table

@item Three forms, special cases for one and two
The header entry would be:

@smallexample
Plural-Forms: nplurals=3; plural=n==1 ? 0 : n==2 ? 1 : 2;
@end smallexample

@noindent
Languages with this property include:

@table @asis
@item Celtic
Gaeilge (Irish) @c 0.4 million speakers
@end table

@item Three forms, special case for numbers ending in 00 or [2-9][0-9]
The header entry would be:

@smallexample
Plural-Forms: nplurals=3; \
    plural=n==1 ? 0 : (n==0 || (n%100 > 0 && n%100 < 20)) ? 1 : 2;
@end smallexample

@noindent
Languages with this property include:

@table @asis
@item Romanic family
Romanian @c    23.4 million speakers
@end table

@item Three forms, special case for numbers ending in 1[2-9]
The header entry would look like this:

@smallexample
Plural-Forms: nplurals=3; \
    plural=n%10==1 && n%100!=11 ? 0 : \
           n%10>=2 && (n%100<10 || n%100>=20) ? 1 : 2;
@end smallexample

@noindent
Languages with this property include:

@table @asis
@item Baltic family
Lithuanian @c  3.2 million speakers
@end table

@item Three forms, special cases for numbers ending in 1 and 2, 3, 4, except those ending in 1[1-4]
The header entry would look like this:

@smallexample
Plural-Forms: nplurals=3; \
    plural=n%10==1 && n%100!=11 ? 0 : \
           n%10>=2 && n%10<=4 && (n%100<10 || n%100>=20) ? 1 : 2;
@end smallexample

@noindent
Languages with this property include:

@table @asis
@item Slavic family
Russian, @c    143.6 million speakers
Ukrainian, @c   37.0 million speakers
Belarusian, @c   8.6 million speakers
Serbian, @c      7.0 million speakers
Croatian @c      5.5 million speakers
@end table

@item Three forms, special cases for 1 and 2, 3, 4
The header entry would look like this:

@smallexample
Plural-Forms: nplurals=3; \
    plural=(n==1) ? 0 : (n>=2 && n<=4) ? 1 : 2;
@end smallexample

@noindent
Languages with this property include:

@table @asis
@item Slavic family
Czech, @c      9.5 million speakers
Slovak @c      5.0 million speakers
@end table

@item Three forms, special case for one and some numbers ending in 2, 3, or 4
The header entry would look like this:

@smallexample
Plural-Forms: nplurals=3; \
    plural=n==1 ? 0 : \
           n%10>=2 && n%10<=4 && (n%100<10 || n%100>=20) ? 1 : 2;
@end smallexample

@noindent
Languages with this property include:

@table @asis
@item Slavic family
Polish @c      40.0 million speakers
@end table

@item Four forms, special case for one and all numbers ending in 02, 03, or 04
The header entry would look like this:

@smallexample
Plural-Forms: nplurals=4; \
    plural=n%100==1 ? 0 : n%100==2 ? 1 : n%100==3 || n%100==4 ? 2 : 3;
@end smallexample

@noindent
Languages with this property include:

@table @asis
@item Slavic family
Slovenian @c   1.9 million speakers
@end table

@item Six forms, special cases for one, two, all numbers ending in 02, 03, @dots{} 10, all numbers ending in 11 @dots{} 99, and others
The header entry would look like this:

@smallexample
Plural-Forms: nplurals=6; \
    plural=n==0 ? 0 : n==1 ? 1 : n==2 ? 2 : n%100>=3 && n%100<=10 ? 3 \
    : n%100>=11 ? 4 : 5;
@end smallexample

@noindent
Languages with this property include:

@table @asis
@item Afroasiatic family
Arabic @c    246.0 million speakers
@end table
@end table

You might now ask, @code{ngettext} handles only numbers @var{n} of type
@samp{unsigned long}.  What about larger integer types?  What about negative
numbers?  What about floating-point numbers?

About larger integer types, such as @samp{uintmax_t} or 
@samp{unsigned long long}: they can be handled by reducing the value to a
range that fits in an @samp{unsigned long}.  Simply casting the value to
@samp{unsigned long} would not do the right thing, since it would treat
@code{ULONG_MAX + 1} like zero, @code{ULONG_MAX + 2} like singular, and
the like.  Here you can exploit the fact that all mentioned plural form
formulas eventually become periodic, with a period that is a divisor of 100
(or 1000 or 1000000).  So, when you reduce a large value to another one in
the range [1000000, 1999999] that ends in the same 6 decimal digits, you
can assume that it will lead to the same plural form selection.  This code
does this:

@smallexample
#include <inttypes.h>
uintmax_t nbytes = ...;
printf (ngettext ("The file has %"PRIuMAX" byte.",
                  "The file has %"PRIuMAX" bytes.",
                  (nbytes > ULONG_MAX
                   ? (nbytes % 1000000) + 1000000
                   : nbytes)),
        nbytes);
@end smallexample

Negative and floating-point values usually represent physical entities for
which singular and plural don't clearly apply.  In such cases, there is no
need to use @code{ngettext}; a simple @code{gettext} call with a form suitable
for all values will do.  For example:

@smallexample
printf (gettext ("Time elapsed: %.3f seconds"),
        num_milliseconds * 0.001);
@end smallexample

@noindent
Even if @var{num_milliseconds} happens to be a multiple of 1000, the output
@smallexample
Time elapsed: 1.000 seconds
@end smallexample
@noindent
is acceptable in English, and similarly for other languages.

The translators' perspective regarding plural forms is explained in
@ref{Translating plural forms}.

@node Optimized gettext,  , Plural forms, gettext
@subsection Optimization of the *gettext functions
@cindex optimization of @code{gettext} functions

At this point of the discussion we should talk about an advantage of the
GNU @code{gettext} implementation.  Some readers might have pointed out
that an internationalized program might have a poor performance if some
string has to be translated in an inner loop.  While this is unavoidable
when the string varies from one run of the loop to the other it is
simply a waste of time when the string is always the same.  Take the
following example:

@example
@group
@{
  while (@dots{})
    @{
      puts (gettext ("Hello world"));
    @}
@}
@end group
@end example

@noindent
When the locale selection does not change between two runs the resulting
string is always the same.  One way to use this is:

@example
@group
@{
  str = gettext ("Hello world");
  while (@dots{})
    @{
      puts (str);
    @}
@}
@end group
@end example

@noindent
But this solution is not usable in all situation (e.g.@: when the locale
selection changes) nor does it lead to legible code.

For this reason, GNU @code{gettext} caches previous translation results.
When the same translation is requested twice, with no new message
catalogs being loaded in between, @code{gettext} will, the second time,
find the result through a single cache lookup.

@node Comparison, Using libintl.a, gettext, Programmers
@section Comparing the Two Interfaces
@cindex @code{gettext} vs @code{catgets}
@cindex comparison of interfaces

@c FIXME: arguments to catgets vs. gettext
@c Partly done 950718 -- drepper

The following discussion is perhaps a little bit colored.  As said
above we implemented GNU @code{gettext} following the Uniforum
proposal and this surely has its reasons.  But it should show how we
came to this decision.

First we take a look at the developing process.  When we write an
application using NLS provided by @code{gettext} we proceed as always.
Only when we come to a string which might be seen by the users and thus
has to be translated we use @code{gettext("@dots{}")} instead of
@code{"@dots{}"}.  At the beginning of each source file (or in a central
header file) we define

@example
#define gettext(String) (String)
@end example

Even this definition can be avoided when the system supports the
@code{gettext} function in its C library.  When we compile this code the
result is the same as if no NLS code is used.  When  you take a look at
the GNU @code{gettext} code you will see that we use @code{_("@dots{}")}
instead of @code{gettext("@dots{}")}.  This reduces the number of
additional characters per translatable string to @emph{3} (in words:
three).

When now a production version of the program is needed we simply replace
the definition

@example
#define _(String) (String)
@end example

@noindent
by

@cindex include file @file{libintl.h}
@example
#include <libintl.h>
#define _(String) gettext (String)
@end example

@noindent
Additionally we run the program @file{xgettext} on all source code file
which contain translatable strings and that's it: we have a running
program which does not depend on translations to be available, but which
can use any that becomes available.

@cindex @code{N_}, a convenience macro
The same procedure can be done for the @code{gettext_noop} invocations
(@pxref{Special cases}).  One usually defines @code{gettext_noop} as a
no-op macro.  So you should consider the following code for your project:

@example
#define gettext_noop(String) String
#define N_(String) gettext_noop (String)
@end example

@code{N_} is a short form similar to @code{_}.  The @file{Makefile} in
the @file{po/} directory of GNU @code{gettext} knows by default both of the
mentioned short forms so you are invited to follow this proposal for
your own ease.

Now to @code{catgets}.  The main problem is the work for the
programmer.  Every time he comes to a translatable string he has to
define a number (or a symbolic constant) which has also be defined in
the message catalog file.  He also has to take care for duplicate
entries, duplicate message IDs etc.  If he wants to have the same
quality in the message catalog as the GNU @code{gettext} program
provides he also has to put the descriptive comments for the strings and
the location in all source code files in the message catalog.  This is
nearly a Mission: Impossible.

But there are also some points people might call advantages speaking for
@code{catgets}.  If you have a single word in a string and this string
is used in different contexts it is likely that in one or the other
language the word has different translations.  Example:

@example
printf ("%s: %d", gettext ("number"), number_of_errors)

printf ("you should see %d %s", number_count,
        number_count == 1 ? gettext ("number") : gettext ("numbers"))
@end example

Here we have to translate two times the string @code{"number"}.  Even
if you do not speak a language beside English it might be possible to
recognize that the two words have a different meaning.  In German the
first appearance has to be translated to @code{"Anzahl"} and the second
to @code{"Zahl"}.

Now you can say that this example is really esoteric.  And you are
right!  This is exactly how we felt about this problem and decide that
it does not weight that much.  The solution for the above problem could
be very easy:

@example
printf ("%s %d", gettext ("number:"), number_of_errors)

printf (number_count == 1 ? gettext ("you should see %d number")
                          : gettext ("you should see %d numbers"),
        number_count)
@end example

We believe that we can solve all conflicts with this method.  If it is
difficult one can also consider changing one of the conflicting string a
little bit.  But it is not impossible to overcome.

@code{catgets} allows same original entry to have different translations,
but @code{gettext} has another, scalable approach for solving ambiguities
of this kind: @xref{Ambiguities}.

@node Using libintl.a, gettext grok, Comparison, Programmers
@section Using libintl.a in own programs

Starting with version 0.9.4 the library @code{libintl.h} should be
self-contained.  I.e., you can use it in your own programs without
providing additional functions.  The @file{Makefile} will put the header
and the library in directories selected using the @code{$(prefix)}.

@node gettext grok, Temp Programmers, Using libintl.a, Programmers
@section Being a @code{gettext} grok

@strong{ NOTE: } This documentation section is outdated and needs to be
revised.

To fully exploit the functionality of the GNU @code{gettext} library it
is surely helpful to read the source code.  But for those who don't want
to spend that much time in reading the (sometimes complicated) code here
is a list comments:

@itemize @bullet
@item Changing the language at runtime
@cindex language selection at runtime

For interactive programs it might be useful to offer a selection of the
used language at runtime.  To understand how to do this one need to know
how the used language is determined while executing the @code{gettext}
function.  The method which is presented here only works correctly
with the GNU implementation of the @code{gettext} functions.

In the function @code{dcgettext} at every call the current setting of
the highest priority environment variable is determined and used.
Highest priority means here the following list with decreasing
priority:

@enumerate
@vindex LANGUAGE@r{, environment variable}
@item @code{LANGUAGE}
@vindex LC_ALL@r{, environment variable}
@item @code{LC_ALL}
@vindex LC_CTYPE@r{, environment variable}
@vindex LC_NUMERIC@r{, environment variable}
@vindex LC_TIME@r{, environment variable}
@vindex LC_COLLATE@r{, environment variable}
@vindex LC_MONETARY@r{, environment variable}
@vindex LC_MESSAGES@r{, environment variable}
@item @code{LC_xxx}, according to selected locale category
@vindex LANG@r{, environment variable}
@item @code{LANG}
@end enumerate

Afterwards the path is constructed using the found value and the
translation file is loaded if available.

What happens now when the value for, say, @code{LANGUAGE} changes?  According
to the process explained above the new value of this variable is found
as soon as the @code{dcgettext} function is called.  But this also means
the (perhaps) different message catalog file is loaded.  In other
words: the used language is changed.

But there is one little hook.  The code for gcc-2.7.0 and up provides
some optimization.  This optimization normally prevents the calling of
the @code{dcgettext} function as long as no new catalog is loaded.  But
if @code{dcgettext} is not called the program also cannot find the
@code{LANGUAGE} variable be changed (@pxref{Optimized gettext}).  A
solution for this is very easy.  Include the following code in the
language switching function.

@example
  /* Change language.  */
  setenv ("LANGUAGE", "fr", 1);

  /* Make change known.  */
  @{
    extern int  _nl_msg_cat_cntr;
    ++_nl_msg_cat_cntr;
  @}
@end example

@cindex @code{_nl_msg_cat_cntr}
The variable @code{_nl_msg_cat_cntr} is defined in @file{loadmsgcat.c}.
You don't need to know what this is for.  But it can be used to detect
whether a @code{gettext} implementation is GNU gettext and not non-GNU
system's native gettext implementation.

@end itemize

@node Temp Programmers,  , gettext grok, Programmers
@section Temporary Notes for the Programmers Chapter

@strong{ NOTE: } This documentation section is outdated and needs to be
revised.

@menu
* Temp Implementations::        Temporary - Two Possible Implementations
* Temp catgets::                Temporary - About @code{catgets}
* Temp WSI::                    Temporary - Why a single implementation
* Temp Notes::                  Temporary - Notes
@end menu

@node Temp Implementations, Temp catgets, Temp Programmers, Temp Programmers
@subsection Temporary - Two Possible Implementations

There are two competing methods for language independent messages:
the X/Open @code{catgets} method, and the Uniforum @code{gettext}
method.  The @code{catgets} method indexes messages by integers; the
@code{gettext} method indexes them by their English translations.
The @code{catgets} method has been around longer and is supported
by more vendors.  The @code{gettext} method is supported by Sun,
and it has been heard that the COSE multi-vendor initiative is
supporting it.  Neither method is a POSIX standard; the POSIX.1
committee had a lot of disagreement in this area.

Neither one is in the POSIX standard.  There was much disagreement
in the POSIX.1 committee about using the @code{gettext} routines
vs. @code{catgets} (XPG).  In the end the committee couldn't
agree on anything, so no messaging system was included as part
of the standard.  I believe the informative annex of the standard
includes the XPG3 messaging interfaces, ``@dots{}as an example of
a messaging system that has been implemented@dots{}''

They were very careful not to say anywhere that you should use one
set of interfaces over the other.  For more on this topic please
see the Programming for Internationalization FAQ.

@node Temp catgets, Temp WSI, Temp Implementations, Temp Programmers
@subsection Temporary - About @code{catgets}

There have been a few discussions of late on the use of
@code{catgets} as a base.  I think it important to present both
sides of the argument and hence am opting to play devil's advocate
for a little bit.

I'll not deny the fact that @code{catgets} could have been designed
a lot better.  It currently has quite a number of limitations and
these have already been pointed out.

However there is a great deal to be said for consistency and
standardization.  A common recurring problem when writing Unix
software is the myriad portability problems across Unix platforms.
It seems as if every Unix vendor had a look at the operating system
and found parts they could improve upon.  Undoubtedly, these
modifications are probably innovative and solve real problems.
However, software developers have a hard time keeping up with all
these changes across so many platforms.

And this has prompted the Unix vendors to begin to standardize their
systems.  Hence the impetus for Spec1170.  Every major Unix vendor
has committed to supporting this standard and every Unix software
developer waits with glee the day they can write software to this
standard and simply recompile (without having to use autoconf)
across different platforms.

As I understand it, Spec1170 is roughly based upon version 4 of the
X/Open Portability Guidelines (XPG4).  Because @code{catgets} and
friends are defined in XPG4, I'm led to believe that @code{catgets}
is a part of Spec1170 and hence will become a standardized component
of all Unix systems.

@node Temp WSI, Temp Notes, Temp catgets, Temp Programmers
@subsection Temporary - Why a single implementation

Now it seems kind of wasteful to me to have two different systems
installed for accessing message catalogs.  If we do want to remedy
@code{catgets} deficiencies why don't we try to expand @code{catgets}
(in a compatible manner) rather than implement an entirely new system.
Otherwise, we'll end up with two message catalog access systems installed
with an operating system - one set of routines for packages using GNU
@code{gettext} for their internationalization, and another set of routines
(catgets) for all other software.  Bloated?

Supposing another catalog access system is implemented.  Which do
we recommend?  At least for Linux, we need to attract as many
software developers as possible.  Hence we need to make it as easy
for them to port their software as possible.  Which means supporting
@code{catgets}.  We will be implementing the @code{libintl} code
within our @code{libc}, but does this mean we also have to incorporate
another message catalog access scheme within our @code{libc} as well?
And what about people who are going to be using the @code{libintl}
+ non-@code{catgets} routines.  When they port their software to
other platforms, they're now going to have to include the front-end
(@code{libintl}) code plus the back-end code (the non-@code{catgets}
access routines) with their software instead of just including the
@code{libintl} code with their software.

Message catalog support is however only the tip of the iceberg.
What about the data for the other locale categories?  They also have
a number of deficiencies.  Are we going to abandon them as well and
develop another duplicate set of routines (should @code{libintl}
expand beyond message catalog support)?

Like many parts of Unix that can be improved upon, we're stuck with balancing
compatibility with the past with useful improvements and innovations for
the future.

@node Temp Notes,  , Temp WSI, Temp Programmers
@subsection Temporary - Notes

X/Open agreed very late on the standard form so that many
implementations differ from the final form.  Both of my system (old
Linux catgets and Ultrix-4) have a strange variation.

OK.  After incorporating the last changes I have to spend some time on
making the GNU/Linux @code{libc} @code{gettext} functions.  So in future
Solaris is not the only system having @code{gettext}.

@node Translators, Maintainers, Programmers, Top
@chapter The Translator's View

@c FIXME: Reorganize whole chapter.

@menu
* Trans Intro 0::               Introduction 0
* Trans Intro 1::               Introduction 1
* Discussions::                 Discussions
* Organization::                Organization
* Information Flow::            Information Flow
* Translating plural forms::    How to fill in @code{msgstr[0]}, @code{msgstr[1]}
* Prioritizing messages::       How to find which messages to translate first
@end menu

@node Trans Intro 0, Trans Intro 1, Translators, Translators
@section Introduction 0

@strong{ NOTE: } This documentation section is outdated and needs to be
revised.

Free software is going international!  The Translation Project is a way
to get maintainers, translators and users all together, so free software
will gradually become able to speak many native languages.

The GNU @code{gettext} tool set contains @emph{everything} maintainers
need for internationalizing their packages for messages.  It also
contains quite useful tools for helping translators at localizing
messages to their native language, once a package has already been
internationalized.

To achieve the Translation Project, we need many interested
people who like their own language and write it well, and who are also
able to synergize with other translators speaking the same language.
If you'd like to volunteer to @emph{work} at translating messages,
please send mail to your translating team.

Each team has its own mailing list, courtesy of Linux
International.  You may reach your translating team at the address
@file{@var{ll}@@li.org}, replacing @var{ll} by the two-letter @w{ISO 639}
code for your language.  Language codes are @emph{not} the same as
country codes given in @w{ISO 3166}.  The following translating teams
exist:

@quotation
Chinese @code{zh}, Czech @code{cs}, Danish @code{da}, Dutch @code{nl},
Esperanto @code{eo}, Finnish @code{fi}, French @code{fr}, Irish
@code{ga}, German @code{de}, Greek @code{el}, Italian @code{it},
Japanese @code{ja}, Indonesian @code{in}, Norwegian @code{no}, Polish
@code{pl}, Portuguese @code{pt}, Russian @code{ru}, Spanish @code{es},
Swedish @code{sv} and Turkish @code{tr}.
@end quotation

@noindent
For example, you may reach the Chinese translating team by writing to
@file{zh@@li.org}.  When you become a member of the translating team
for your own language, you may subscribe to its list.  For example,
Swedish people can send a message to @w{@file{sv-request@@li.org}},
having this message body:

@example
subscribe
@end example

Keep in mind that team members should be interested in @emph{working}
at translations, or at solving translational difficulties, rather than
merely lurking around.  If your team does not exist yet and you want to
start one, please write to @w{@file{coordinator@@translationproject.org}};
you will then reach the coordinator for all translator teams.

A handful of GNU packages have already been adapted and provided
with message translations for several languages.  Translation
teams have begun to organize, using these packages as a starting
point.  But there are many more packages and many languages for
which we have no volunteer translators.  If you would like to
volunteer to work at translating messages, please send mail to
@file{coordinator@@translationproject.org} indicating what language(s)
you can work on.

@node Trans Intro 1, Discussions, Trans Intro 0, Translators
@section Introduction 1

@strong{ NOTE: } This documentation section is outdated and needs to be
revised.

This is now official, GNU is going international!  Here is the
announcement submitted for the January 1995 GNU Bulletin:

@quotation
A handful of GNU packages have already been adapted and provided
with message translations for several languages.  Translation
teams have begun to organize, using these packages as a starting
point.  But there are many more packages and many languages
for which we have no volunteer translators.  If you'd like to
volunteer to work at translating messages, please send mail to
@samp{coordinator@@translationproject.org} indicating what language(s)
you can work on.
@end quotation

This document should answer many questions for those who are curious about
the process or would like to contribute.  Please at least skim over it,
hoping to cut down a little of the high volume of e-mail generated by this
collective effort towards internationalization of free software.

Most free programming which is widely shared is done in English, and
currently, English is used as the main communicating language between
national communities collaborating to free software.  This very document
is written in English.  This will not change in the foreseeable future.

However, there is a strong appetite from national communities for
having more software able to write using national language and habits,
and there is an on-going effort to modify free software in such a way
that it becomes able to do so.  The experiments driven so far raised
an enthusiastic response from pretesters, so we believe that
internationalization of free software is dedicated to succeed.

For suggestion clarifications, additions or corrections to this
document, please e-mail to @file{coordinator@@translationproject.org}.

@node Discussions, Organization, Trans Intro 1, Translators
@section Discussions

@strong{ NOTE: } This documentation section is outdated and needs to be
revised.

Facing this internationalization effort, a few users expressed their
concerns.  Some of these doubts are presented and discussed, here.

@itemize @bullet
@item Smaller groups

Some languages are not spoken by a very large number of people, so people
speaking them sometimes consider that there may not be all that much
demand such versions of free software packages.  Moreover, many people
being @emph{into computers}, in some countries, generally seem to prefer
English versions of their software.

On the other end, people might enjoy their own language a lot, and be
very motivated at providing to themselves the pleasure of having their
beloved free software speaking their mother tongue.  They do themselves
a personal favor, and do not pay that much attention to the number of
people benefiting of their work.

@item Misinterpretation

Other users are shy to push forward their own language, seeing in this
some kind of misplaced propaganda.  Someone thought there must be some
users of the language over the networks pestering other people with it.

But any spoken language is worth localization, because there are
people behind the language for whom the language is important and
dear to their hearts.

@item Odd translations

The biggest problem is to find the right translations so that
everybody can understand the messages.  Translations are usually a
little odd.  Some people get used to English, to the extent they may
find translations into their own language ``rather pushy, obnoxious
and sometimes even hilarious.''  As a French speaking man, I have
the experience of those instruction manuals for goods, so poorly
translated in French in Korea or Taiwan@dots{}

The fact is that we sometimes have to create a kind of national
computer culture, and this is not easy without the collaboration of
many people liking their mother tongue.  This is why translations are
better achieved by people knowing and loving their own language, and
ready to work together at improving the results they obtain.

@item Dependencies over the GPL or LGPL

Some people wonder if using GNU @code{gettext} necessarily brings their
package under the protective wing of the GNU General Public License or
the GNU Lesser General Public License, when they do not want to make
their program free, or want other kinds of freedom.  The simplest
answer is ``normally not''.

The @code{gettext-runtime} part of GNU @code{gettext}, i.e.@: the
contents of @code{libintl}, is covered by the GNU Lesser General Public
License.  The @code{gettext-tools} part of GNU @code{gettext}, i.e.@: the
rest of the GNU @code{gettext} package, is covered by the GNU General
Public License.

The mere marking of localizable strings in a package, or conditional
inclusion of a few lines for initialization, is not really including
GPL'ed or LGPL'ed code.  However, since the localization routines in
@code{libintl} are under the LGPL, the LGPL needs to be considered.
It gives the right to distribute the complete unmodified source of
@code{libintl} even with non-free programs.  It also gives the right
to use @code{libintl} as a shared library, even for non-free programs.
But it gives the right to use @code{libintl} as a static library or
to incorporate @code{libintl} into another library only to free
software.

@end itemize

@node Organization, Information Flow, Discussions, Translators
@section Organization

@strong{ NOTE: } This documentation section is outdated and needs to be
revised.

On a larger scale, the true solution would be to organize some kind of
fairly precise set up in which volunteers could participate.  I gave
some thought to this idea lately, and realize there will be some
touchy points.  I thought of writing to Richard Stallman to launch
such a project, but feel it might be good to shake out the ideas
between ourselves first.  Most probably that Linux International has
some experience in the field already, or would like to orchestrate
the volunteer work, maybe.  Food for thought, in any case!

I guess we have to setup something early, somehow, that will help
many possible contributors of the same language to interlock and avoid
work duplication, and further be put in contact for solving together
problems particular to their tongue (in most languages, there are many
difficulties peculiar to translating technical English).  My Swedish
contributor acknowledged these difficulties, and I'm well aware of
them for French.

This is surely not a technical issue, but we should manage so the
effort of locale contributors be maximally useful, despite the national
team layer interface between contributors and maintainers.

The Translation Project needs some setup for coordinating language
coordinators.  Localizing evolving programs will surely
become a permanent and continuous activity in the free software community,
once well started.
The setup should be minimally completed and tested before GNU
@code{gettext} becomes an official reality.  The e-mail address
@file{coordinator@@translationproject.org} has been set up for receiving
offers from volunteers and general e-mail on these topics.  This address
reaches the Translation Project coordinator.

@menu
* Central Coordination::        Central Coordination
* National Teams::              National Teams
* Mailing Lists::               Mailing Lists
@end menu

@node Central Coordination, National Teams, Organization, Organization
@subsection Central Coordination

I also think GNU will need sooner than it thinks, that someone set up
a way to organize and coordinate these groups.  Some kind of group
of groups.  My opinion is that it would be good that GNU delegates
this task to a small group of collaborating volunteers, shortly.
Perhaps in @file{gnu.announce} a list of this national committee's
can be published.

My role as coordinator would simply be to refer to Ulrich any German
speaking volunteer interested to localization of free software packages, and
maybe helping national groups to initially organize, while maintaining
national registries for until national groups are ready to take over.
In fact, the coordinator should ease volunteers to get in contact with
one another for creating national teams, which should then select
one coordinator per language, or country (regionalized language).
If well done, the coordination should be useful without being an
overwhelming task, the time to put delegations in place.

@node National Teams, Mailing Lists, Central Coordination, Organization
@subsection National Teams

I suggest we look for volunteer coordinators/editors for individual
languages.  These people will scan contributions of translation files
for various programs, for their own languages, and will ensure high
and uniform standards of diction.

From my current experience with other people in these days, those who
provide localizations are very enthusiastic about the process, and are
more interested in the localization process than in the program they
localize, and want to do many programs, not just one.  This seems
to confirm that having a coordinator/editor for each language is a
good idea.

We need to choose someone who is good at writing clear and concise
prose in the language in question.  That is hard---we can't check
it ourselves.  So we need to ask a few people to judge each others'
writing and select the one who is best.

I announce my prerelease to a few dozen people, and you would not
believe all the discussions it generated already.  I shudder to think
what will happen when this will be launched, for true, officially,
world wide.  Who am I to arbitrate between two Czekolsovak users
contradicting each other, for example?

I assume that your German is not much better than my French so that
I would not be able to judge about these formulations.  What I would
suggest is that for each language there is a group for people who
maintain the PO files and judge about changes.  I suspect there will
be cultural differences between how such groups of people will behave.
Some will have relaxed ways, reach consensus easily, and have anyone
of the group relate to the maintainers, while others will fight to
death, organize heavy administrations up to national standards, and
use strict channels.

The German team is putting out a good example.  Right now, they are
maybe half a dozen people revising translations of each other and
discussing the linguistic issues.  I do not even have all the names.
Ulrich Drepper is taking care of coordinating the German team.
He subscribed to all my pretest lists, so I do not even have to warn
him specifically of incoming releases.

I'm sure, that is a good idea to get teams for each language working
on translations.  That will make the translations better and more
consistent.

@menu
* Sub-Cultures::                Sub-Cultures
* Organizational Ideas::        Organizational Ideas
@end menu

@node Sub-Cultures, Organizational Ideas, National Teams, National Teams
@subsubsection Sub-Cultures

Taking French for example, there are a few sub-cultures around computers
which developed diverging vocabularies.  Picking volunteers here and
there without addressing this problem in an organized way, soon in the
project, might produce a distasteful mix of internationalized programs,
and possibly trigger endless quarrels among those who really care.

Keeping some kind of unity in the way French localization of
internationalized programs is achieved is a difficult (and delicate) job.
Knowing the latin character of French people (:-), if we take this
the wrong way, we could end up nowhere, or spoil a lot of energies.
Maybe we should begin to address this problem seriously @emph{before}
GNU @code{gettext} become officially published.  And I suspect that this
means soon!

@node Organizational Ideas,  , Sub-Cultures, National Teams
@subsubsection Organizational Ideas

I expect the next big changes after the official release.  Please note
that I use the German translation of the short GPL message.  We need
to set a few good examples before the localization goes out for true
in the free software community.  Here are a few points to discuss:

@itemize @bullet
@item
Each group should have one FTP server (at least one master).

@item
The files on the server should reflect the latest version (of
course!) and it should also contain a RCS directory with the
corresponding archives (I don't have this now).

@item
There should also be a ChangeLog file (this is more useful than the
RCS archive but can be generated automatically from the later by
Emacs).

@item
A @dfn{core group} should judge about questionable changes (for now
this group consists solely by me but I ask some others occasionally;
this also seems to work).

@end itemize

@node Mailing Lists,  , National Teams, Organization
@subsection Mailing Lists

If we get any inquiries about GNU @code{gettext}, send them on to:

@example
@file{coordinator@@translationproject.org}
@end example

The @file{*-pretest} lists are quite useful to me, maybe the idea could
be generalized to many GNU, and non-GNU packages.  But each maintainer
his/her way!

Fran@,{c}ois, we have a mechanism in place here at
@file{gnu.ai.mit.edu} to track teams, support mailing lists for
them and log members.  We have a slight preference that you use it.
If this is OK with you, I can get you clued in.

Things are changing!  A few years ago, when Daniel Fekete and I
asked for a mailing list for GNU localization, nested at the FSF, we
were politely invited to organize it anywhere else, and so did we.
For communicating with my pretesters, I later made a handful of
mailing lists located at iro.umontreal.ca and administrated by
@code{majordomo}.  These lists have been @emph{very} dependable
so far@dots{}

I suspect that the German team will organize itself a mailing list
located in Germany, and so forth for other countries.  But before they
organize for true, it could surely be useful to offer mailing lists
located at the FSF to each national team.  So yes, please explain me
how I should proceed to create and handle them.

We should create temporary mailing lists, one per country, to help
people organize.  Temporary, because once regrouped and structured, it
would be fair the volunteers from country bring back @emph{their} list
in there and manage it as they want.  My feeling is that, in the long
run, each team should run its own list, from within their country.
There also should be some central list to which all teams could
subscribe as they see fit, as long as each team is represented in it.

@node Information Flow, Translating plural forms, Organization, Translators
@section Information Flow

@strong{ NOTE: } This documentation section is outdated and needs to be
revised.

There will surely be some discussion about this messages after the
packages are finally released.  If people now send you some proposals
for better messages, how do you proceed?  Jim, please note that
right now, as I put forward nearly a dozen of localizable programs, I
receive both the translations and the coordination concerns about them.

If I put one of my things to pretest, Ulrich receives the announcement
and passes it on to the German team, who make last minute revisions.
Then he submits the translation files to me @emph{as the maintainer}.
For free packages I do not maintain, I would not even hear about it.
This scheme could be made to work for the whole Translation Project,
I think.  For security reasons, maybe Ulrich (national coordinators,
in fact) should update central registry kept at the Translation Project
(Jim, me, or Len's recruits) once in a while.

In December/January, I was aggressively ready to internationalize
all of GNU, giving myself the duty of one small GNU package per week
or so, taking many weeks or months for bigger packages.  But it does
not work this way.  I first did all the things I'm responsible for.
I've nothing against some missionary work on other maintainers, but
I'm also losing a lot of energy over it---same debates over again.

And when the first localized packages are released we'll get a lot of
responses about ugly translations :-).  Surely, and we need to have
beforehand a fairly good idea about how to handle the information
flow between the national teams and the package maintainers.

Please start saving somewhere a quick history of each PO file.  I know
for sure that the file format will change, allowing for comments.
It would be nice that each file has a kind of log, and references for
those who want to submit comments or gripes, or otherwise contribute.
I sent a proposal for a fast and flexible format, but it is not
receiving acceptance yet by the GNU deciders.  I'll tell you when I
have more information about this.

@node Translating plural forms, Prioritizing messages, Information Flow, Translators
@section Translating plural forms

@cindex plural forms, translating
Suppose you are translating a PO file, and it contains an entry like this:

@smallexample
#, c-format
msgid "One file removed"
msgid_plural "%d files removed"
msgstr[0] ""
msgstr[1] ""
@end smallexample

@noindent
What does this mean? How do you fill it in?

Such an entry denotes a message with plural forms, that is, a message where
the text depends on a cardinal number.  The general form of the message,
in English, is the @code{msgid_plural} line.  The @code{msgid} line is the
English singular form, that is, the form for when the number is equal to 1.
More details about plural forms are explained in @ref{Plural forms}.

The first thing you need to look at is the @code{Plural-Forms} line in the
header entry of the PO file.  It contains the number of plural forms and a
formula.  If the PO file does not yet have such a line, you have to add it.
It only depends on the language into which you are translating.  You can
get this info by using the @code{msginit} command (see @ref{Creating}) --
it contains a database of known plural formulas -- or by asking other
members of your translation team.

Suppose the line looks as follows:

@smallexample
"Plural-Forms: nplurals=3; plural=n%10==1 && n%100!=11 ? 0 : n%10>=2 && n"
"%10<=4 && (n%100<10 || n%100>=20) ? 1 : 2;\n"
@end smallexample

It's logically one line; recall that the PO file formatting is allowed to
break long lines so that each physical line fits in 80 monospaced columns.

The value of @code{nplurals} here tells you that there are three plural
forms.  The first thing you need to do is to ensure that the entry contains
an @code{msgstr} line for each of the forms:

@smallexample
#, c-format
msgid "One file removed"
msgid_plural "%d files removed"
msgstr[0] ""
msgstr[1] ""
msgstr[2] ""
@end smallexample

Then translate the @code{msgid_plural} line and fill it in into each
@code{msgstr} line:

@smallexample
#, c-format
msgid "One file removed"
msgid_plural "%d files removed"
msgstr[0] "%d slika uklonjenih"
msgstr[1] "%d slika uklonjenih"
msgstr[2] "%d slika uklonjenih"
@end smallexample

Now you can refine the translation so that it matches the plural form.
According to the formula above, @code{msgstr[0]} is used when the number
ends in 1 but does not end in 11; @code{msgstr[1]} is used when the number
ends in 2, 3, 4, but not in 12, 13, 14; and @code{msgstr[2]} is used in
all other cases.  With this knowledge, you can refine the translations:

@smallexample
#, c-format
msgid "One file removed"
msgid_plural "%d files removed"
msgstr[0] "%d slika je uklonjena"
msgstr[1] "%d datoteke uklonjenih"
msgstr[2] "%d slika uklonjenih"
@end smallexample

You noticed that in the English singular form (@code{msgid}) the number
placeholder could be omitted and replaced by the numeral word ``one''.
Can you do this in your translation as well?

@smallexample
msgstr[0] "jednom datotekom je uklonjen"
@end smallexample

@noindent
Well, it depends on whether @code{msgstr[0]} applies only to the number 1,
or to other numbers as well.  If, according to the plural formula,
@code{msgstr[0]} applies only to @code{n == 1}, then you can use the
specialized translation without the number placeholder.  In our case,
however, @code{msgstr[0]} also applies to the numbers 21, 31, 41, etc.,
and therefore you cannot omit the placeholder.

@node Prioritizing messages,  , Translating plural forms, Translators
@section Prioritizing messages: How to determine which messages to translate first

A translator sometimes has only a limited amount of time per week to
spend on a package, and some packages have quite large message catalogs
(over 1000 messages).  Therefore she wishes to translate the messages
first that are the most visible to the user, or that occur most frequently.
This section describes how to determine these "most urgent" messages.
It also applies to determine the "next most urgent" messages after the
message catalog has already been partially translated.

In a first step, she uses the programs like a user would do.  While she
does this, the GNU @code{gettext} library logs into a file the not yet
translated messages for which a translation was requested from the program.

In a second step, she uses the PO mode to translate precisely this set
of messages.

@vindex GETTEXT_LOG_UNTRANSLATED@r{, environment variable}
Here a more details.  The GNU @code{libintl} library (but not the
corresponding functions in GNU @code{libc}) supports an environment variable
@code{GETTEXT_LOG_UNTRANSLATED}.  The GNU @code{libintl} library will
log into this file the messages for which @code{gettext()} and related
functions couldn't find the translation.  If the file doesn't exist, it
will be created as needed.  On systems with GNU @code{libc} a shared library
@samp{preloadable_libintl.so} is provided that can be used with the ELF
@samp{LD_PRELOAD} mechanism.

So, in the first step, the translator uses these commands on systems with
GNU @code{libc}:

@smallexample
$ LD_PRELOAD=/usr/local/lib/preloadable_libintl.so
$ export LD_PRELOAD
$ GETTEXT_LOG_UNTRANSLATED=$HOME/gettextlogused
$ export GETTEXT_LOG_UNTRANSLATED
@end smallexample

@noindent
and these commands on other systems:

@smallexample
$ GETTEXT_LOG_UNTRANSLATED=$HOME/gettextlogused
$ export GETTEXT_LOG_UNTRANSLATED
@end smallexample

Then she uses and peruses the programs.  (It is a good and recommended
practice to use the programs for which you provide translations: it
gives you the needed context.)  When done, she removes the environment
variables:

@smallexample
$ unset LD_PRELOAD
$ unset GETTEXT_LOG_UNTRANSLATED
@end smallexample

The second step starts with removing duplicates:

@smallexample
$ msguniq $HOME/gettextlogused > missing.po
@end smallexample

The result is a PO file, but needs some preprocessing before a PO file editor
can be used with it.  First, it is a multi-domain PO file, containing
messages from many translation domains.  Second, it lacks all translator
comments and source references.  Here is how to get a list of the affected
translation domains:

@smallexample
$ sed -n -e 's,^domain "\(.*\)"$,\1,p' < missing.po | sort | uniq
@end smallexample

Then the translator can handle the domains one by one.  For simplicity,
let's use environment variables to denote the language, domain and source
package.

@smallexample
$ lang=nl             # your language
$ domain=coreutils    # the name of the domain to be handled
$ package=/usr/src/gnu/coreutils-4.5.4   # the package where it comes from
@end smallexample

She takes the latest copy of @file{$lang.po} from the Translation Project,
or from the package (in most cases, @file{$package/po/$lang.po}), or
creates a fresh one if she's the first translator (see @ref{Creating}).
She then uses the following commands to mark the not urgent messages as
"obsolete".  (This doesn't mean that these messages - translated and
untranslated ones - will go away.  It simply means that the PO file editor
will ignore them in the following editing session.)

@smallexample
$ msggrep --domain=$domain missing.po | grep -v '^domain' \
  > $domain-missing.po
$ msgattrib --set-obsolete --ignore-file $domain-missing.po $domain.$lang.po \
  > $domain.$lang-urgent.po
@end smallexample

The she translates @file{$domain.$lang-urgent.po} by use of a PO file editor
(@pxref{Editing}).
(FIXME: I don't know whether @code{KBabel} and @code{gtranslator} also
preserve obsolete messages, as they should.)
Finally she restores the not urgent messages (with their earlier
translations, for those which were already translated) through this command:

@smallexample
$ msgmerge --no-fuzzy-matching $domain.$lang-urgent.po $package/po/$domain.pot \
  > $domain.$lang.po
@end smallexample

Then she can submit @file{$domain.$lang.po} and proceed to the next domain.

@node Maintainers, Installers, Translators, Top
@chapter The Maintainer's View
@cindex package maintainer's view of @code{gettext}

The maintainer of a package has many responsibilities.  One of them
is ensuring that the package will install easily on many platforms,
and that the magic we described earlier (@pxref{Users}) will work
for installers and end users.

Of course, there are many possible ways by which GNU @code{gettext}
might be integrated in a distribution, and this chapter does not cover
them in all generality.  Instead, it details one possible approach which
is especially adequate for many free software distributions following GNU
standards, or even better, Gnits standards, because GNU @code{gettext}
is purposely for helping the internationalization of the whole GNU
project, and as many other good free packages as possible.  So, the
maintainer's view presented here presumes that the package already has
a @file{configure.ac} file and uses GNU Autoconf.

Nevertheless, GNU @code{gettext} may surely be useful for free packages
not following GNU standards and conventions, but the maintainers of such
packages might have to show imagination and initiative in organizing
their distributions so @code{gettext} work for them in all situations.
There are surely many, out there.

Even if @code{gettext} methods are now stabilizing, slight adjustments
might be needed between successive @code{gettext} versions, so you
should ideally revise this chapter in subsequent releases, looking
for changes.

@menu
* Flat and Non-Flat::           Flat or Non-Flat Directory Structures
* Prerequisites::               Prerequisite Works
* gettextize Invocation::       Invoking the @code{gettextize} Program
* Adjusting Files::             Files You Must Create or Alter
* autoconf macros::             Autoconf macros for use in @file{configure.ac}
* Version Control Issues::      
* Release Management::          Creating a Distribution Tarball
@end menu

@node Flat and Non-Flat, Prerequisites, Maintainers, Maintainers
@section Flat or Non-Flat Directory Structures

Some free software packages are distributed as @code{tar} files which unpack
in a single directory, these are said to be @dfn{flat} distributions.
Other free software packages have a one level hierarchy of subdirectories, using
for example a subdirectory named @file{doc/} for the Texinfo manual and
man pages, another called @file{lib/} for holding functions meant to
replace or complement C libraries, and a subdirectory @file{src/} for
holding the proper sources for the package.  These other distributions
are said to be @dfn{non-flat}.

We cannot say much about flat distributions.  A flat
directory structure has the disadvantage of increasing the difficulty
of updating to a new version of GNU @code{gettext}.  Also, if you have
many PO files, this could somewhat pollute your single directory.
Also, GNU @code{gettext}'s libintl sources consist of C sources, shell
scripts, @code{sed} scripts and complicated Makefile rules, which don't
fit well into an existing flat structure.  For these reasons, we
recommend to use non-flat approach in this case as well.

Maybe because GNU @code{gettext} itself has a non-flat structure,
we have more experience with this approach, and this is what will be
described in the remaining of this chapter.  Some maintainers might
use this as an opportunity to unflatten their package structure.

@node Prerequisites, gettextize Invocation, Flat and Non-Flat, Maintainers
@section Prerequisite Works
@cindex converting a package to use @code{gettext}
@cindex migration from earlier versions of @code{gettext}
@cindex upgrading to new versions of @code{gettext}

There are some works which are required for using GNU @code{gettext}
in one of your package.  These works have some kind of generality
that escape the point by point descriptions used in the remainder
of this chapter.  So, we describe them here.

@itemize @bullet
@item
Before attempting to use @code{gettextize} you should install some
other packages first.
Ensure that recent versions of GNU @code{m4}, GNU Autoconf and GNU
@code{gettext} are already installed at your site, and if not, proceed
to do this first.  If you get to install these things, beware that
GNU @code{m4} must be fully installed before GNU Autoconf is even
@emph{configured}.

To further ease the task of a package maintainer the @code{automake}
package was designed and implemented.  GNU @code{gettext} now uses this
tool and the @file{Makefile}s in the @file{intl/} and @file{po/}
therefore know about all the goals necessary for using @code{automake}
and @file{libintl} in one project.

Those four packages are only needed by you, as a maintainer; the
installers of your own package and end users do not really need any of
GNU @code{m4}, GNU Autoconf, GNU @code{gettext}, or GNU @code{automake}
for successfully installing and running your package, with messages
properly translated.  But this is not completely true if you provide
internationalized shell scripts within your own package: GNU
@code{gettext} shall then be installed at the user site if the end users
want to see the translation of shell script messages.

@item
Your package should use Autoconf and have a @file{configure.ac} or
@file{configure.in} file.
If it does not, you have to learn how.  The Autoconf documentation
is quite well written, it is a good idea that you print it and get
familiar with it.

@item
Your C sources should have already been modified according to
instructions given earlier in this manual.  @xref{Sources}.

@item
Your @file{po/} directory should receive all PO files submitted to you
by the translator teams, each having @file{@var{ll}.po} as a name.
This is not usually easy to get translation
work done before your package gets internationalized and available!
Since the cycle has to start somewhere, the easiest for the maintainer
is to start with absolutely no PO files, and wait until various
translator teams get interested in your package, and submit PO files.

@end itemize

It is worth adding here a few words about how the maintainer should
ideally behave with PO files submissions.  As a maintainer, your role is
to authenticate the origin of the submission as being the representative
of the appropriate translating teams of the Translation Project (forward
the submission to @file{coordinator@@translationproject.org} in case of doubt),
to ensure that the PO file format is not severely broken and does not
prevent successful installation, and for the rest, to merely put these
PO files in @file{po/} for distribution.

As a maintainer, you do not have to take on your shoulders the
responsibility of checking if the translations are adequate or
complete, and should avoid diving into linguistic matters.  Translation
teams drive themselves and are fully responsible of their linguistic
choices for the Translation Project.  Keep in mind that translator teams are @emph{not}
driven by maintainers.  You can help by carefully redirecting all
communications and reports from users about linguistic matters to the
appropriate translation team, or explain users how to reach or join
their team.  The simplest might be to send them the @file{ABOUT-NLS} file.

Maintainers should @emph{never ever} apply PO file bug reports
themselves, short-cutting translation teams.  If some translator has
difficulty to get some of her points through her team, it should not be
an option for her to directly negotiate translations with maintainers.
Teams ought to settle their problems themselves, if any.  If you, as
a maintainer, ever think there is a real problem with a team, please
never try to @emph{solve} a team's problem on your own.

@node gettextize Invocation, Adjusting Files, Prerequisites, Maintainers
@section Invoking the @code{gettextize} Program

@include gettextize.texi

@node Adjusting Files, autoconf macros, gettextize Invocation, Maintainers
@section Files You Must Create or Alter
@cindex @code{gettext} files

Besides files which are automatically added through @code{gettextize},
there are many files needing revision for properly interacting with
GNU @code{gettext}.  If you are closely following GNU standards for
Makefile engineering and auto-configuration, the adaptations should
be easier to achieve.  Here is a point by point description of the
changes needed in each.

So, here comes a list of files, each one followed by a description of
all alterations it needs.  Many examples are taken out from the GNU
@code{gettext} @value{VERSION} distribution itself, or from the GNU
@code{hello} distribution (@uref{http://www.gnu.org/software/hello}).
You may indeed refer to the source code of the GNU @code{gettext} and
GNU @code{hello} packages, as they are intended to be good examples for
using GNU gettext functionality.

@menu
* po/POTFILES.in::              @file{POTFILES.in} in @file{po/}
* po/LINGUAS::                  @file{LINGUAS} in @file{po/}
* po/Makevars::                 @file{Makevars} in @file{po/}
* po/Rules-*::                  Extending @file{Makefile} in @file{po/}
* configure.ac::                @file{configure.ac} at top level
* config.guess::                @file{config.guess}, @file{config.sub} at top level
* mkinstalldirs::               @file{mkinstalldirs} at top level
* aclocal::                     @file{aclocal.m4} at top level
* acconfig::                    @file{acconfig.h} at top level
* config.h.in::                 @file{config.h.in} at top level
* Makefile::                    @file{Makefile.in} at top level
* src/Makefile::                @file{Makefile.in} in @file{src/}
* lib/gettext.h::               @file{gettext.h} in @file{lib/}
@end menu

@node po/POTFILES.in, po/LINGUAS, Adjusting Files, Adjusting Files
@subsection @file{POTFILES.in} in @file{po/}
@cindex @file{POTFILES.in} file

The @file{po/} directory should receive a file named
@file{POTFILES.in}.  This file tells which files, among all program
sources, have marked strings needing translation.  Here is an example
of such a file:

@example
@group
# List of source files containing translatable strings.
# Copyright (C) 1995 Free Software Foundation, Inc.

# Common library files
lib/error.c
lib/getopt.c
lib/xmalloc.c

# Package source files
src/gettext.c
src/msgfmt.c
src/xgettext.c
@end group
@end example

@noindent
Hash-marked comments and white lines are ignored.  All other lines
list those source files containing strings marked for translation
(@pxref{Mark Keywords}), in a notation relative to the top level
of your whole distribution, rather than the location of the
@file{POTFILES.in} file itself.

When a C file is automatically generated by a tool, like @code{flex} or
@code{bison}, that doesn't introduce translatable strings by itself,
it is recommended to list in @file{po/POTFILES.in} the real source file
(ending in @file{.l} in the case of @code{flex}, or in @file{.y} in the
case of @code{bison}), not the generated C file.

@node po/LINGUAS, po/Makevars, po/POTFILES.in, Adjusting Files
@subsection @file{LINGUAS} in @file{po/}
@cindex @file{LINGUAS} file

The @file{po/} directory should also receive a file named
@file{LINGUAS}.  This file contains the list of available translations.
It is a whitespace separated list.  Hash-marked comments and white lines
are ignored.  Here is an example file:

@example
@group
# Set of available languages.
de fr
@end group
@end example

@noindent
This example means that German and French PO files are available, so
that these languages are currently supported by your package.  If you
want to further restrict, at installation time, the set of installed
languages, this should not be done by modifying the @file{LINGUAS} file,
but rather by using the @code{LINGUAS} environment variable
(@pxref{Installers}).

It is recommended that you add the "languages" @samp{en@@quot} and
@samp{en@@boldquot} to the @code{LINGUAS} file.  @code{en@@quot} is a
variant of English message catalogs (@code{en}) which uses real quotation
marks instead of the ugly looking asymmetric ASCII substitutes @samp{`}
and @samp{'}.  @code{en@@boldquot} is a variant of @code{en@@quot} that
additionally outputs quoted pieces of text in a bold font, when used in
a terminal emulator which supports the VT100 escape sequences (such as
@code{xterm} or the Linux console, but not Emacs in @kbd{M-x shell} mode).

These extra message catalogs @samp{en@@quot} and @samp{en@@boldquot}
are constructed automatically, not by translators; to support them, you
need the files @file{Rules-quot}, @file{quot.sed}, @file{boldquot.sed},
@file{en@@quot.header}, @file{en@@boldquot.header}, @file{insert-header.sin}
in the @file{po/} directory.  You can copy them from GNU gettext's @file{po/}
directory; they are also installed by running @code{gettextize}.

@node po/Makevars, po/Rules-*, po/LINGUAS, Adjusting Files
@subsection @file{Makevars} in @file{po/}
@cindex @file{Makevars} file

The @file{po/} directory also has a file named @file{Makevars}.  It
contains variables that are specific to your project.  @file{po/Makevars}
gets inserted into the @file{po/Makefile} when the latter is created.
The variables thus take effect when the POT file is created or updated,
and when the message catalogs get installed.

The first three variables can be left unmodified if your package has a
single message domain and, accordingly, a single @file{po/} directory.
Only packages which have multiple @file{po/} directories at different
locations need to adjust the three first variables defined in
@file{Makevars}.

As an alternative to the @code{XGETTEXT_OPTIONS} variables, it is also
possible to specify @code{xgettext} options through the
@code{AM_XGETTEXT_OPTION} autoconf macro.  See @ref{AM_XGETTEXT_OPTION}.

@node po/Rules-*, configure.ac, po/Makevars, Adjusting Files
@subsection Extending @file{Makefile} in @file{po/}
@cindex @file{Makefile.in.in} extensions

All files called @file{Rules-*} in the @file{po/} directory get appended to
the @file{po/Makefile} when it is created.  They present an opportunity to
add rules for special PO files to the Makefile, without needing to mess
with @file{po/Makefile.in.in}.

@cindex quotation marks
@vindex LANGUAGE@r{, environment variable}
GNU gettext comes with a @file{Rules-quot} file, containing rules for
building catalogs @file{en@@quot.po} and @file{en@@boldquot.po}.  The
effect of @file{en@@quot.po} is that people who set their @code{LANGUAGE}
environment variable to @samp{en@@quot} will get messages with proper
looking symmetric Unicode quotation marks instead of abusing the ASCII
grave accent and the ASCII apostrophe for indicating quotations.  To
enable this catalog, simply add @code{en@@quot} to the @file{po/LINGUAS}
file.  The effect of @file{en@@boldquot.po} is that people who set
@code{LANGUAGE} to @samp{en@@boldquot} will get not only proper quotation
marks, but also the quoted text will be shown in a bold font on terminals
and consoles.  This catalog is useful only for command-line programs, not
GUI programs.  To enable it, similarly add @code{en@@boldquot} to the
@file{po/LINGUAS} file.

Similarly, you can create rules for building message catalogs for the
@file{sr@@latin} locale -- Serbian written with the Latin alphabet --
from those for the @file{sr} locale -- Serbian written with Cyrillic
letters.  See @ref{msgfilter Invocation}.

@node configure.ac, config.guess, po/Rules-*, Adjusting Files
@subsection @file{configure.ac} at top level

@file{configure.ac} or @file{configure.in} - this is the source from which
@code{autoconf} generates the @file{configure} script.

@enumerate
@item Declare the package and version.
@cindex package and version declaration in @file{configure.ac}

This is done by a set of lines like these:

@example
PACKAGE=gettext
VERSION=@value{VERSION}
AC_DEFINE_UNQUOTED(PACKAGE, "$PACKAGE")
AC_DEFINE_UNQUOTED(VERSION, "$VERSION")
AC_SUBST(PACKAGE)
AC_SUBST(VERSION)
@end example

@noindent
or, if you are using GNU @code{automake}, by a line like this:

@example
AM_INIT_AUTOMAKE(gettext, @value{VERSION})
@end example

@noindent
Of course, you replace @samp{gettext} with the name of your package,
and @samp{@value{VERSION}} by its version numbers, exactly as they
should appear in the packaged @code{tar} file name of your distribution
(@file{gettext-@value{VERSION}.tar.gz}, here).

@item Check for internationalization support.

Here is the main @code{m4} macro for triggering internationalization
support.  Just add this line to @file{configure.ac}:

@example
AM_GNU_GETTEXT
@end example

@noindent
This call is purposely simple, even if it generates a lot of configure
time checking and actions.

If you have suppressed the @file{intl/} subdirectory by calling
@code{gettextize} without @samp{--intl} option, this call should read

@example
AM_GNU_GETTEXT([external])
@end example

@item Have output files created.

The @code{AC_OUTPUT} directive, at the end of your @file{configure.ac}
file, needs to be modified in two ways:

@example
AC_OUTPUT([@var{existing configuration files} intl/Makefile po/Makefile.in],
[@var{existing additional actions}])
@end example

The modification to the first argument to @code{AC_OUTPUT} asks
for substitution in the @file{intl/} and @file{po/} directories.
Note the @samp{.in} suffix used for @file{po/} only.  This is because
the distributed file is really @file{po/Makefile.in.in}.

If you have suppressed the @file{intl/} subdirectory by calling
@code{gettextize} without @samp{--intl} option, then you don't need to
add @code{intl/Makefile} to the @code{AC_OUTPUT} line.

@end enumerate

If, after doing the recommended modifications, a command like
@samp{aclocal -I m4} or @samp{autoconf} or @samp{autoreconf} fails with
a trace similar to this:

@smallexample
configure.ac:44: warning: AC_COMPILE_IFELSE was called before AC_GNU_SOURCE
../../lib/autoconf/specific.m4:335: AC_GNU_SOURCE is expanded from...
m4/lock.m4:224: gl_LOCK is expanded from...
m4/gettext.m4:571: gt_INTL_SUBDIR_CORE is expanded from...
m4/gettext.m4:472: AM_INTL_SUBDIR is expanded from...
m4/gettext.m4:347: AM_GNU_GETTEXT is expanded from...
configure.ac:44: the top level
configure.ac:44: warning: AC_RUN_IFELSE was called before AC_GNU_SOURCE
@end smallexample

@noindent
you need to add an explicit invocation of @samp{AC_GNU_SOURCE} in the
@file{configure.ac} file - after @samp{AC_PROG_CC} but before
@samp{AM_GNU_GETTEXT}, most likely very close to the @samp{AC_PROG_CC}
invocation.  This is necessary because of ordering restrictions imposed
by GNU autoconf.

@node config.guess, mkinstalldirs, configure.ac, Adjusting Files
@subsection @file{config.guess}, @file{config.sub} at top level

If you haven't suppressed the @file{intl/} subdirectory,
you need to add the GNU @file{config.guess} and @file{config.sub} files
to your distribution.  They are needed because the @file{intl/} directory
has platform dependent support for determining the locale's character
encoding and therefore needs to identify the platform.

You can obtain the newest version of @file{config.guess} and
@file{config.sub} from the @samp{config} project at
@file{http://savannah.gnu.org/}. The commands to fetch them are
@smallexample
$ wget -O config.guess 'http://git.savannah.gnu.org/gitweb/?p=config.git;a=blob_plain;f=config.guess;hb=HEAD'
$ wget -O config.sub 'http://git.savannah.gnu.org/gitweb/?p=config.git;a=blob_plain;f=config.sub;hb=HEAD'
@end smallexample
@noindent
Less recent versions are also contained in the GNU @code{automake} and
GNU @code{libtool} packages.

Normally, @file{config.guess} and @file{config.sub} are put at the
top level of a distribution.  But it is also possible to put them in a
subdirectory, altogether with other configuration support files like
@file{install-sh}, @file{ltconfig}, @file{ltmain.sh} or @file{missing}.
All you need to do, other than moving the files, is to add the following line
to your @file{configure.ac}.

@example
AC_CONFIG_AUX_DIR([@var{subdir}])
@end example

@node mkinstalldirs, aclocal, config.guess, Adjusting Files
@subsection @file{mkinstalldirs} at top level
@cindex @file{mkinstalldirs} file

With earlier versions of GNU gettext, you needed to add the GNU
@file{mkinstalldirs} script to your distribution.  This is not needed any
more.  You can remove it if you not also using an automake version older than
automake 1.9.

@node aclocal, acconfig, mkinstalldirs, Adjusting Files
@subsection @file{aclocal.m4} at top level
@cindex @file{aclocal.m4} file

If you do not have an @file{aclocal.m4} file in your distribution,
the simplest is to concatenate the files @file{codeset.m4}, @file{fcntl-o.m4},
@file{gettext.m4}, @file{glibc2.m4}, @file{glibc21.m4}, @file{iconv.m4},
@file{intdiv0.m4}, @file{intl.m4}, @file{intldir.m4}, @file{intlmacosx.m4},
@file{intmax.m4}, @file{inttypes_h.m4}, @file{inttypes-pri.m4},
@file{lcmessage.m4}, @file{lib-ld.m4}, @file{lib-link.m4},
@file{lib-prefix.m4}, @file{lock.m4}, @file{longlong.m4}, @file{nls.m4},
@file{po.m4}, @file{printf-posix.m4}, @file{progtest.m4}, @file{size_max.m4},
@file{stdint_h.m4}, @file{threadlib.m4}, @file{uintmax_t.m4},
@file{visibility.m4}, @file{wchar_t.m4}, @file{wint_t.m4}, @file{xsize.m4}
from GNU @code{gettext}'s
@file{m4/} directory into a single file.  If you have suppressed the
@file{intl/} directory, only @file{gettext.m4}, @file{iconv.m4},
@file{lib-ld.m4}, @file{lib-link.m4}, @file{lib-prefix.m4},
@file{nls.m4}, @file{po.m4}, @file{progtest.m4} need to be concatenated.

If you are not using GNU @code{automake} 1.8 or newer, you will need to
add a file @file{mkdirp.m4} from a newer automake distribution to the
list of files above.

If you already have an @file{aclocal.m4} file, then you will have
to merge the said macro files into your @file{aclocal.m4}.  Note that if
you are upgrading from a previous release of GNU @code{gettext}, you
should most probably @emph{replace} the macros (@code{AM_GNU_GETTEXT},
etc.), as they usually
change a little from one release of GNU @code{gettext} to the next.
Their contents may vary as we get more experience with strange systems
out there.

If you are using GNU @code{automake} 1.5 or newer, it is enough to put
these macro files into a subdirectory named @file{m4/} and add the line

@example
ACLOCAL_AMFLAGS = -I m4
@end example

@noindent
to your top level @file{Makefile.am}.

If you are using GNU @code{automake} 1.10 or newer, it is even easier:
Add the line

@example
ACLOCAL_AMFLAGS = --install -I m4
@end example

@noindent
to your top level @file{Makefile.am}, and run @samp{aclocal --install -I m4}.
This will copy the needed files to the @file{m4/} subdirectory automatically,
before updating @file{aclocal.m4}.

These macros check for the internationalization support functions
and related informations.  Hopefully, once stabilized, these macros
might be integrated in the standard Autoconf set, because this
piece of @code{m4} code will be the same for all projects using GNU
@code{gettext}.

@node acconfig, config.h.in, aclocal, Adjusting Files
@subsection @file{acconfig.h} at top level
@cindex @file{acconfig.h} file

Earlier GNU @code{gettext} releases required to put definitions for
@code{ENABLE_NLS}, @code{HAVE_GETTEXT} and @code{HAVE_LC_MESSAGES},
@code{HAVE_STPCPY}, @code{PACKAGE} and @code{VERSION} into an
@file{acconfig.h} file.  This is not needed any more; you can remove
them from your @file{acconfig.h} file unless your package uses them
independently from the @file{intl/} directory.

@node config.h.in, Makefile, acconfig, Adjusting Files
@subsection @file{config.h.in} at top level
@cindex @file{config.h.in} file

The include file template that holds the C macros to be defined by
@code{configure} is usually called @file{config.h.in} and may be
maintained either manually or automatically.

If @code{gettextize} has created an @file{intl/} directory, this file
must be called @file{config.h.in} and must be at the top level.  If,
however, you have suppressed the @file{intl/} directory by calling
@code{gettextize} without @samp{--intl} option, then you can choose the
name of this file and its location freely.

If it is maintained automatically, by use of the @samp{autoheader}
program, you need to do nothing about it.  This is the case in particular
if you are using GNU @code{automake}.

If it is maintained manually, and if @code{gettextize} has created an
@file{intl/} directory, you should switch to using @samp{autoheader}.
The list of C macros to be added for the sake of the @file{intl/}
directory is just too long to be maintained manually; it also changes
between different versions of GNU @code{gettext}.

If it is maintained manually, and if on the other hand you have
suppressed the @file{intl/} directory by calling @code{gettextize}
without @samp{--intl} option, then you can get away by adding the
following lines to @file{config.h.in}:

@example
/* Define to 1 if translation of program messages to the user's
   native language is requested. */
#undef ENABLE_NLS
@end example

@node Makefile, src/Makefile, config.h.in, Adjusting Files
@subsection @file{Makefile.in} at top level

Here are a few modifications you need to make to your main, top-level
@file{Makefile.in} file.

@enumerate
@item
Add the following lines near the beginning of your @file{Makefile.in},
so the @samp{dist:} goal will work properly (as explained further down):

@example
PACKAGE = @@PACKAGE@@
VERSION = @@VERSION@@
@end example

@item
Add file @file{ABOUT-NLS} to the @code{DISTFILES} definition, so the file gets
distributed.

@item
Wherever you process subdirectories in your @file{Makefile.in}, be sure
you also process the subdirectories @samp{intl} and @samp{po}.  Special
rules in the @file{Makefiles} take care for the case where no
internationalization is wanted.

If you are using Makefiles, either generated by automake, or hand-written
so they carefully follow the GNU coding standards, the effected goals for
which the new subdirectories must be handled include @samp{installdirs},
@samp{install}, @samp{uninstall}, @samp{clean}, @samp{distclean}.

Here is an example of a canonical order of processing.  In this
example, we also define @code{SUBDIRS} in @code{Makefile.in} for it
to be further used in the @samp{dist:} goal.

@example
SUBDIRS = doc intl lib src po
@end example

Note that you must arrange for @samp{make} to descend into the
@code{intl} directory before descending into other directories containing
code which make use of the @code{libintl.h} header file.  For this
reason, here we mention @code{intl} before @code{lib} and @code{src}.

@item
A delicate point is the @samp{dist:} goal, as both
@file{intl/Makefile} and @file{po/Makefile} will later assume that the
proper directory has been set up from the main @file{Makefile}.  Here is
an example at what the @samp{dist:} goal might look like:

@example
distdir = $(PACKAGE)-$(VERSION)
dist: Makefile
        rm -fr $(distdir)
        mkdir $(distdir)
        chmod 777 $(distdir)
        for file in $(DISTFILES); do \
          ln $$file $(distdir) 2>/dev/null || cp -p $$file $(distdir); \
        done
        for subdir in $(SUBDIRS); do \
          mkdir $(distdir)/$$subdir || exit 1; \
          chmod 777 $(distdir)/$$subdir; \
          (cd $$subdir && $(MAKE) $@@) || exit 1; \
        done
        tar chozf $(distdir).tar.gz $(distdir)
        rm -fr $(distdir)
@end example

@end enumerate

Note that if you are using GNU @code{automake}, @file{Makefile.in} is
automatically generated from @file{Makefile.am}, and all needed changes
to @file{Makefile.am} are already made by running @samp{gettextize}.

@node src/Makefile, lib/gettext.h, Makefile, Adjusting Files
@subsection @file{Makefile.in} in @file{src/}

Some of the modifications made in the main @file{Makefile.in} will
also be needed in the @file{Makefile.in} from your package sources,
which we assume here to be in the @file{src/} subdirectory.  Here are
all the modifications needed in @file{src/Makefile.in}:

@enumerate
@item
In view of the @samp{dist:} goal, you should have these lines near the
beginning of @file{src/Makefile.in}:

@example
PACKAGE = @@PACKAGE@@
VERSION = @@VERSION@@
@end example

@item
If not done already, you should guarantee that @code{top_srcdir}
gets defined.  This will serve for @code{cpp} include files.  Just add
the line:

@example
top_srcdir = @@top_srcdir@@
@end example

@item
You might also want to define @code{subdir} as @samp{src}, later
allowing for almost uniform @samp{dist:} goals in all your
@file{Makefile.in}.  At list, the @samp{dist:} goal below assume that
you used:

@example
subdir = src
@end example

@item
The @code{main} function of your program will normally call
@code{bindtextdomain} (see @pxref{Triggering}), like this:

@example
bindtextdomain (@var{PACKAGE}, LOCALEDIR);
textdomain (@var{PACKAGE});
@end example

To make LOCALEDIR known to the program, add the following lines to
@file{Makefile.in} if you are using Autoconf version 2.60 or newer:

@example
datadir = @@datadir@@
datarootdir= @@datarootdir@@
localedir = @@localedir@@
DEFS = -DLOCALEDIR=\"$(localedir)\" @@DEFS@@
@end example

or these lines if your version of Autoconf is older than 2.60:

@example
datadir = @@datadir@@
localedir = $(datadir)/locale
DEFS = -DLOCALEDIR=\"$(localedir)\" @@DEFS@@
@end example

Note that @code{@@datadir@@} defaults to @samp{$(prefix)/share}, thus
@code{$(localedir)} defaults to @samp{$(prefix)/share/locale}.

@item
You should ensure that the final linking will use @code{@@LIBINTL@@} or
@code{@@LTLIBINTL@@} as a library.  @code{@@LIBINTL@@} is for use without
@code{libtool}, @code{@@LTLIBINTL@@} is for use with @code{libtool}.  An
easy way to achieve this is to manage that it gets into @code{LIBS}, like
this:

@example
LIBS = @@LIBINTL@@ @@LIBS@@
@end example

In most packages internationalized with GNU @code{gettext}, one will
find a directory @file{lib/} in which a library containing some helper
functions will be build.  (You need at least the few functions which the
GNU @code{gettext} Library itself needs.)  However some of the functions
in the @file{lib/} also give messages to the user which of course should be
translated, too.  Taking care of this, the support library (say
@file{libsupport.a}) should be placed before @code{@@LIBINTL@@} and
@code{@@LIBS@@} in the above example.  So one has to write this:

@example
LIBS = ../lib/libsupport.a @@LIBINTL@@ @@LIBS@@
@end example

@item
You should also ensure that directory @file{intl/} will be searched for
C preprocessor include files in all circumstances.  So, you have to
manage so both @samp{-I../intl} and @samp{-I$(top_srcdir)/intl} will
be given to the C compiler.

@item
Your @samp{dist:} goal has to conform with others.  Here is a
reasonable definition for it:

@example
distdir = ../$(PACKAGE)-$(VERSION)/$(subdir)
dist: Makefile $(DISTFILES)
        for file in $(DISTFILES); do \
          ln $$file $(distdir) 2>/dev/null || cp -p $$file $(distdir) || exit 1; \
        done
@end example

@end enumerate

Note that if you are using GNU @code{automake}, @file{Makefile.in} is
automatically generated from @file{Makefile.am}, and the first three
changes and the last change are not necessary.  The remaining needed
@file{Makefile.am} modifications are the following:

@enumerate
@item
To make LOCALEDIR known to the program, add the following to
@file{Makefile.am}:

@example
<module>_CPPFLAGS = -DLOCALEDIR=\"$(localedir)\"
@end example

@noindent
for each specific module or compilation unit, or

@example
AM_CPPFLAGS = -DLOCALEDIR=\"$(localedir)\"
@end example

for all modules and compilation units together.  Furthermore, if you are
using an Autoconf version older then 2.60, add this line to define
@samp{localedir}:

@example
localedir = $(datadir)/locale
@end example

@item
To ensure that the final linking will use @code{@@LIBINTL@@} or
@code{@@LTLIBINTL@@} as a library, add the following to
@file{Makefile.am}:

@example
<program>_LDADD = @@LIBINTL@@
@end example

@noindent
for each specific program, or

@example
LDADD = @@LIBINTL@@
@end example

for all programs together.  Remember that when you use @code{libtool}
to link a program, you need to use @@LTLIBINTL@@ instead of @@LIBINTL@@
for that program.

@item
If you have an @file{intl/} directory, whose contents is created by
@code{gettextize}, then to ensure that it will be searched for
C preprocessor include files in all circumstances, add something like
this to @file{Makefile.am}:

@example
AM_CPPFLAGS = -I../intl -I$(top_srcdir)/intl
@end example

@end enumerate

@node lib/gettext.h,  , src/Makefile, Adjusting Files
@subsection @file{gettext.h} in @file{lib/}
@cindex @file{gettext.h} file
@cindex turning off NLS support
@cindex disabling NLS

Internationalization of packages, as provided by GNU @code{gettext}, is
optional.  It can be turned off in two situations:

@itemize @bullet
@item
When the installer has specified @samp{./configure --disable-nls}.  This
can be useful when small binaries are more important than features, for
example when building utilities for boot diskettes.  It can also be useful
in order to get some specific C compiler warnings about code quality with
some older versions of GCC (older than 3.0).

@item
When the package does not include the @code{intl/} subdirectory, and the
libintl.h header (with its associated libintl library, if any) is not
already installed on the system, it is preferable that the package builds
without internationalization support, rather than to give a compilation
error.
@end itemize

A C preprocessor macro can be used to detect these two cases.  Usually,
when @code{libintl.h} was found and not explicitly disabled, the
@code{ENABLE_NLS} macro will be defined to 1 in the autoconf generated
configuration file (usually called @file{config.h}).  In the two negative
situations, however, this macro will not be defined, thus it will evaluate
to 0 in C preprocessor expressions.

@cindex include file @file{libintl.h}
@file{gettext.h} is a convenience header file for conditional use of
@file{<libintl.h>}, depending on the @code{ENABLE_NLS} macro.  If
@code{ENABLE_NLS} is set, it includes @file{<libintl.h>}; otherwise it
defines no-op substitutes for the libintl.h functions.  We recommend
the use of @code{"gettext.h"} over direct use of @file{<libintl.h>},
so that portability to older systems is guaranteed and installers can
turn off internationalization if they want to.  In the C code, you will
then write

@example
#include "gettext.h"
@end example

@noindent
instead of

@example
#include <libintl.h>
@end example

The location of @code{gettext.h} is usually in a directory containing
auxiliary include files.  In many GNU packages, there is a directory
@file{lib/} containing helper functions; @file{gettext.h} fits there.
In other packages, it can go into the @file{src} directory.

Do not install the @code{gettext.h} file in public locations.  Every
package that needs it should contain a copy of it on its own.

@node autoconf macros, Version Control Issues, Adjusting Files, Maintainers
@section Autoconf macros for use in @file{configure.ac}
@cindex autoconf macros for @code{gettext}

GNU @code{gettext} installs macros for use in a package's
@file{configure.ac} or @file{configure.in}.
@xref{Top, , Introduction, autoconf, The Autoconf Manual}.
The primary macro is, of course, @code{AM_GNU_GETTEXT}.

@menu
* AM_GNU_GETTEXT::              AM_GNU_GETTEXT in @file{gettext.m4}
* AM_GNU_GETTEXT_VERSION::      AM_GNU_GETTEXT_VERSION in @file{gettext.m4}
* AM_GNU_GETTEXT_NEED::         AM_GNU_GETTEXT_NEED in @file{gettext.m4}
* AM_GNU_GETTEXT_INTL_SUBDIR::  AM_GNU_GETTEXT_INTL_SUBDIR in @file{intldir.m4}
* AM_PO_SUBDIRS::               AM_PO_SUBDIRS in @file{po.m4}
* AM_XGETTEXT_OPTION::          AM_XGETTEXT_OPTION in @file{po.m4}
* AM_ICONV::                    AM_ICONV in @file{iconv.m4}
@end menu

@node AM_GNU_GETTEXT, AM_GNU_GETTEXT_VERSION, autoconf macros, autoconf macros
@subsection AM_GNU_GETTEXT in @file{gettext.m4}

@amindex AM_GNU_GETTEXT
The @code{AM_GNU_GETTEXT} macro tests for the presence of the GNU gettext
function family in either the C library or a separate @code{libintl}
library (shared or static libraries are both supported) or in the package's
@file{intl/} directory.  It also invokes @code{AM_PO_SUBDIRS}, thus preparing
the @file{po/} directories of the package for building.

@code{AM_GNU_GETTEXT} accepts up to three optional arguments.  The general
syntax is

@example
AM_GNU_GETTEXT([@var{intlsymbol}], [@var{needsymbol}], [@var{intldir}])
@end example

@c We don't document @var{intlsymbol} = @samp{use-libtool} here, because
@c it is of no use for packages other than GNU gettext itself.  (Such packages
@c are not allowed to install the shared libintl.  But if they use libtool,
@c then it is in order to install shared libraries that depend on libintl.)
@var{intlsymbol} can be @samp{external} or @samp{no-libtool}.  The default
(if it is not specified or empty) is @samp{no-libtool}.  @var{intlsymbol}
should be @samp{external} for packages with no @file{intl/} directory.
For packages with an @file{intl/} directory, you can either use an
@var{intlsymbol} equal to @samp{no-libtool}, or you can use @samp{external}
and override by using the macro @code{AM_GNU_GETTEXT_INTL_SUBDIR} elsewhere.
The two ways to specify the existence of an @file{intl/} directory are
equivalent.  At build time, a static library
@code{$(top_builddir)/intl/libintl.a} will then be created.

If @var{needsymbol} is specified and is @samp{need-ngettext}, then GNU
gettext implementations (in libc or libintl) without the @code{ngettext()}
function will be ignored.  If @var{needsymbol} is specified and is
@samp{need-formatstring-macros}, then GNU gettext implementations that don't
support the ISO C 99 @file{<inttypes.h>} formatstring macros will be ignored.
Only one @var{needsymbol} can be specified.  These requirements can also be
specified by using the macro @code{AM_GNU_GETTEXT_NEED} elsewhere.  To specify
more than one requirement, just specify the strongest one among them, or
invoke the @code{AM_GNU_GETTEXT_NEED} macro several times.  The hierarchy
among the various alternatives is as follows: @samp{need-formatstring-macros}
implies @samp{need-ngettext}.

@var{intldir} is used to find the intl libraries.  If empty, the value
@samp{$(top_builddir)/intl/} is used.

The @code{AM_GNU_GETTEXT} macro determines whether GNU gettext is
available and should be used.  If so, it sets the @code{USE_NLS} variable
to @samp{yes}; it defines @code{ENABLE_NLS} to 1 in the autoconf
generated configuration file (usually called @file{config.h}); it sets
the variables @code{LIBINTL} and @code{LTLIBINTL} to the linker options
for use in a Makefile (@code{LIBINTL} for use without libtool,
@code{LTLIBINTL} for use with libtool); it adds an @samp{-I} option to
@code{CPPFLAGS} if necessary.  In the negative case, it sets
@code{USE_NLS} to @samp{no}; it sets @code{LIBINTL} and @code{LTLIBINTL}
to empty and doesn't change @code{CPPFLAGS}.

The complexities that @code{AM_GNU_GETTEXT} deals with are the following:

@itemize @bullet
@item
@cindex @code{libintl} library
Some operating systems have @code{gettext} in the C library, for example
glibc.  Some have it in a separate library @code{libintl}.  GNU @code{libintl}
might have been installed as part of the GNU @code{gettext} package.

@item
GNU @code{libintl}, if installed, is not necessarily already in the search
path (@code{CPPFLAGS} for the include file search path, @code{LDFLAGS} for
the library search path).

@item
Except for glibc, the operating system's native @code{gettext} cannot
exploit the GNU mo files, doesn't have the necessary locale dependency
features, and cannot convert messages from the catalog's text encoding
to the user's locale encoding.

@item
GNU @code{libintl}, if installed, is not necessarily already in the
run time library search path.  To avoid the need for setting an environment
variable like @code{LD_LIBRARY_PATH}, the macro adds the appropriate
run time search path options to the @code{LIBINTL} and @code{LTLIBINTL}
variables.  This works on most systems, but not on some operating systems
with limited shared library support, like SCO.

@item
GNU @code{libintl} relies on POSIX/XSI @code{iconv}.  The macro checks for
linker options needed to use iconv and appends them to the @code{LIBINTL}
and @code{LTLIBINTL} variables.
@end itemize

@node AM_GNU_GETTEXT_VERSION, AM_GNU_GETTEXT_NEED, AM_GNU_GETTEXT, autoconf macros
@subsection AM_GNU_GETTEXT_VERSION in @file{gettext.m4}

@amindex AM_GNU_GETTEXT_VERSION
The @code{AM_GNU_GETTEXT_VERSION} macro declares the version number of
the GNU gettext infrastructure that is used by the package.

The use of this macro is optional; only the @code{autopoint} program makes
use of it (@pxref{Version Control Issues}).

@node AM_GNU_GETTEXT_NEED, AM_GNU_GETTEXT_INTL_SUBDIR, AM_GNU_GETTEXT_VERSION, autoconf macros
@subsection AM_GNU_GETTEXT_NEED in @file{gettext.m4}

@amindex AM_GNU_GETTEXT_NEED
The @code{AM_GNU_GETTEXT_NEED} macro declares a constraint regarding the
GNU gettext implementation.  The syntax is

@example
AM_GNU_GETTEXT_NEED([@var{needsymbol}])
@end example

If @var{needsymbol} is @samp{need-ngettext}, then GNU gettext implementations
(in libc or libintl) without the @code{ngettext()} function will be ignored.
If @var{needsymbol} is @samp{need-formatstring-macros}, then GNU gettext
implementations that don't support the ISO C 99 @file{<inttypes.h>}
formatstring macros will be ignored.

The optional second argument of @code{AM_GNU_GETTEXT} is also taken into
account.

The @code{AM_GNU_GETTEXT_NEED} invocations can occur before or after
the @code{AM_GNU_GETTEXT} invocation; the order doesn't matter.

@node AM_GNU_GETTEXT_INTL_SUBDIR, AM_PO_SUBDIRS, AM_GNU_GETTEXT_NEED, autoconf macros
@subsection AM_GNU_GETTEXT_INTL_SUBDIR in @file{intldir.m4}

@amindex AM_GNU_GETTEXT_INTL_SUBDIR
The @code{AM_GNU_GETTEXT_INTL_SUBDIR} macro specifies that the
@code{AM_GNU_GETTEXT} macro, although invoked with the first argument
@samp{external}, should also prepare for building the @file{intl/}
subdirectory.

The @code{AM_GNU_GETTEXT_INTL_SUBDIR} invocation can occur before or after
the @code{AM_GNU_GETTEXT} invocation; the order doesn't matter.

The use of this macro requires GNU automake 1.10 or newer and
GNU autoconf 2.61 or newer.

@node AM_PO_SUBDIRS, AM_XGETTEXT_OPTION, AM_GNU_GETTEXT_INTL_SUBDIR, autoconf macros
@subsection AM_PO_SUBDIRS in @file{po.m4}

@amindex AM_PO_SUBDIRS
The @code{AM_PO_SUBDIRS} macro prepares the @file{po/} directories of the
package for building.  This macro should be used in internationalized
programs written in other programming languages than C, C++, Objective C,
for example @code{sh}, @code{Python}, @code{Lisp}.  See @ref{Programming
Languages} for a list of programming languages that support localization
through PO files.

The @code{AM_PO_SUBDIRS} macro determines whether internationalization
should be used.  If so, it sets the @code{USE_NLS} variable to @samp{yes},
otherwise to @samp{no}.  It also determines the right values for Makefile
variables in each @file{po/} directory.

@node AM_XGETTEXT_OPTION, AM_ICONV, AM_PO_SUBDIRS, autoconf macros
@subsection AM_XGETTEXT_OPTION in @file{po.m4}

@amindex AM_XGETTEXT_OPTION
The @code{AM_XGETTEXT_OPTION} macro registers a command-line option to be
used in the invocations of @code{xgettext} in the @file{po/} directories
of the package.

For example, if you have a source file that defines a function
@samp{error_at_line} whose fifth argument is a format string, you can use
@example
AM_XGETTEXT_OPTION([--flag=error_at_line:5:c-format])
@end example
@noindent
to instruct @code{xgettext} to mark all translatable strings in @samp{gettext}
invocations that occur as fifth argument to this function as @samp{c-format}.

See @ref{xgettext Invocation} for the list of options that @code{xgettext}
accepts.

The use of this macro is an alternative to the use of the
@samp{XGETTEXT_OPTIONS} variable in @file{po/Makevars}.

@node AM_ICONV,  , AM_XGETTEXT_OPTION, autoconf macros
@subsection AM_ICONV in @file{iconv.m4}

@amindex AM_ICONV
The @code{AM_ICONV} macro tests for the presence of the POSIX/XSI
@code{iconv} function family in either the C library or a separate
@code{libiconv} library.  If found, it sets the @code{am_cv_func_iconv}
variable to @samp{yes}; it defines @code{HAVE_ICONV} to 1 in the autoconf
generated configuration file (usually called @file{config.h}); it defines
@code{ICONV_CONST} to @samp{const} or to empty, depending on whether the
second argument of @code{iconv()} is of type @samp{const char **} or
@samp{char **}; it sets the variables @code{LIBICONV} and
@code{LTLIBICONV} to the linker options for use in a Makefile
(@code{LIBICONV} for use without libtool, @code{LTLIBICONV} for use with
libtool); it adds an @samp{-I} option to @code{CPPFLAGS} if
necessary.  If not found, it sets @code{LIBICONV} and @code{LTLIBICONV} to
empty and doesn't change @code{CPPFLAGS}.

The complexities that @code{AM_ICONV} deals with are the following:

@itemize @bullet
@item
@cindex @code{libiconv} library
Some operating systems have @code{iconv} in the C library, for example
glibc.  Some have it in a separate library @code{libiconv}, for example
OSF/1 or FreeBSD.  Regardless of the operating system, GNU @code{libiconv}
might have been installed.  In that case, it should be used instead of the
operating system's native @code{iconv}.

@item
GNU @code{libiconv}, if installed, is not necessarily already in the search
path (@code{CPPFLAGS} for the include file search path, @code{LDFLAGS} for
the library search path).

@item
GNU @code{libiconv} is binary incompatible with some operating system's
native @code{iconv}, for example on FreeBSD.  Use of an @file{iconv.h}
and @file{libiconv.so} that don't fit together would produce program
crashes.

@item
GNU @code{libiconv}, if installed, is not necessarily already in the
run time library search path.  To avoid the need for setting an environment
variable like @code{LD_LIBRARY_PATH}, the macro adds the appropriate
run time search path options to the @code{LIBICONV} variable.  This works
on most systems, but not on some operating systems with limited shared
library support, like SCO.
@end itemize

@file{iconv.m4} is distributed with the GNU gettext package because
@file{gettext.m4} relies on it.

@node Version Control Issues, Release Management, autoconf macros, Maintainers
@section Integrating with Version Control Systems

Many projects use version control systems for distributed development
and source backup.  This section gives some advice how to manage the
uses of @code{gettextize}, @code{autopoint} and @code{autoconf} on
version controlled files.

@menu
* Distributed Development::     Avoiding version mismatch in distributed development
* Files under Version Control::  Files to put under version control
* Translations under Version Control::  Put PO Files under Version Control
* autopoint Invocation::        Invoking the @code{autopoint} Program
@end menu

@node Distributed Development, Files under Version Control, Version Control Issues, Version Control Issues
@subsection Avoiding version mismatch in distributed development

In a project development with multiple developers, there should be a
single developer who occasionally - when there is desire to upgrade to
a new @code{gettext} version - runs @code{gettextize} and performs the
changes listed in @ref{Adjusting Files}, and then commits his changes
to the repository.

It is highly recommended that all developers on a project use the same
version of GNU @code{gettext} in the package.  In other words, if a
developer runs @code{gettextize}, he should go the whole way, make the
necessary remaining changes and commit his changes to the repository.
Otherwise the following damages will likely occur:

@itemize @bullet
@item
Apparent version mismatch between developers.  Since some @code{gettext}
specific portions in @file{configure.ac}, @file{configure.in} and
@code{Makefile.am}, @code{Makefile.in} files depend on the @code{gettext}
version, the use of infrastructure files belonging to different
@code{gettext} versions can easily lead to build errors.

@item
Hidden version mismatch.  Such version mismatch can also lead to
malfunctioning of the package, that may be undiscovered by the developers.
The worst case of hidden version mismatch is that internationalization
of the package doesn't work at all.

@item
Release risks.  All developers implicitly perform constant testing on
a package.  This is important in the days and weeks before a release.
If the guy who makes the release tar files uses a different version
of GNU @code{gettext} than the other developers, the distribution will
be less well tested than if all had been using the same @code{gettext}
version.  For example, it is possible that a platform specific bug goes
undiscovered due to this constellation.
@end itemize

@node Files under Version Control, Translations under Version Control, Distributed Development, Version Control Issues
@subsection Files to put under version control

There are basically three ways to deal with generated files in the
context of a version controlled repository, such as @file{configure}
generated from @file{configure.ac}, @code{@var{parser}.c} generated
from @code{@var{parser}.y}, or @code{po/Makefile.in.in} autoinstalled
by @code{gettextize} or @code{autopoint}.

@enumerate
@item
All generated files are always committed into the repository.

@item
All generated files are committed into the repository occasionally,
for example each time a release is made.

@item
Generated files are never committed into the repository.
@end enumerate

Each of these three approaches has different advantages and drawbacks.

@enumerate
@item
The advantage is that anyone can check out the source at any moment and
gets a working build.  The drawbacks are:  1a. It requires some frequent
"push" actions by the maintainers.  1b. The repository grows in size
quite fast.

@item
The advantage is that anyone can check out the source, and the usual
"./configure; make" will work.  The drawbacks are: 2a. The one who
checks out the repository needs tools like GNU @code{automake}, GNU
@code{autoconf}, GNU @code{m4} installed in his PATH; sometimes he
even needs particular versions of them.  2b. When a release is made
and a commit is made on the generated files, the other developers get
conflicts on the generated files when merging the local work back to
the repository.  Although these conflicts are easy to resolve, they
are annoying.

@item
The advantage is less work for the maintainers.  The drawback is that
anyone who checks out the source not only needs tools like GNU
@code{automake}, GNU @code{autoconf}, GNU @code{m4} installed in his
PATH, but also that he needs to perform a package specific pre-build
step before being able to "./configure; make".
@end enumerate

For the first and second approach, all files modified or brought in
by the occasional @code{gettextize} invocation and update should be
committed into the repository.

For the third approach, the maintainer can omit from the repository
all the files that @code{gettextize} mentions as "copy".  Instead, he
adds to the @file{configure.ac} or @file{configure.in} a line of the
form

@example
AM_GNU_GETTEXT_VERSION(@value{ARCHIVE-VERSION})
@end example

@noindent
and adds to the package's pre-build script an invocation of
@samp{autopoint}.  For everyone who checks out the source, this
@code{autopoint} invocation will copy into the right place the
@code{gettext} infrastructure files that have been omitted from the repository.

The version number used as argument to @code{AM_GNU_GETTEXT_VERSION} is
the version of the @code{gettext} infrastructure that the package wants
to use.  It is also the minimum version number of the @samp{autopoint}
program.  So, if you write @code{AM_GNU_GETTEXT_VERSION(0.11.5)} then the
developers can have any version >= 0.11.5 installed; the package will work
with the 0.11.5 infrastructure in all developers' builds.  When the
maintainer then runs gettextize from, say, version 0.12.1 on the package,
the occurrence of @code{AM_GNU_GETTEXT_VERSION(0.11.5)} will be changed
into @code{AM_GNU_GETTEXT_VERSION(0.12.1)}, and all other developers that
use the CVS will henceforth need to have GNU @code{gettext} 0.12.1 or newer
installed.

@node Translations under Version Control, autopoint Invocation, Files under Version Control, Version Control Issues
@subsection Put PO Files under Version Control

Since translations are valuable assets as well as the source code, it
would make sense to put them under version control.  The GNU gettext
infrastructure supports two ways to deal with translations in the
context of a version controlled repository.

@enumerate
@item
Both POT file and PO files are committed into the repository.

@item
Only PO files are committed into the repository.

@end enumerate

If a POT file is absent when building, it will be generated by
scanning the source files with @code{xgettext}, and then the PO files
are regenerated as a dependency.  On the other hand, some maintainers
want to keep the POT file unchanged during the development phase.  So,
even if a POT file is present and older than the source code, it won't
be updated automatically.  You can manually update it with @code{make
$(DOMAIN).pot-update}, and commit it at certain point.

Special advices for particular version control systems:

@itemize @bullet
@item
Recent version control systems, Git for instance, ignore file's
timestamp.  In that case, PO files can be accidentally updated even if
a POT file is not updated.  To prevent this, you can set
@samp{PO_DEPENDS_ON_POT} variable to @code{no} in the @file{Makevars}
file and do @code{make update-po} manually.

@item
Location comments such as @code{#: lib/error.c:116} are sometimes
annoying, since these comments are volatile and may introduce unwanted
change to the working copy when building.  To mitigate this, you can
decide to omit those comments from the PO files in the repository.

This is possible with the @code{--no-location} option of the
@code{msgmerge} command @footnote{you can also use it through the
@samp{MSGMERGE_OPTIONS} option from @file{Makevars}}.  The drawback is
that, if the location information is needed, translators have to
recover the location comments by running @code{msgmerge} again.

@end itemize

@node autopoint Invocation,  , Translations under Version Control, Version Control Issues
@subsection Invoking the @code{autopoint} Program

@include autopoint.texi

@node Release Management,  , Version Control Issues, Maintainers
@section Creating a Distribution Tarball

@cindex release
@cindex distribution tarball
In projects that use GNU @code{automake}, the usual commands for creating
a distribution tarball, @samp{make dist} or @samp{make distcheck},
automatically update the PO files as needed.

If GNU @code{automake} is not used, the maintainer needs to perform this
update before making a release:

@example
$ ./configure
$ (cd po; make update-po)
$ make distclean
@end example

@node Installers, Programming Languages, Maintainers, Top
@chapter The Installer's and Distributor's View
@cindex package installer's view of @code{gettext}
@cindex package distributor's view of @code{gettext}
@cindex package build and installation options
@cindex setting up @code{gettext} at build time

By default, packages fully using GNU @code{gettext}, internally,
are installed in such a way as to allow translation of
messages.  At @emph{configuration} time, those packages should
automatically detect whether the underlying host system already provides
the GNU @code{gettext} functions.  If not,
the GNU @code{gettext} library should be automatically prepared
and used.  Installers may use special options at configuration
time for changing this behavior.  The command @samp{./configure
--with-included-gettext} bypasses system @code{gettext} to
use the included GNU @code{gettext} instead,
while @samp{./configure --disable-nls}
produces programs totally unable to translate messages.

@vindex LINGUAS@r{, environment variable}
Internationalized packages have usually many @file{@var{ll}.po}
files.  Unless
translations are disabled, all those available are installed together
with the package.  However, the environment variable @code{LINGUAS}
may be set, prior to configuration, to limit the installed set.
@code{LINGUAS} should then contain a space separated list of two-letter
codes, stating which languages are allowed.

@node Programming Languages, Conclusion, Installers, Top
@chapter Other Programming Languages

While the presentation of @code{gettext} focuses mostly on C and
implicitly applies to C++ as well, its scope is far broader than that:
Many programming languages, scripting languages and other textual data
like GUI resources or package descriptions can make use of the gettext
approach.

@menu
* Language Implementors::       The Language Implementor's View
* Programmers for other Languages::  The Programmer's View
* Translators for other Languages::  The Translator's View
* Maintainers for other Languages::  The Maintainer's View
* List of Programming Languages::  Individual Programming Languages
* List of Data Formats::        Internationalizable Data
@end menu

@node Language Implementors, Programmers for other Languages, Programming Languages, Programming Languages
@section The Language Implementor's View
@cindex programming languages
@cindex scripting languages

All programming and scripting languages that have the notion of strings
are eligible to supporting @code{gettext}.  Supporting @code{gettext}
means the following:

@enumerate
@item
You should add to the language a syntax for translatable strings.  In
principle, a function call of @code{gettext} would do, but a shorthand
syntax helps keeping the legibility of internationalized programs.  For
example, in C we use the syntax @code{_("string")}, and in GNU awk we use
the shorthand @code{_"string"}.

@item
You should arrange that evaluation of such a translatable string at
runtime calls the @code{gettext} function, or performs equivalent
processing.

@item
Similarly, you should make the functions @code{ngettext},
@code{dcgettext}, @code{dcngettext} available from within the language.
These functions are less often used, but are nevertheless necessary for
particular purposes: @code{ngettext} for correct plural handling, and
@code{dcgettext} and @code{dcngettext} for obeying other locale-related
environment variables than @code{LC_MESSAGES}, such as @code{LC_TIME} or
@code{LC_MONETARY}.  For these latter functions, you need to make the
@code{LC_*} constants, available in the C header @code{<locale.h>},
referenceable from within the language, usually either as enumeration
values or as strings.

@item
You should allow the programmer to designate a message domain, either by
making the @code{textdomain} function available from within the
language, or by introducing a magic variable called @code{TEXTDOMAIN}.
Similarly, you should allow the programmer to designate where to search
for message catalogs, by providing access to the @code{bindtextdomain}
function.

@item
You should either perform a @code{setlocale (LC_ALL, "")} call during
the startup of your language runtime, or allow the programmer to do so.
Remember that gettext will act as a no-op if the @code{LC_MESSAGES} and
@code{LC_CTYPE} locale categories are not both set.

@item
A programmer should have a way to extract translatable strings from a
program into a PO file.  The GNU @code{xgettext} program is being
extended to support very different programming languages.  Please
contact the GNU @code{gettext} maintainers to help them doing this.  If
the string extractor is best integrated into your language's parser, GNU
@code{xgettext} can function as a front end to your string extractor.

@item
The language's library should have a string formatting facility where
the arguments of a format string are denoted by a positional number or a
name.  This is needed because for some languages and some messages with
more than one substitutable argument, the translation will need to
output the substituted arguments in different order.  @xref{c-format Flag}.

@item
If the language has more than one implementation, and not all of the
implementations use @code{gettext}, but the programs should be portable
across implementations, you should provide a no-i18n emulation, that
makes the other implementations accept programs written for yours,
without actually translating the strings.

@item
To help the programmer in the task of marking translatable strings,
which is sometimes performed using the Emacs PO mode (@pxref{Marking}),
you are welcome to
contact the GNU @code{gettext} maintainers, so they can add support for
your language to @file{po-mode.el}.
@end enumerate

On the implementation side, three approaches are possible, with
different effects on portability and copyright:

@itemize @bullet
@item
You may integrate the GNU @code{gettext}'s @file{intl/} directory in
your package, as described in @ref{Maintainers}.  This allows you to
have internationalization on all kinds of platforms.  Note that when you
then distribute your package, it legally falls under the GNU General
Public License, and the GNU project will be glad about your contribution
to the Free Software pool.

@item
You may link against GNU @code{gettext} functions if they are found in
the C library.  For example, an autoconf test for @code{gettext()} and
@code{ngettext()} will detect this situation.  For the moment, this test
will succeed on GNU systems and not on other platforms.  No severe
copyright restrictions apply.

@item
You may emulate or reimplement the GNU @code{gettext} functionality.
This has the advantage of full portability and no copyright
restrictions, but also the drawback that you have to reimplement the GNU
@code{gettext} features (such as the @code{LANGUAGE} environment
variable, the locale aliases database, the automatic charset conversion,
and plural handling).
@end itemize

@node Programmers for other Languages, Translators for other Languages, Language Implementors, Programming Languages
@section The Programmer's View

For the programmer, the general procedure is the same as for the C
language.  The Emacs PO mode marking supports other languages, and the GNU
@code{xgettext} string extractor recognizes other languages based on the
file extension or a command-line option.  In some languages,
@code{setlocale} is not needed because it is already performed by the
underlying language runtime.

@node Translators for other Languages, Maintainers for other Languages, Programmers for other Languages, Programming Languages
@section The Translator's View

The translator works exactly as in the C language case.  The only
difference is that when translating format strings, she has to be aware
of the language's particular syntax for positional arguments in format
strings.

@menu
* c-format::                    C Format Strings
* objc-format::                 Objective C Format Strings
* sh-format::                   Shell Format Strings
* python-format::               Python Format Strings
* lisp-format::                 Lisp Format Strings
* elisp-format::                Emacs Lisp Format Strings
* librep-format::               librep Format Strings
* scheme-format::               Scheme Format Strings
* smalltalk-format::            Smalltalk Format Strings
* java-format::                 Java Format Strings
* csharp-format::               C# Format Strings
* awk-format::                  awk Format Strings
* object-pascal-format::        Object Pascal Format Strings
* ycp-format::                  YCP Format Strings
* tcl-format::                  Tcl Format Strings
* perl-format::                 Perl Format Strings
* php-format::                  PHP Format Strings
* gcc-internal-format::         GCC internal Format Strings
* gfc-internal-format::         GFC internal Format Strings
* qt-format::                   Qt Format Strings
* qt-plural-format::            Qt Plural Format Strings
* kde-format::                  KDE Format Strings
* kde-kuit-format::             KUIT Format Strings
* boost-format::                Boost Format Strings
* lua-format::                  Lua Format Strings
* javascript-format::           JavaScript Format Strings
@end menu

@node c-format, objc-format, Translators for other Languages, Translators for other Languages
@subsection C Format Strings

C format strings are described in POSIX (IEEE P1003.1 2001), section
XSH 3 fprintf(),
@uref{http://www.opengroup.org/onlinepubs/007904975/functions/fprintf.html}.
See also the fprintf() manual page,
@uref{http://www.linuxvalley.it/encyclopedia/ldp/manpage/man3/printf.3.php},
@uref{http://informatik.fh-wuerzburg.de/student/i510/man/printf.html}.

Although format strings with positions that reorder arguments, such as

@example
"Only %2$d bytes free on '%1$s'."
@end example

@noindent
which is semantically equivalent to

@example
"'%s' has only %d bytes free."
@end example

@noindent
are a POSIX/XSI feature and not specified by ISO C 99, translators can rely
on this reordering ability: On the few platforms where @code{printf()},
@code{fprintf()} etc. don't support this feature natively, @file{libintl.a}
or @file{libintl.so} provides replacement functions, and GNU @code{<libintl.h>}
activates these replacement functions automatically.

@cindex outdigits
@cindex Arabic digits
As a special feature for Farsi (Persian) and maybe Arabic, translators can
insert an @samp{I} flag into numeric format directives.  For example, the
translation of @code{"%d"} can be @code{"%Id"}.  The effect of this flag,
on systems with GNU @code{libc}, is that in the output, the ASCII digits are
replaced with the @samp{outdigits} defined in the @code{LC_CTYPE} locale
category.  On other systems, the @code{gettext} function removes this flag,
so that it has no effect.

Note that the programmer should @emph{not} put this flag into the
untranslated string.  (Putting the @samp{I} format directive flag into an
@var{msgid} string would lead to undefined behaviour on platforms without
glibc when NLS is disabled.)

@node objc-format, sh-format, c-format, Translators for other Languages
@subsection Objective C Format Strings

Objective C format strings are like C format strings.  They support an
additional format directive: "%@@", which when executed consumes an argument
of type @code{Object *}.

@node sh-format, python-format, objc-format, Translators for other Languages
@subsection Shell Format Strings

Shell format strings, as supported by GNU gettext and the @samp{envsubst}
program, are strings with references to shell variables in the form
@code{$@var{variable}} or @code{$@{@var{variable}@}}.  References of the form
@code{$@{@var{variable}-@var{default}@}},
@code{$@{@var{variable}:-@var{default}@}},
@code{$@{@var{variable}=@var{default}@}},
@code{$@{@var{variable}:=@var{default}@}},
@code{$@{@var{variable}+@var{replacement}@}},
@code{$@{@var{variable}:+@var{replacement}@}},
@code{$@{@var{variable}?@var{ignored}@}},
@code{$@{@var{variable}:?@var{ignored}@}},
that would be valid inside shell scripts, are not supported.  The
@var{variable} names must consist solely of alphanumeric or underscore
ASCII characters, not start with a digit and be nonempty; otherwise such
a variable reference is ignored.

@node python-format, lisp-format, sh-format, Translators for other Languages
@subsection Python Format Strings

There are two kinds of format strings in Python: those acceptable to
the Python built-in format operator @code{%}, labelled as
@samp{python-format}, and those acceptable to the @code{format} method
of the @samp{str} object.

Python @code{%} format strings are described in
@w{Python Library reference} /
@w{5. Built-in Types} /
@w{5.6. Sequence Types} /
@w{5.6.2. String Formatting Operations}.
@uref{http://docs.python.org/2/library/stdtypes.html#string-formatting-operations}.

Python brace format strings are described in @w{PEP 3101 -- Advanced
String Formatting}, @uref{http://www.python.org/dev/peps/pep-3101/}.

@node lisp-format, elisp-format, python-format, Translators for other Languages
@subsection Lisp Format Strings

Lisp format strings are described in the Common Lisp HyperSpec,
chapter 22.3 @w{Formatted Output},
@uref{http://www.lisp.org/HyperSpec/Body/sec_22-3.html}.

@node elisp-format, librep-format, lisp-format, Translators for other Languages
@subsection Emacs Lisp Format Strings

Emacs Lisp format strings are documented in the Emacs Lisp reference,
section @w{Formatting Strings},
@uref{http://www.gnu.org/manual/elisp-manual-21-2.8/html_chapter/elisp_4.html#SEC75}.
Note that as of version 21, XEmacs supports numbered argument specifications
in format strings while FSF Emacs doesn't.

@node librep-format, scheme-format, elisp-format, Translators for other Languages
@subsection librep Format Strings

librep format strings are documented in the librep manual, section
@w{Formatted Output},
@url{http://librep.sourceforge.net/librep-manual.html#Formatted%20Output},
@url{http://www.gwinnup.org/research/docs/librep.html#SEC122}.

@node scheme-format, smalltalk-format, librep-format, Translators for other Languages
@subsection Scheme Format Strings

Scheme format strings are documented in the SLIB manual, section
@w{Format Specification}.

@node smalltalk-format, java-format, scheme-format, Translators for other Languages
@subsection Smalltalk Format Strings

Smalltalk format strings are described in the GNU Smalltalk documentation,
class @code{CharArray}, methods @samp{bindWith:} and
@samp{bindWithArguments:}.
@uref{http://www.gnu.org/software/smalltalk/gst-manual/gst_68.html#SEC238}.
In summary, a directive starts with @samp{%} and is followed by @samp{%}
or a nonzero digit (@samp{1} to @samp{9}).

@node java-format, csharp-format, smalltalk-format, Translators for other Languages
@subsection Java Format Strings

Java format strings are described in the JDK documentation for class
@code{java.text.MessageFormat},
@uref{http://java.sun.com/j2se/1.4/docs/api/java/text/MessageFormat.html}.
See also the ICU documentation
@uref{http://oss.software.ibm.com/icu/apiref/classMessageFormat.html}.

@node csharp-format, awk-format, java-format, Translators for other Languages
@subsection C# Format Strings

C# format strings are described in the .NET documentation for class
@code{System.String} and in
@uref{http://msdn.microsoft.com/library/default.asp?url=/library/en-us/cpguide/html/cpConFormattingOverview.asp}.

@node awk-format, object-pascal-format, csharp-format, Translators for other Languages
@subsection awk Format Strings

awk format strings are described in the gawk documentation, section
@w{Printf},
@uref{http://www.gnu.org/manual/gawk/html_node/Printf.html#Printf}.

@node object-pascal-format, ycp-format, awk-format, Translators for other Languages
@subsection Object Pascal Format Strings

Object Pascal format strings are described in the documentation of the
Free Pascal runtime library, section Format,
@uref{http://www.freepascal.org/docs-html/rtl/sysutils/format.html}.

@node ycp-format, tcl-format, object-pascal-format, Translators for other Languages
@subsection YCP Format Strings

YCP sformat strings are described in the libycp documentation
@uref{file:/usr/share/doc/packages/libycp/YCP-builtins.html}.
In summary, a directive starts with @samp{%} and is followed by @samp{%}
or a nonzero digit (@samp{1} to @samp{9}).

@node tcl-format, perl-format, ycp-format, Translators for other Languages
@subsection Tcl Format Strings

Tcl format strings are described in the @file{format.n} manual page,
@uref{http://www.scriptics.com/man/tcl8.3/TclCmd/format.htm}.

@node perl-format, php-format, tcl-format, Translators for other Languages
@subsection Perl Format Strings

There are two kinds format strings in Perl: those acceptable to the
Perl built-in function @code{printf}, labelled as @samp{perl-format},
and those acceptable to the @code{libintl-perl} function @code{__x},
labelled as @samp{perl-brace-format}.

Perl @code{printf} format strings are described in the @code{sprintf}
section of @samp{man perlfunc}.

Perl brace format strings are described in the
@file{Locale::TextDomain(3pm)} manual page of the CPAN package
libintl-perl.  In brief, Perl format uses placeholders put between
braces (@samp{@{} and @samp{@}}).  The placeholder must have the syntax
of simple identifiers.

@node php-format, gcc-internal-format, perl-format, Translators for other Languages
@subsection PHP Format Strings

PHP format strings are described in the documentation of the PHP function
@code{sprintf}, in @file{phpdoc/manual/function.sprintf.html} or
@uref{http://www.php.net/manual/en/function.sprintf.php}.

@node gcc-internal-format, gfc-internal-format, php-format, Translators for other Languages
@subsection GCC internal Format Strings

These format strings are used inside the GCC sources.  In such a format
string, a directive starts with @samp{%}, is optionally followed by a
size specifier @samp{l}, an optional flag @samp{+}, another optional flag
@samp{#}, and is finished by a specifier: @samp{%} denotes a literal
percent sign, @samp{c} denotes a character, @samp{s} denotes a string,
@samp{i} and @samp{d} denote an integer, @samp{o}, @samp{u}, @samp{x}
denote an unsigned integer, @samp{.*s} denotes a string preceded by a
width specification, @samp{H} denotes a @samp{location_t *} pointer,
@samp{D} denotes a general declaration, @samp{F} denotes a function
declaration, @samp{T} denotes a type, @samp{A} denotes a function argument,
@samp{C} denotes a tree code, @samp{E} denotes an expression, @samp{L}
denotes a programming language, @samp{O} denotes a binary operator,
@samp{P} denotes a function parameter, @samp{Q} denotes an assignment
operator, @samp{V} denotes a const/volatile qualifier.

@node gfc-internal-format, qt-format, gcc-internal-format, Translators for other Languages
@subsection GFC internal Format Strings

These format strings are used inside the GNU Fortran Compiler sources,
that is, the Fortran frontend in the GCC sources.  In such a format
string, a directive starts with @samp{%} and is finished by a
specifier: @samp{%} denotes a literal percent sign, @samp{C} denotes the
current source location, @samp{L} denotes a source location, @samp{c}
denotes a character, @samp{s} denotes a string, @samp{i} and @samp{d}
denote an integer, @samp{u} denotes an unsigned integer.  @samp{i},
@samp{d}, and @samp{u} may be preceded by a size specifier @samp{l}.

@node qt-format, qt-plural-format, gfc-internal-format, Translators for other Languages
@subsection Qt Format Strings

Qt format strings are described in the documentation of the QString class
@uref{file:/usr/lib/qt-4.3.0/doc/html/qstring.html}.
In summary, a directive consists of a @samp{%} followed by a digit. The same
directive cannot occur more than once in a format string.

@node qt-plural-format, kde-format, qt-format, Translators for other Languages
@subsection Qt Format Strings

Qt format strings are described in the documentation of the QObject::tr method
@uref{file:/usr/lib/qt-4.3.0/doc/html/qobject.html}.
In summary, the only allowed directive is @samp{%n}.

@node kde-format, kde-kuit-format, qt-plural-format, Translators for other Languages
@subsection KDE Format Strings

KDE 4 format strings are defined as follows:
A directive consists of a @samp{%} followed by a non-zero decimal number.
If a @samp{%n} occurs in a format strings, all of @samp{%1}, ..., @samp{%(n-1)}
must occur as well, except possibly one of them.

@node kde-kuit-format, boost-format, kde-format, Translators for other Languages
@subsection KUIT Format Strings

KUIT (KDE User Interface Text) is compatible with KDE 4 format strings,
while it also allows programmers to add semantic information to a format
string, through XML markup tags.  For example, if the first format
directive in a string is a filename, programmers could indicate that
with a @samp{filename} tag, like @samp{<filename>%1</filename>}.

KUIT format strings are described in
@uref{http://api.kde.org/frameworks-api/frameworks5-apidocs/ki18n/html/prg_guide.html#kuit_markup}.

@node boost-format, lua-format, kde-kuit-format, Translators for other Languages
@subsection Boost Format Strings

Boost format strings are described in the documentation of the
@code{boost::format} class, at
@uref{http://www.boost.org/libs/format/doc/format.html}.
In summary, a directive has either the same syntax as in a C format string,
such as @samp{%1$+5d}, or may be surrounded by vertical bars, such as
@samp{%|1$+5d|} or @samp{%|1$+5|}, or consists of just an argument number
between percent signs, such as @samp{%1%}.

@node lua-format, javascript-format, boost-format, Translators for other Languages
@subsection Lua Format Strings

Lua format strings are described in the Lua reference manual, section @w{String Manipulation},
@uref{http://www.lua.org/manual/5.1/manual.html#pdf-string.format}.

@node javascript-format,  , lua-format, Translators for other Languages
@subsection JavaScript Format Strings

Although JavaScript specification itself does not define any format
strings, many JavaScript implementations provide printf-like
functions.  @code{xgettext} understands a set of common format strings
used in popular JavaScript implementations including Gjs, Seed, and
Node.JS.  In such a format string, a directive starts with @samp{%}
and is finished by a specifier: @samp{%} denotes a literal percent
sign, @samp{c} denotes a character, @samp{s} denotes a string,
@samp{b}, @samp{d}, @samp{o}, @samp{x}, @samp{X} denote an integer,
@samp{f} denotes floating-point number, @samp{j} denotes a JSON
object.


@node Maintainers for other Languages, List of Programming Languages, Translators for other Languages, Programming Languages
@section The Maintainer's View

For the maintainer, the general procedure differs from the C language
case in two ways.

@itemize @bullet
@item
For those languages that don't use GNU gettext, the @file{intl/} directory
is not needed and can be omitted.  This means that the maintainer calls the
@code{gettextize} program without the @samp{--intl} option, and that he
invokes the @code{AM_GNU_GETTEXT} autoconf macro via
@samp{AM_GNU_GETTEXT([external])}.

@item
If only a single programming language is used, the @code{XGETTEXT_OPTIONS}
variable in @file{po/Makevars} (@pxref{po/Makevars}) should be adjusted to
match the @code{xgettext} options for that particular programming language.
If the package uses more than one programming language with @code{gettext}
support, it becomes necessary to change the POT file construction rule
in @file{po/Makefile.in.in}.  It is recommended to make one @code{xgettext}
invocation per programming language, each with the options appropriate for
that language, and to combine the resulting files using @code{msgcat}.
@end itemize

@node List of Programming Languages, List of Data Formats, Maintainers for other Languages, Programming Languages
@section Individual Programming Languages

@c Here is a list of programming languages, as used for Free Software projects
@c on SourceForge/Freshmeat, as of February 2002.  Those supported by gettext
@c are marked with a star.
@c   C                       3580     *
@c   Perl                    1911     *
@c   C++                     1379     *
@c   Java                    1200     *
@c   PHP                     1051     *
@c   Python                   613     *
@c   Unix Shell               357     *
@c   Tcl                      266     *
@c   SQL                      174
@c   JavaScript               118
@c   Assembly                 108
@c   Scheme                    51
@c   Ruby                      47
@c   Lisp                      45     *
@c   Objective C               39     *
@c   PL/SQL                    29
@c   Fortran                   25
@c   Ada                       24
@c   Delphi                    22
@c   Awk                       19     *
@c   Pascal                    19
@c   ML                        19
@c   Eiffel                    17
@c   Emacs-Lisp                14     *
@c   Zope                      14
@c   ASP                       12
@c   Forth                     12
@c   Cold Fusion               10
@c   Haskell                    9
@c   Visual Basic               9
@c   C#                         6     *
@c   Smalltalk                  6     *
@c   Basic                      5
@c   Erlang                     5
@c   Modula                     5
@c   Object Pascal              5     *
@c   Rexx                       5
@c   Dylan                      4
@c   Prolog                     4
@c   APL                        3
@c   PROGRESS                   2
@c   Euler                      1
@c   Euphoria                   1
@c   Pliant                     1
@c   Simula                     1
@c   XBasic                     1
@c   Logo                       0
@c   Other Scripting Engines   49
@c   Other                    116

@menu
* C::                           C, C++, Objective C
* sh::                          sh - Shell Script
* bash::                        bash - Bourne-Again Shell Script
* Python::                      Python
* Common Lisp::                 GNU clisp - Common Lisp
* clisp C::                     GNU clisp C sources
* Emacs Lisp::                  Emacs Lisp
* librep::                      librep
* Scheme::                      GNU guile - Scheme
* Smalltalk::                   GNU Smalltalk
* Java::                        Java
* C#::                          C#
* gawk::                        GNU awk
* Pascal::                      Pascal - Free Pascal Compiler
* wxWidgets::                   wxWidgets library
* YCP::                         YCP - YaST2 scripting language
* Tcl::                         Tcl - Tk's scripting language
* Perl::                        Perl
* PHP::                         PHP Hypertext Preprocessor
* Pike::                        Pike
* GCC-source::                  GNU Compiler Collection sources
* Lua::                         Lua
* JavaScript::                  JavaScript
* Vala::                        Vala
@end menu

@node C, sh, List of Programming Languages, List of Programming Languages
@subsection C, C++, Objective C
@cindex C and C-like languages

@table @asis
@item RPMs
gcc, gpp, gobjc, glibc, gettext

@item File extension
For C: @code{c}, @code{h}.
@*For C++: @code{C}, @code{c++}, @code{cc}, @code{cxx}, @code{cpp}, @code{hpp}.
@*For Objective C: @code{m}.

@item String syntax
@code{"abc"}

@item gettext shorthand
@code{_("abc")}

@item gettext/ngettext functions
@code{gettext}, @code{dgettext}, @code{dcgettext}, @code{ngettext},
@code{dngettext}, @code{dcngettext}

@item textdomain
@code{textdomain} function

@item bindtextdomain
@code{bindtextdomain} function

@item setlocale
Programmer must call @code{setlocale (LC_ALL, "")}

@item Prerequisite
@code{#include <libintl.h>}
@*@code{#include <locale.h>}
@*@code{#define _(string) gettext (string)}

@item Use or emulate GNU gettext
Use

@item Extractor
@code{xgettext -k_}

@item Formatting with positions
@code{fprintf "%2$d %1$d"}
@*In C++: @code{autosprintf "%2$d %1$d"}
(@pxref{Top, , Introduction, autosprintf, GNU autosprintf})

@item Portability
autoconf (gettext.m4) and #if ENABLE_NLS

@item po-mode marking
yes
@end table

The following examples are available in the @file{examples} directory:
@code{hello-c}, @code{hello-c-gnome}, @code{hello-c++}, @code{hello-c++-qt}, 
@code{hello-c++-kde}, @code{hello-c++-gnome}, @code{hello-c++-wxwidgets},
@code{hello-objc}, @code{hello-objc-gnustep}, @code{hello-objc-gnome}.

@node sh, bash, C, List of Programming Languages
@subsection sh - Shell Script
@cindex shell scripts

@table @asis
@item RPMs
bash, gettext

@item File extension
@code{sh}

@item String syntax
@code{"abc"}, @code{'abc'}, @code{abc}

@item gettext shorthand
@code{"`gettext \"abc\"`"}

@item gettext/ngettext functions
@pindex gettext
@pindex ngettext
@code{gettext}, @code{ngettext} programs
@*@code{eval_gettext}, @code{eval_ngettext} shell functions

@item textdomain
@vindex TEXTDOMAIN@r{, environment variable}
environment variable @code{TEXTDOMAIN}

@item bindtextdomain
@vindex TEXTDOMAINDIR@r{, environment variable}
environment variable @code{TEXTDOMAINDIR}

@item setlocale
automatic

@item Prerequisite
@code{. gettext.sh}

@item Use or emulate GNU gettext
use

@item Extractor
@code{xgettext}

@item Formatting with positions
---

@item Portability
fully portable

@item po-mode marking
---
@end table

An example is available in the @file{examples} directory: @code{hello-sh}.

@menu
* Preparing Shell Scripts::     Preparing Shell Scripts for Internationalization
* gettext.sh::                  Contents of @code{gettext.sh}
* gettext Invocation::          Invoking the @code{gettext} program
* ngettext Invocation::         Invoking the @code{ngettext} program
* envsubst Invocation::         Invoking the @code{envsubst} program
* eval_gettext Invocation::     Invoking the @code{eval_gettext} function
* eval_ngettext Invocation::    Invoking the @code{eval_ngettext} function
@end menu

@node Preparing Shell Scripts, gettext.sh, sh, sh
@subsubsection Preparing Shell Scripts for Internationalization
@cindex preparing shell scripts for translation

Preparing a shell script for internationalization is conceptually similar
to the steps described in @ref{Sources}.  The concrete steps for shell
scripts are as follows.

@enumerate
@item
Insert the line

@smallexample
. gettext.sh
@end smallexample

near the top of the script.  @code{gettext.sh} is a shell function library
that provides the functions
@code{eval_gettext} (see @ref{eval_gettext Invocation}) and
@code{eval_ngettext} (see @ref{eval_ngettext Invocation}).
You have to ensure that @code{gettext.sh} can be found in the @code{PATH}.

@item
Set and export the @code{TEXTDOMAIN} and @code{TEXTDOMAINDIR} environment
variables.  Usually @code{TEXTDOMAIN} is the package or program name, and
@code{TEXTDOMAINDIR} is the absolute pathname corresponding to
@code{$prefix/share/locale}, where @code{$prefix} is the installation location.

@smallexample
TEXTDOMAIN=@@PACKAGE@@
export TEXTDOMAIN
TEXTDOMAINDIR=@@LOCALEDIR@@
export TEXTDOMAINDIR
@end smallexample

@item
Prepare the strings for translation, as described in @ref{Preparing Strings}.

@item
Simplify translatable strings so that they don't contain command substitution
(@code{"`...`"} or @code{"$(...)"}), variable access with defaulting (like
@code{$@{@var{variable}-@var{default}@}}), access to positional arguments
(like @code{$0}, @code{$1}, ...) or highly volatile shell variables (like
@code{$?}). This can always be done through simple local code restructuring.
For example,

@smallexample
echo "Usage: $0 [OPTION] FILE..."
@end smallexample

becomes

@smallexample
program_name=$0
echo "Usage: $program_name [OPTION] FILE..."
@end smallexample

Similarly,

@smallexample
echo "Remaining files: `ls | wc -l`"
@end smallexample

becomes

@smallexample
filecount="`ls | wc -l`"
echo "Remaining files: $filecount"
@end smallexample

@item
For each translatable string, change the output command @samp{echo} or
@samp{$echo} to @samp{gettext} (if the string contains no references to
shell variables) or to @samp{eval_gettext} (if it refers to shell variables),
followed by a no-argument @samp{echo} command (to account for the terminating
newline). Similarly, for cases with plural handling, replace a conditional
@samp{echo} command with an invocation of @samp{ngettext} or
@samp{eval_ngettext}, followed by a no-argument @samp{echo} command.

When doing this, you also need to add an extra backslash before the dollar
sign in references to shell variables, so that the @samp{eval_gettext}
function receives the translatable string before the variable values are
substituted into it. For example,

@smallexample
echo "Remaining files: $filecount"
@end smallexample

becomes

@smallexample
eval_gettext "Remaining files: \$filecount"; echo
@end smallexample

If the output command is not @samp{echo}, you can make it use @samp{echo}
nevertheless, through the use of backquotes. However, note that inside
backquotes, backslashes must be doubled to be effective (because the
backquoting eats one level of backslashes). For example, assuming that
@samp{error} is a shell function that signals an error,

@smallexample
error "file not found: $filename"
@end smallexample

is first transformed into

@smallexample
error "`echo \"file not found: \$filename\"`"
@end smallexample

which then becomes

@smallexample
error "`eval_gettext \"file not found: \\\$filename\"`"
@end smallexample
@end enumerate

@node gettext.sh, gettext Invocation, Preparing Shell Scripts, sh
@subsubsection Contents of @code{gettext.sh}

@code{gettext.sh}, contained in the run-time package of GNU gettext, provides
the following:

@itemize @bullet
@item $echo
The variable @code{echo} is set to a command that outputs its first argument
and a newline, without interpreting backslashes in the argument string.

@item eval_gettext
See @ref{eval_gettext Invocation}.

@item eval_ngettext
See @ref{eval_ngettext Invocation}.
@end itemize

@node gettext Invocation, ngettext Invocation, gettext.sh, sh
@subsubsection Invoking the @code{gettext} program

@include rt-gettext.texi

Note: @code{xgettext} supports only the one-argument form of the
@code{gettext} invocation, where no options are present and the
@var{textdomain} is implicit, from the environment.

@node ngettext Invocation, envsubst Invocation, gettext Invocation, sh
@subsubsection Invoking the @code{ngettext} program

@include rt-ngettext.texi

Note: @code{xgettext} supports only the three-arguments form of the
@code{ngettext} invocation, where no options are present and the
@var{textdomain} is implicit, from the environment.

@node envsubst Invocation, eval_gettext Invocation, ngettext Invocation, sh
@subsubsection Invoking the @code{envsubst} program

@include rt-envsubst.texi

@node eval_gettext Invocation, eval_ngettext Invocation, envsubst Invocation, sh
@subsubsection Invoking the @code{eval_gettext} function

@cindex @code{eval_gettext} function, usage
@example
eval_gettext @var{msgid}
@end example

@cindex lookup message translation
This function outputs the native language translation of a textual message,
performing dollar-substitution on the result.  Note that only shell variables
mentioned in @var{msgid} will be dollar-substituted in the result.

@node eval_ngettext Invocation,  , eval_gettext Invocation, sh
@subsubsection Invoking the @code{eval_ngettext} function

@cindex @code{eval_ngettext} function, usage
@example
eval_ngettext @var{msgid} @var{msgid-plural} @var{count}
@end example

@cindex lookup plural message translation
This function outputs the native language translation of a textual message
whose grammatical form depends on a number, performing dollar-substitution
on the result.  Note that only shell variables mentioned in @var{msgid} or
@var{msgid-plural} will be dollar-substituted in the result.

@node bash, Python, sh, List of Programming Languages
@subsection bash - Bourne-Again Shell Script
@cindex bash

GNU @code{bash} 2.0 or newer has a special shorthand for translating a
string and substituting variable values in it: @code{$"msgid"}.  But
the use of this construct is @strong{discouraged}, due to the security
holes it opens and due to its portability problems.

The security holes of @code{$"..."} come from the fact that after looking up
the translation of the string, @code{bash} processes it like it processes
any double-quoted string: dollar and backquote processing, like @samp{eval}
does.

@enumerate
@item
In a locale whose encoding is one of BIG5, BIG5-HKSCS, GBK, GB18030, SHIFT_JIS,
JOHAB, some double-byte characters have a second byte whose value is
@code{0x60}.  For example, the byte sequence @code{\xe0\x60} is a single
character in these locales.  Many versions of @code{bash} (all versions
up to bash-2.05, and newer versions on platforms without @code{mbsrtowcs()}
function) don't know about character boundaries and see a backquote character
where there is only a particular Chinese character.  Thus it can start
executing part of the translation as a command list.  This situation can occur
even without the translator being aware of it: if the translator provides
translations in the UTF-8 encoding, it is the @code{gettext()} function which
will, during its conversion from the translator's encoding to the user's
locale's encoding, produce the dangerous @code{\x60} bytes.

@item
A translator could - voluntarily or inadvertently - use backquotes
@code{"`...`"} or dollar-parentheses @code{"$(...)"} in her translations.
The enclosed strings would be executed as command lists by the shell.
@end enumerate

The portability problem is that @code{bash} must be built with
internationalization support; this is normally not the case on systems
that don't have the @code{gettext()} function in libc.

@node Python, Common Lisp, bash, List of Programming Languages
@subsection Python
@cindex Python

@table @asis
@item RPMs
python

@item File extension
@code{py}

@item String syntax
@code{'abc'}, @code{u'abc'}, @code{r'abc'}, @code{ur'abc'},
@*@code{"abc"}, @code{u"abc"}, @code{r"abc"}, @code{ur"abc"},
@*@code{'''abc'''}, @code{u'''abc'''}, @code{r'''abc'''}, @code{ur'''abc'''},
@*@code{"""abc"""}, @code{u"""abc"""}, @code{r"""abc"""}, @code{ur"""abc"""}

@item gettext shorthand
@code{_('abc')} etc.

@item gettext/ngettext functions
@code{gettext.gettext}, @code{gettext.dgettext},
@code{gettext.ngettext}, @code{gettext.dngettext},
also @code{ugettext}, @code{ungettext}

@item textdomain
@code{gettext.textdomain} function, or
@code{gettext.install(@var{domain})} function

@item bindtextdomain
@code{gettext.bindtextdomain} function, or
@code{gettext.install(@var{domain},@var{localedir})} function

@item setlocale
not used by the gettext emulation

@item Prerequisite
@code{import gettext}

@item Use or emulate GNU gettext
emulate

@item Extractor
@code{xgettext}

@item Formatting with positions
@code{'...%(ident)d...' % @{ 'ident': value @}}

@item Portability
fully portable

@item po-mode marking
---
@end table

An example is available in the @file{examples} directory: @code{hello-python}.

A note about format strings: Python supports format strings with unnamed
arguments, such as @code{'...%d...'}, and format strings with named arguments,
such as @code{'...%(ident)d...'}.  The latter are preferable for
internationalized programs, for two reasons:

@itemize @bullet
@item
When a format string takes more than one argument, the translator can provide
a translation that uses the arguments in a different order, if the format
string uses named arguments.  For example, the translator can reformulate
@smallexample
"'%(volume)s' has only %(freespace)d bytes free."
@end smallexample
@noindent
to
@smallexample
"Only %(freespace)d bytes free on '%(volume)s'."
@end smallexample
@noindent
Additionally, the identifiers also provide some context to the translator.

@item
In the context of plural forms, the format string used for the singular form
does not use the numeric argument in many languages.  Even in English, one
prefers to write @code{"one hour"} instead of @code{"1 hour"}.  Omitting
individual arguments from format strings like this is only possible with
the named argument syntax.  (With unnamed arguments, Python -- unlike C --
verifies that the format string uses all supplied arguments.)
@end itemize

@node Common Lisp, clisp C, Python, List of Programming Languages
@subsection GNU clisp - Common Lisp
@cindex Common Lisp
@cindex Lisp
@cindex clisp

@table @asis
@item RPMs
clisp 2.28 or newer

@item File extension
@code{lisp}

@item String syntax
@code{"abc"}

@item gettext shorthand
@code{(_ "abc")}, @code{(ENGLISH "abc")}

@item gettext/ngettext functions
@code{i18n:gettext}, @code{i18n:ngettext}

@item textdomain
@code{i18n:textdomain}

@item bindtextdomain
@code{i18n:textdomaindir}

@item setlocale
automatic

@item Prerequisite
---

@item Use or emulate GNU gettext
use

@item Extractor
@code{xgettext -k_ -kENGLISH}

@item Formatting with positions
@code{format "~1@@*~D ~0@@*~D"}

@item Portability
On platforms without gettext, no translation.

@item po-mode marking
---
@end table

An example is available in the @file{examples} directory: @code{hello-clisp}.

@node clisp C, Emacs Lisp, Common Lisp, List of Programming Languages
@subsection GNU clisp C sources
@cindex clisp C sources

@table @asis
@item RPMs
clisp

@item File extension
@code{d}

@item String syntax
@code{"abc"}

@item gettext shorthand
@code{ENGLISH ? "abc" : ""}
@*@code{GETTEXT("abc")}
@*@code{GETTEXTL("abc")}

@item gettext/ngettext functions
@code{clgettext}, @code{clgettextl}

@item textdomain
---

@item bindtextdomain
---

@item setlocale
automatic

@item Prerequisite
@code{#include "lispbibl.c"}

@item Use or emulate GNU gettext
use

@item Extractor
@code{clisp-xgettext}

@item Formatting with positions
@code{fprintf "%2$d %1$d"}

@item Portability
On platforms without gettext, no translation.

@item po-mode marking
---
@end table

@node Emacs Lisp, librep, clisp C, List of Programming Languages
@subsection Emacs Lisp
@cindex Emacs Lisp

@table @asis
@item RPMs
emacs, xemacs

@item File extension
@code{el}

@item String syntax
@code{"abc"}

@item gettext shorthand
@code{(_"abc")}

@item gettext/ngettext functions
@code{gettext}, @code{dgettext} (xemacs only)

@item textdomain
@code{domain} special form (xemacs only)

@item bindtextdomain
@code{bind-text-domain} function (xemacs only)

@item setlocale
automatic

@item Prerequisite
---

@item Use or emulate GNU gettext
use

@item Extractor
@code{xgettext}

@item Formatting with positions
@code{format "%2$d %1$d"}

@item Portability
Only XEmacs.  Without @code{I18N3} defined at build time, no translation.

@item po-mode marking
---
@end table

@node librep, Scheme, Emacs Lisp, List of Programming Languages
@subsection librep
@cindex @code{librep} Lisp

@table @asis
@item RPMs
librep 0.15.3 or newer

@item File extension
@code{jl}

@item String syntax
@code{"abc"}

@item gettext shorthand
@code{(_"abc")}

@item gettext/ngettext functions
@code{gettext}

@item textdomain
@code{textdomain} function

@item bindtextdomain
@code{bindtextdomain} function

@item setlocale
---

@item Prerequisite
@code{(require 'rep.i18n.gettext)}

@item Use or emulate GNU gettext
use

@item Extractor
@code{xgettext}

@item Formatting with positions
@code{format "%2$d %1$d"}

@item Portability
On platforms without gettext, no translation.

@item po-mode marking
---
@end table

An example is available in the @file{examples} directory: @code{hello-librep}.

@node Scheme, Smalltalk, librep, List of Programming Languages
@subsection GNU guile - Scheme
@cindex Scheme
@cindex guile

@table @asis
@item RPMs
guile

@item File extension
@code{scm}

@item String syntax
@code{"abc"}

@item gettext shorthand
@code{(_ "abc")}, @code{_"abc"} (GIMP script-fu extension)

@item gettext/ngettext functions
@code{gettext}, @code{ngettext}

@item textdomain
@code{textdomain}

@item bindtextdomain
@code{bindtextdomain}

@item setlocale
@code{(catch #t (lambda () (setlocale LC_ALL "")) (lambda args #f))}

@item Prerequisite
@code{(use-modules (ice-9 format))}

@item Use or emulate GNU gettext
use

@item Extractor
@code{xgettext -k_}

@item Formatting with positions
@c @code{format "~1@@*~D ~0@@*~D~2@@*"}, requires @code{(use-modules (ice-9 format))}
@c not yet supported
---

@item Portability
On platforms without gettext, no translation.

@item po-mode marking
---
@end table

An example is available in the @file{examples} directory: @code{hello-guile}.

@node Smalltalk, Java, Scheme, List of Programming Languages
@subsection GNU Smalltalk
@cindex Smalltalk

@table @asis
@item RPMs
smalltalk

@item File extension
@code{st}

@item String syntax
@code{'abc'}

@item gettext shorthand
@code{NLS ? 'abc'}

@item gettext/ngettext functions
@code{LcMessagesDomain>>#at:}, @code{LcMessagesDomain>>#at:plural:with:}

@item textdomain
@code{LcMessages>>#domain:localeDirectory:} (returns a @code{LcMessagesDomain}
object).@*
Example: @code{I18N Locale default messages domain: 'gettext' localeDirectory: /usr/local/share/locale'}

@item bindtextdomain
@code{LcMessages>>#domain:localeDirectory:}, see above.

@item setlocale
Automatic if you use @code{I18N Locale default}.

@item Prerequisite
@code{PackageLoader fileInPackage: 'I18N'!}

@item Use or emulate GNU gettext
emulate

@item Extractor
@code{xgettext}

@item Formatting with positions
@code{'%1 %2' bindWith: 'Hello' with: 'world'}

@item Portability
fully portable

@item po-mode marking
---
@end table

An example is available in the @file{examples} directory:
@code{hello-smalltalk}.

@node Java, C#, Smalltalk, List of Programming Languages
@subsection Java
@cindex Java

@table @asis
@item RPMs
java, java2

@item File extension
@code{java}

@item String syntax
"abc"

@item gettext shorthand
_("abc")

@item gettext/ngettext functions
@code{GettextResource.gettext}, @code{GettextResource.ngettext},
@code{GettextResource.pgettext}, @code{GettextResource.npgettext}

@item textdomain
---, use @code{ResourceBundle.getResource} instead

@item bindtextdomain
---, use CLASSPATH instead

@item setlocale
automatic

@item Prerequisite
---

@item Use or emulate GNU gettext
---, uses a Java specific message catalog format

@item Extractor
@code{xgettext -k_}

@item Formatting with positions
@code{MessageFormat.format "@{1,number@} @{0,number@}"}

@item Portability
fully portable

@item po-mode marking
---
@end table

Before marking strings as internationalizable, uses of the string
concatenation operator need to be converted to @code{MessageFormat}
applications.  For example, @code{"file "+filename+" not found"} becomes
@code{MessageFormat.format("file @{0@} not found", new Object[] @{ filename @})}.
Only after this is done, can the strings be marked and extracted.

GNU gettext uses the native Java internationalization mechanism, namely
@code{ResourceBundle}s.  There are two formats of @code{ResourceBundle}s:
@code{.properties} files and @code{.class} files.  The @code{.properties}
format is a text file which the translators can directly edit, like PO
files, but which doesn't support plural forms.  Whereas the @code{.class}
format is compiled from @code{.java} source code and can support plural
forms (provided it is accessed through an appropriate API, see below).

To convert a PO file to a @code{.properties} file, the @code{msgcat}
program can be used with the option @code{--properties-output}.  To convert
a @code{.properties} file back to a PO file, the @code{msgcat} program
can be used with the option @code{--properties-input}.  All the tools
that manipulate PO files can work with @code{.properties} files as well,
if given the @code{--properties-input} and/or @code{--properties-output}
option.

To convert a PO file to a ResourceBundle class, the @code{msgfmt} program
can be used with the option @code{--java} or @code{--java2}.  To convert a
ResourceBundle back to a PO file, the @code{msgunfmt} program can be used
with the option @code{--java}.

Two different programmatic APIs can be used to access ResourceBundles.
Note that both APIs work with all kinds of ResourceBundles, whether
GNU gettext generated classes, or other @code{.class} or @code{.properties}
files.

@enumerate
@item
The @code{java.util.ResourceBundle} API.

In particular, its @code{getString} function returns a string translation.
Note that a missing translation yields a @code{MissingResourceException}.

This has the advantage of being the standard API.  And it does not require
any additional libraries, only the @code{msgcat} generated @code{.properties}
files or the @code{msgfmt} generated @code{.class} files.  But it cannot do
plural handling, even if the resource was generated by @code{msgfmt} from
a PO file with plural handling.

@item
The @code{gnu.gettext.GettextResource} API.

Reference documentation in Javadoc 1.1 style format is in the
@uref{javadoc2/index.html,javadoc2 directory}.

Its @code{gettext} function returns a string translation.  Note that when
a translation is missing, the @var{msgid} argument is returned unchanged.

This has the advantage of having the @code{ngettext} function for plural
handling and the @code{pgettext} and @code{npgettext} for strings constraint
to a particular context.

@cindex @code{libintl} for Java
To use this API, one needs the @code{libintl.jar} file which is part of
the GNU gettext package and distributed under the LGPL.
@end enumerate

Four examples, using the second API, are available in the @file{examples}
directory: @code{hello-java}, @code{hello-java-awt}, @code{hello-java-swing},
@code{hello-java-qtjambi}.

Now, to make use of the API and define a shorthand for @samp{getString},
there are three idioms that you can choose from:

@itemize @bullet
@item
(This one assumes Java 1.5 or newer.)
In a unique class of your project, say @samp{Util}, define a static variable
holding the @code{ResourceBundle} instance and the shorthand:

@smallexample
private static ResourceBundle myResources =
  ResourceBundle.getBundle("domain-name");
public static String _(String s) @{
  return myResources.getString(s);
@}
@end smallexample

All classes containing internationalized strings then contain

@smallexample
import static Util._;
@end smallexample

@noindent
and the shorthand is used like this:

@smallexample
System.out.println(_("Operation completed."));
@end smallexample

@item
In a unique class of your project, say @samp{Util}, define a static variable
holding the @code{ResourceBundle} instance:

@smallexample
public static ResourceBundle myResources =
  ResourceBundle.getBundle("domain-name");
@end smallexample

All classes containing internationalized strings then contain

@smallexample
private static ResourceBundle res = Util.myResources;
private static String _(String s) @{ return res.getString(s); @}
@end smallexample

@noindent
and the shorthand is used like this:

@smallexample
System.out.println(_("Operation completed."));
@end smallexample

@item
You add a class with a very short name, say @samp{S}, containing just the
definition of the resource bundle and of the shorthand:

@smallexample
public class S @{
  public static ResourceBundle myResources =
    ResourceBundle.getBundle("domain-name");
  public static String _(String s) @{
    return myResources.getString(s);
  @}
@}
@end smallexample

@noindent
and the shorthand is used like this:

@smallexample
System.out.println(S._("Operation completed."));
@end smallexample
@end itemize

Which of the three idioms you choose, will depend on whether your project
requires portability to Java versions prior to Java 1.5 and, if so, whether
copying two lines of codes into every class is more acceptable in your project
than a class with a single-letter name.

@node C#, gawk, Java, List of Programming Languages
@subsection C#
@cindex C#

@table @asis
@item RPMs
pnet, pnetlib 0.6.2 or newer, or mono 0.29 or newer

@item File extension
@code{cs}

@item String syntax
@code{"abc"}, @code{@@"abc"}

@item gettext shorthand
_("abc")

@item gettext/ngettext functions
@code{GettextResourceManager.GetString},
@code{GettextResourceManager.GetPluralString}
@code{GettextResourceManager.GetParticularString}
@code{GettextResourceManager.GetParticularPluralString}

@item textdomain
@code{new GettextResourceManager(domain)}

@item bindtextdomain
---, compiled message catalogs are located in subdirectories of the directory
containing the executable

@item setlocale
automatic

@item Prerequisite
---

@item Use or emulate GNU gettext
---, uses a C# specific message catalog format

@item Extractor
@code{xgettext -k_}

@item Formatting with positions
@code{String.Format "@{1@} @{0@}"}

@item Portability
fully portable

@item po-mode marking
---
@end table

Before marking strings as internationalizable, uses of the string
concatenation operator need to be converted to @code{String.Format}
invocations.  For example, @code{"file "+filename+" not found"} becomes
@code{String.Format("file @{0@} not found", filename)}.
Only after this is done, can the strings be marked and extracted.

GNU gettext uses the native C#/.NET internationalization mechanism, namely
the classes @code{ResourceManager} and @code{ResourceSet}.  Applications
use the @code{ResourceManager} methods to retrieve the native language
translation of strings.  An instance of @code{ResourceSet} is the in-memory
representation of a message catalog file.  The @code{ResourceManager} loads
and accesses @code{ResourceSet} instances as needed to look up the
translations.

There are two formats of @code{ResourceSet}s that can be directly loaded by
the C# runtime: @code{.resources} files and @code{.dll} files.

@itemize @bullet
@item
The @code{.resources} format is a binary file usually generated through the
@code{resgen} or @code{monoresgen} utility, but which doesn't support plural
forms.  @code{.resources} files can also be embedded in .NET @code{.exe} files.
This only affects whether a file system access is performed to load the message
catalog; it doesn't affect the contents of the message catalog.

@item
On the other hand, the @code{.dll} format is a binary file that is compiled
from @code{.cs} source code and can support plural forms (provided it is
accessed through the GNU gettext API, see below).
@end itemize

Note that these .NET @code{.dll} and @code{.exe} files are not tied to a
particular platform; their file format and GNU gettext for C# can be used
on any platform.

To convert a PO file to a @code{.resources} file, the @code{msgfmt} program
can be used with the option @samp{--csharp-resources}.  To convert a
@code{.resources} file back to a PO file, the @code{msgunfmt} program can be
used with the option @samp{--csharp-resources}.  You can also, in some cases,
use the @code{resgen} program (from the @code{pnet} package) or the
@code{monoresgen} program (from the @code{mono}/@code{mcs} package).  These
programs can also convert a @code{.resources} file back to a PO file.  But
beware: as of this writing (January 2004), the @code{monoresgen} converter is
quite buggy and the @code{resgen} converter ignores the encoding of the PO
files.

To convert a PO file to a @code{.dll} file, the @code{msgfmt} program can be
used with the option @code{--csharp}.  The result will be a @code{.dll} file
containing a subclass of @code{GettextResourceSet}, which itself is a subclass
of @code{ResourceSet}.  To convert a @code{.dll} file containing a
@code{GettextResourceSet} subclass back to a PO file, the @code{msgunfmt}
program can be used with the option @code{--csharp}.

The advantages of the @code{.dll} format over the @code{.resources} format
are:

@enumerate
@item
Freedom to localize: Users can add their own translations to an application
after it has been built and distributed.  Whereas when the programmer uses
a @code{ResourceManager} constructor provided by the system, the set of
@code{.resources} files for an application must be specified when the
application is built and cannot be extended afterwards.
@c If this were the only issue with the @code{.resources} format, one could
@c use the @code{ResourceManager.CreateFileBasedResourceManager} function.

@item
Plural handling: A message catalog in @code{.dll} format supports the plural
handling function @code{GetPluralString}.  Whereas @code{.resources} files can
only contain data and only support lookups that depend on a single string.

@item
Context handling: A message catalog in @code{.dll} format supports the
query-with-context functions @code{GetParticularString} and
@code{GetParticularPluralString}.  Whereas @code{.resources} files can
only contain data and only support lookups that depend on a single string.

@item
The @code{GettextResourceManager} that loads the message catalogs in
@code{.dll} format also provides for inheritance on a per-message basis.
For example, in Austrian (@code{de_AT}) locale, translations from the German
(@code{de}) message catalog will be used for messages not found in the
Austrian message catalog.  This has the consequence that the Austrian
translators need only translate those few messages for which the translation
into Austrian differs from the German one.  Whereas when working with
@code{.resources} files, each message catalog must provide the translations
of all messages by itself.

@item
The @code{GettextResourceManager} that loads the message catalogs in
@code{.dll} format also provides for a fallback: The English @var{msgid} is
returned when no translation can be found.  Whereas when working with
@code{.resources} files, a language-neutral @code{.resources} file must
explicitly be provided as a fallback.
@end enumerate

On the side of the programmatic APIs, the programmer can use either the
standard @code{ResourceManager} API and the GNU @code{GettextResourceManager}
API.  The latter is an extension of the former, because
@code{GettextResourceManager} is a subclass of @code{ResourceManager}.

@enumerate
@item
The @code{System.Resources.ResourceManager} API.

This API works with resources in @code{.resources} format.

The creation of the @code{ResourceManager} is done through
@smallexample
  new ResourceManager(domainname, Assembly.GetExecutingAssembly())
@end smallexample
@noindent

The @code{GetString} function returns a string's translation.  Note that this
function returns null when a translation is missing (i.e.@: not even found in
the fallback resource file).

@item
The @code{GNU.Gettext.GettextResourceManager} API.

This API works with resources in @code{.dll} format.

Reference documentation is in the
@uref{csharpdoc/index.html,csharpdoc directory}.

The creation of the @code{ResourceManager} is done through
@smallexample
  new GettextResourceManager(domainname)
@end smallexample

The @code{GetString} function returns a string's translation.  Note that when
a translation is missing, the @var{msgid} argument is returned unchanged.

The @code{GetPluralString} function returns a string translation with plural
handling, like the @code{ngettext} function in C.

The @code{GetParticularString} function returns a string's translation,
specific to a particular context, like the @code{pgettext} function in C.
Note that when a translation is missing, the @var{msgid} argument is returned
unchanged.

The @code{GetParticularPluralString} function returns a string translation,
specific to a particular context, with plural handling, like the
@code{npgettext} function in C.

@cindex @code{libintl} for C#
To use this API, one needs the @code{GNU.Gettext.dll} file which is part of
the GNU gettext package and distributed under the LGPL.
@end enumerate

You can also mix both approaches: use the
@code{GNU.Gettext.GettextResourceManager} constructor, but otherwise use
only the @code{ResourceManager} type and only the @code{GetString} method.
This is appropriate when you want to profit from the tools for PO files,
but don't want to change an existing source code that uses
@code{ResourceManager} and don't (yet) need the @code{GetPluralString} method.

Two examples, using the second API, are available in the @file{examples}
directory: @code{hello-csharp}, @code{hello-csharp-forms}.

Now, to make use of the API and define a shorthand for @samp{GetString},
there are two idioms that you can choose from:

@itemize @bullet
@item
In a unique class of your project, say @samp{Util}, define a static variable
holding the @code{ResourceManager} instance:

@smallexample
public static GettextResourceManager MyResourceManager =
  new GettextResourceManager("domain-name");
@end smallexample

All classes containing internationalized strings then contain

@smallexample
private static GettextResourceManager Res = Util.MyResourceManager;
private static String _(String s) @{ return Res.GetString(s); @}
@end smallexample

@noindent
and the shorthand is used like this:

@smallexample
Console.WriteLine(_("Operation completed."));
@end smallexample

@item
You add a class with a very short name, say @samp{S}, containing just the
definition of the resource manager and of the shorthand:

@smallexample
public class S @{
  public static GettextResourceManager MyResourceManager =
    new GettextResourceManager("domain-name");
  public static String _(String s) @{
     return MyResourceManager.GetString(s);
  @}
@}
@end smallexample

@noindent
and the shorthand is used like this:

@smallexample
Console.WriteLine(S._("Operation completed."));
@end smallexample
@end itemize

Which of the two idioms you choose, will depend on whether copying two lines
of codes into every class is more acceptable in your project than a class
with a single-letter name.

@node gawk, Pascal, C#, List of Programming Languages
@subsection GNU awk
@cindex awk
@cindex gawk

@table @asis
@item RPMs
gawk 3.1 or newer

@item File extension
@code{awk}, @code{gawk}, @code{twjr}.
The file extension @code{twjr} is used by TexiWeb Jr
(@uref{https://github.com/arnoldrobbins/texiwebjr}).

@item String syntax
@code{"abc"}

@item gettext shorthand
@code{_"abc"}

@item gettext/ngettext functions
@code{dcgettext}, missing @code{dcngettext} in gawk-3.1.0

@item textdomain
@code{TEXTDOMAIN} variable

@item bindtextdomain
@code{bindtextdomain} function

@item setlocale
automatic, but missing @code{setlocale (LC_MESSAGES, "")} in gawk-3.1.0

@item Prerequisite
---

@item Use or emulate GNU gettext
use

@item Extractor
@code{xgettext}

@item Formatting with positions
@code{printf "%2$d %1$d"} (GNU awk only)

@item Portability
On platforms without gettext, no translation.  On non-GNU awks, you must
define @code{dcgettext}, @code{dcngettext} and @code{bindtextdomain}
yourself.

@item po-mode marking
---
@end table

An example is available in the @file{examples} directory: @code{hello-gawk}.

@node Pascal, wxWidgets, gawk, List of Programming Languages
@subsection Pascal - Free Pascal Compiler
@cindex Pascal
@cindex Free Pascal
@cindex Object Pascal

@table @asis
@item RPMs
fpk

@item File extension
@code{pp}, @code{pas}

@item String syntax
@code{'abc'}

@item gettext shorthand
automatic

@item gettext/ngettext functions
---, use @code{ResourceString} data type instead

@item textdomain
---, use @code{TranslateResourceStrings} function instead

@item bindtextdomain
---, use @code{TranslateResourceStrings} function instead

@item setlocale
automatic, but uses only LANG, not LC_MESSAGES or LC_ALL

@item Prerequisite
@code{@{$mode delphi@}} or @code{@{$mode objfpc@}}@*@code{uses gettext;}

@item Use or emulate GNU gettext
emulate partially

@item Extractor
@code{ppc386} followed by @code{xgettext} or @code{rstconv}

@item Formatting with positions
@code{uses sysutils;}@*@code{format "%1:d %0:d"}

@item Portability
?

@item po-mode marking
---
@end table

The Pascal compiler has special support for the @code{ResourceString} data
type.  It generates a @code{.rst} file.  This is then converted to a
@code{.pot} file by use of @code{xgettext} or @code{rstconv}.  At runtime,
a @code{.mo} file corresponding to translations of this @code{.pot} file
can be loaded using the @code{TranslateResourceStrings} function in the
@code{gettext} unit.

An example is available in the @file{examples} directory: @code{hello-pascal}.

@node wxWidgets, YCP, Pascal, List of Programming Languages
@subsection wxWidgets library
@cindex @code{wxWidgets} library

@table @asis
@item RPMs
wxGTK, gettext

@item File extension
@code{cpp}

@item String syntax
@code{"abc"}

@item gettext shorthand
@code{_("abc")}

@item gettext/ngettext functions
@code{wxLocale::GetString}, @code{wxGetTranslation}

@item textdomain
@code{wxLocale::AddCatalog}

@item bindtextdomain
@code{wxLocale::AddCatalogLookupPathPrefix}

@item setlocale
@code{wxLocale::Init}, @code{wxSetLocale}

@item Prerequisite
@code{#include <wx/intl.h>}

@item Use or emulate GNU gettext
emulate, see @code{include/wx/intl.h} and @code{src/common/intl.cpp}

@item Extractor
@code{xgettext}

@item Formatting with positions
wxString::Format supports positions if and only if the system has
@code{wprintf()}, @code{vswprintf()} functions and they support positions
according to POSIX.

@item Portability
fully portable

@item po-mode marking
yes
@end table

@node YCP, Tcl, wxWidgets, List of Programming Languages
@subsection YCP - YaST2 scripting language
@cindex YCP
@cindex YaST2 scripting language

@table @asis
@item RPMs
libycp, libycp-devel, yast2-core, yast2-core-devel

@item File extension
@code{ycp}

@item String syntax
@code{"abc"}

@item gettext shorthand
@code{_("abc")}

@item gettext/ngettext functions
@code{_()} with 1 or 3 arguments

@item textdomain
@code{textdomain} statement

@item bindtextdomain
---

@item setlocale
---

@item Prerequisite
---

@item Use or emulate GNU gettext
use

@item Extractor
@code{xgettext}

@item Formatting with positions
@code{sformat "%2 %1"}

@item Portability
fully portable

@item po-mode marking
---
@end table

An example is available in the @file{examples} directory: @code{hello-ycp}.

@node Tcl, Perl, YCP, List of Programming Languages
@subsection Tcl - Tk's scripting language
@cindex Tcl
@cindex Tk's scripting language

@table @asis
@item RPMs
tcl

@item File extension
@code{tcl}

@item String syntax
@code{"abc"}

@item gettext shorthand
@code{[_ "abc"]}

@item gettext/ngettext functions
@code{::msgcat::mc}

@item textdomain
---

@item bindtextdomain
---, use @code{::msgcat::mcload} instead

@item setlocale
automatic, uses LANG, but ignores LC_MESSAGES and LC_ALL

@item Prerequisite
@code{package require msgcat}
@*@code{proc _ @{s@} @{return [::msgcat::mc $s]@}}

@item Use or emulate GNU gettext
---, uses a Tcl specific message catalog format

@item Extractor
@code{xgettext -k_}

@item Formatting with positions
@code{format "%2\$d %1\$d"}

@item Portability
fully portable

@item po-mode marking
---
@end table

Two examples are available in the @file{examples} directory:
@code{hello-tcl}, @code{hello-tcl-tk}.

Before marking strings as internationalizable, substitutions of variables
into the string need to be converted to @code{format} applications.  For
example, @code{"file $filename not found"} becomes
@code{[format "file %s not found" $filename]}.
Only after this is done, can the strings be marked and extracted.
After marking, this example becomes
@code{[format [_ "file %s not found"] $filename]} or
@code{[msgcat::mc "file %s not found" $filename]}.  Note that the
@code{msgcat::mc} function implicitly calls @code{format} when more than one
argument is given.

@node Perl, PHP, Tcl, List of Programming Languages
@subsection Perl
@cindex Perl

@table @asis
@item RPMs
perl

@item File extension
@code{pl}, @code{PL}, @code{pm}, @code{perl}, @code{cgi}

@item String syntax
@itemize @bullet

@item @code{"abc"}

@item @code{'abc'}

@item @code{qq (abc)}

@item @code{q (abc)}

@item @code{qr /abc/}

@item @code{qx (/bin/date)}

@item @code{/pattern match/}

@item @code{?pattern match?}

@item @code{s/substitution/operators/}

@item @code{$tied_hash@{"message"@}}

@item @code{$tied_hash_reference->@{"message"@}}

@item etc., issue the command @samp{man perlsyn} for details

@end itemize

@item gettext shorthand
@code{__} (double underscore)

@item gettext/ngettext functions
@code{gettext}, @code{dgettext}, @code{dcgettext}, @code{ngettext},
@code{dngettext}, @code{dcngettext}

@item textdomain
@code{textdomain} function

@item bindtextdomain
@code{bindtextdomain} function

@item bind_textdomain_codeset 
@code{bind_textdomain_codeset} function

@item setlocale
Use @code{setlocale (LC_ALL, "");}

@item Prerequisite
@code{use POSIX;}
@*@code{use Locale::TextDomain;} (included in the package libintl-perl
which is available on the Comprehensive Perl Archive Network CPAN,
http://www.cpan.org/).

@item Use or emulate GNU gettext
platform dependent: gettext_pp emulates, gettext_xs uses GNU gettext

@item Extractor
@code{xgettext -k__ -k\$__ -k%__ -k__x -k__n:1,2 -k__nx:1,2 -k__xn:1,2 -kN__ -k}

@item Formatting with positions
Both kinds of format strings support formatting with positions.
@*@code{printf "%2\$d %1\$d", ...} (requires Perl 5.8.0 or newer)
@*@code{__expand("[new] replaces [old]", old => $oldvalue, new => $newvalue)}

@item Portability
The @code{libintl-perl} package is platform independent but is not
part of the Perl core.  The programmer is responsible for
providing a dummy implementation of the required functions if the 
package is not installed on the target system.

@item po-mode marking
---

@item Documentation
Included in @code{libintl-perl}, available on CPAN
(http://www.cpan.org/).

@end table

An example is available in the @file{examples} directory: @code{hello-perl}.

@cindex marking Perl sources

The @code{xgettext} parser backend for Perl differs significantly from
the parser backends for other programming languages, just as Perl
itself differs significantly from other programming languages.  The
Perl parser backend offers many more string marking facilities than
the other backends but it also has some Perl specific limitations, the
worst probably being its imperfectness.

@menu
* General Problems::            General Problems Parsing Perl Code
* Default Keywords::            Which Keywords Will xgettext Look For?
* Special Keywords::            How to Extract Hash Keys
* Quote-like Expressions::      What are Strings And Quote-like Expressions?
* Interpolation I::             Invalid String Interpolation
* Interpolation II::            Valid String Interpolation
* Parentheses::                 When To Use Parentheses
* Long Lines::                  How To Grok with Long Lines
* Perl Pitfalls::               Bugs, Pitfalls, and Things That Do Not Work
@end menu

@node General Problems, Default Keywords, Perl, Perl
@subsubsection General Problems Parsing Perl Code

It is often heard that only Perl can parse Perl.  This is not true.
Perl cannot be @emph{parsed} at all, it can only be @emph{executed}.
Perl has various built-in ambiguities that can only be resolved at runtime.

The following example may illustrate one common problem:

@example
print gettext "Hello World!";
@end example

Although this example looks like a bullet-proof case of a function
invocation, it is not:

@example
open gettext, ">testfile" or die;
print gettext "Hello world!"
@end example

In this context, the string @code{gettext} looks more like a
file handle.  But not necessarily:

@example
use Locale::Messages qw (:libintl_h);
open gettext ">testfile" or die;
print gettext "Hello world!";
@end example

Now, the file is probably syntactically incorrect, provided that the module
@code{Locale::Messages} found first in the Perl include path exports a
function @code{gettext}.  But what if the module
@code{Locale::Messages} really looks like this?

@example
use vars qw (*gettext);

1;
@end example

In this case, the string @code{gettext} will be interpreted as a file
handle again, and the above example will create a file @file{testfile}
and write the string ``Hello world!'' into it.  Even advanced
control flow analysis will not really help:

@example
if (0.5 < rand) @{
   eval "use Sane";
@} else @{
   eval "use InSane";
@}
print gettext "Hello world!";
@end example

If the module @code{Sane} exports a function @code{gettext} that does
what we expect, and the module @code{InSane} opens a file for writing
and associates the @emph{handle} @code{gettext} with this output
stream, we are clueless again about what will happen at runtime.  It is
completely unpredictable.  The truth is that Perl has so many ways to
fill its symbol table at runtime that it is impossible to interpret a
particular piece of code without executing it.

Of course, @code{xgettext} will not execute your Perl sources while
scanning for translatable strings, but rather use heuristics in order
to guess what you meant.

Another problem is the ambiguity of the slash and the question mark.
Their interpretation depends on the context:

@example
# A pattern match.
print "OK\n" if /foobar/;

# A division.
print 1 / 2;

# Another pattern match.
print "OK\n" if ?foobar?;

# Conditional.
print $x ? "foo" : "bar";
@end example

The slash may either act as the division operator or introduce a
pattern match, whereas the question mark may act as the ternary
conditional operator or as a pattern match, too.  Other programming
languages like @code{awk} present similar problems, but the consequences of a
misinterpretation are particularly nasty with Perl sources.  In @code{awk}
for instance, a statement can never exceed one line and the parser
can recover from a parsing error at the next newline and interpret
the rest of the input stream correctly.  Perl is different, as a
pattern match is terminated by the next appearance of the delimiter
(the slash or the question mark) in the input stream, regardless of
the semantic context.  If a slash is really a division sign but
mis-interpreted as a pattern match, the rest of the input file is most
probably parsed incorrectly.

There are certain cases, where the ambiguity cannot be resolved at all:

@example
$x = wantarray ? 1 : 0;
@end example

The Perl built-in function @code{wantarray} does not accept any arguments.
The Perl parser therefore knows that the question mark does not start
a regular expression but is the ternary conditional operator.

@example 
sub wantarrays @{@}
$x = wantarrays ? 1 : 0;
@end example

Now the situation is different.  The function @code{wantarrays} takes
a variable number of arguments (like any non-prototyped Perl function).
The question mark is now the delimiter of a pattern match, and hence
the piece of code does not compile.

@example
sub wantarrays() @{@}
$x = wantarrays ? 1 : 0;
@end example

Now the function is prototyped, Perl knows that it does not accept any
arguments, and the question mark is therefore interpreted as the
ternaray operator again.  But that unfortunately outsmarts @code{xgettext}.

The Perl parser in @code{xgettext} cannot know whether a function has
a prototype and what that prototype would look like.  It therefore makes
an educated guess.  If a function is known to be a Perl built-in and
this function does not accept any arguments, a following question mark
or slash is treated as an operator, otherwise as the delimiter of a
following regular expression.  The Perl built-ins that do not accept
arguments are @code{wantarray}, @code{fork}, @code{time}, @code{times},
@code{getlogin}, @code{getppid}, @code{getpwent}, @code{getgrent},
@code{gethostent}, @code{getnetent}, @code{getprotoent}, @code{getservent},
@code{setpwent}, @code{setgrent}, @code{endpwent}, @code{endgrent},
@code{endhostent}, @code{endnetent}, @code{endprotoent}, and
@code{endservent}.

If you find that @code{xgettext} fails to extract strings from
portions of your sources, you should therefore look out for slashes
and/or question marks preceding these sections.  You may have come
across a bug in @code{xgettext}'s Perl parser (and of course you
should report that bug).  In the meantime you should consider to
reformulate your code in a manner less challenging to @code{xgettext}.

In particular, if the parser is too dumb to see that a function
does not accept arguments, use parentheses:

@example
$x = somefunc() ? 1 : 0;
$y = (somefunc) ? 1 : 0;
@end example

In fact the Perl parser itself has similar problems and warns you
about such constructs.

@node Default Keywords, Special Keywords, General Problems, Perl
@subsubsection Which keywords will xgettext look for?
@cindex Perl default keywords

Unless you instruct @code{xgettext} otherwise by invoking it with one
of the options @code{--keyword} or @code{-k}, it will recognize the
following keywords in your Perl sources:

@itemize @bullet

@item @code{gettext}

@item @code{dgettext}

@item @code{dcgettext}

@item @code{ngettext:1,2}

The first (singular) and the second (plural) argument will be
extracted.

@item @code{dngettext:1,2}

The first (singular) and the second (plural) argument will be
extracted.

@item @code{dcngettext:1,2}

The first (singular) and the second (plural) argument will be
extracted.

@item @code{gettext_noop}

@item @code{%gettext}

The keys of lookups into the hash @code{%gettext} will be extracted.

@item @code{$gettext}

The keys of lookups into the hash reference @code{$gettext} will be extracted.

@end itemize

@node Special Keywords, Quote-like Expressions, Default Keywords, Perl
@subsubsection How to Extract Hash Keys
@cindex Perl special keywords for hash-lookups

Translating messages at runtime is normally performed by looking up the
original string in the translation database and returning the
translated version.  The ``natural'' Perl implementation is a hash
lookup, and, of course, @code{xgettext} supports such practice.

@example
print __"Hello world!";
print $__@{"Hello world!"@};
print $__->@{"Hello world!"@};
print $$__@{"Hello world!"@};
@end example  

The above four lines all do the same thing.  The Perl module 
@code{Locale::TextDomain} exports by default a hash @code{%__} that
is tied to the function @code{__()}.  It also exports a reference
@code{$__} to @code{%__}.

If an argument to the @code{xgettext} option @code{--keyword},
resp. @code{-k} starts with a percent sign, the rest of the keyword is
interpreted as the name of a hash.  If it starts with a dollar
sign, the rest of the keyword is interpreted as a reference to a
hash.

Note that you can omit the quotation marks (single or double) around
the hash key (almost) whenever Perl itself allows it:

@example
print $gettext@{Error@};
@end example

The exact rule is: You can omit the surrounding quotes, when the hash
key is a valid C (!) identifier, i.e.@: when it starts with an
underscore or an ASCII letter and is followed by an arbitrary number
of underscores, ASCII letters or digits.  Other Unicode characters
are @emph{not} allowed, regardless of the @code{use utf8} pragma.

@node Quote-like Expressions, Interpolation I, Special Keywords, Perl
@subsubsection What are Strings And Quote-like Expressions?
@cindex Perl quote-like expressions

Perl offers a plethora of different string constructs.  Those that can
be used either as arguments to functions or inside braces for hash
lookups are generally supported by @code{xgettext}.  

@itemize @bullet
@item @strong{double-quoted strings}
@*
@example
print gettext "Hello World!";
@end example

@item @strong{single-quoted strings}
@*
@example
print gettext 'Hello World!';
@end example

@item @strong{the operator qq}
@*
@example
print gettext qq |Hello World!|;
print gettext qq <E-mail: <guido\@@imperia.net>>;
@end example

The operator @code{qq} is fully supported.  You can use arbitrary
delimiters, including the four bracketing delimiters (round, angle,
square, curly) that nest.

@item @strong{the operator q}
@*
@example
print gettext q |Hello World!|;
print gettext q <E-mail: <guido@@imperia.net>>;
@end example

The operator @code{q} is fully supported.  You can use arbitrary
delimiters, including the four bracketing delimiters (round, angle,
square, curly) that nest.

@item @strong{the operator qx}
@*
@example
print gettext qx ;LANGUAGE=C /bin/date;
print gettext qx [/usr/bin/ls | grep '^[A-Z]*'];
@end example

The operator @code{qx} is fully supported.  You can use arbitrary
delimiters, including the four bracketing delimiters (round, angle,
square, curly) that nest.

The example is actually a useless use of @code{gettext}.  It will
invoke the @code{gettext} function on the output of the command
specified with the @code{qx} operator.  The feature was included
in order to make the interface consistent (the parser will extract
all strings and quote-like expressions).

@item @strong{here documents}
@*
@example
@group
print gettext <<'EOF';
program not found in $PATH
EOF

print ngettext <<EOF, <<"EOF";
one file deleted
EOF
several files deleted
EOF
@end group
@end example

Here-documents are recognized.  If the delimiter is enclosed in single
quotes, the string is not interpolated.  If it is enclosed in double
quotes or has no quotes at all, the string is interpolated.

Delimiters that start with a digit are not supported!

@end itemize

@node Interpolation I, Interpolation II, Quote-like Expressions, Perl
@subsubsection Invalid Uses Of String Interpolation
@cindex Perl invalid string interpolation

Perl is capable of interpolating variables into strings.  This offers
some nice features in localized programs but can also lead to
problems.

A common error is a construct like the following:

@example
print gettext "This is the program $0!\n";
@end example

Perl will interpolate at runtime the value of the variable @code{$0}
into the argument of the @code{gettext()} function.  Hence, this
argument is not a string constant but a variable argument (@code{$0}
is a global variable that holds the name of the Perl script being
executed).  The interpolation is performed by Perl before the string
argument is passed to @code{gettext()} and will therefore depend on
the name of the script which can only be determined at runtime.
Consequently, it is almost impossible that a translation can be looked
up at runtime (except if, by accident, the interpolated string is found
in the message catalog).

The @code{xgettext} program will therefore terminate parsing with a fatal
error if it encounters a variable inside of an extracted string.  In
general, this will happen for all kinds of string interpolations that
cannot be safely performed at compile time.  If you absolutely know
what you are doing, you can always circumvent this behavior:

@example
my $know_what_i_am_doing = "This is program $0!\n";
print gettext $know_what_i_am_doing;
@end example

Since the parser only recognizes strings and quote-like expressions,
but not variables or other terms, the above construct will be
accepted.  You will have to find another way, however, to let your
original string make it into your message catalog.

If invoked with the option @code{--extract-all}, resp. @code{-a},
variable interpolation will be accepted.  Rationale: You will
generally use this option in order to prepare your sources for
internationalization.

Please see the manual page @samp{man perlop} for details of strings and
quote-like expressions that are subject to interpolation and those
that are not.  Safe interpolations (that will not lead to a fatal
error) are:

@itemize @bullet

@item the escape sequences @code{\t} (tab, HT, TAB), @code{\n}
(newline, NL), @code{\r} (return, CR), @code{\f} (form feed, FF),
@code{\b} (backspace, BS), @code{\a} (alarm, bell, BEL), and @code{\e}
(escape, ESC).

@item octal chars, like @code{\033}
@*
Note that octal escapes in the range of 400-777 are translated into a 
UTF-8 representation, regardless of the presence of the @code{use utf8} pragma.

@item hex chars, like @code{\x1b}

@item wide hex chars, like @code{\x@{263a@}}
@*
Note that this escape is translated into a UTF-8 representation,
regardless of the presence of the @code{use utf8} pragma.

@item control chars, like @code{\c[} (CTRL-[)

@item named Unicode chars, like @code{\N@{LATIN CAPITAL LETTER C WITH CEDILLA@}}
@*
Note that this escape is translated into a UTF-8 representation,
regardless of the presence of the @code{use utf8} pragma.
@end itemize

The following escapes are considered partially safe:

@itemize @bullet

@item @code{\l} lowercase next char

@item @code{\u} uppercase next char

@item @code{\L} lowercase till \E

@item @code{\U} uppercase till \E

@item @code{\E} end case modification

@item @code{\Q} quote non-word characters till \E

@end itemize

These escapes are only considered safe if the string consists of
ASCII characters only.  Translation of characters outside the range
defined by ASCII is locale-dependent and can actually only be performed 
at runtime; @code{xgettext} doesn't do these locale-dependent translations
at extraction time.

Except for the modifier @code{\Q}, these translations, albeit valid,
are generally useless and only obfuscate your sources.  If a
translation can be safely performed at compile time you can just as
well write what you mean.

@node Interpolation II, Parentheses, Interpolation I, Perl
@subsubsection Valid Uses Of String Interpolation
@cindex Perl valid string interpolation

Perl is often used to generate sources for other programming languages
or arbitrary file formats.  Web applications that output HTML code
make a prominent example for such usage.

You will often come across situations where you want to intersperse
code written in the target (programming) language with translatable
messages, like in the following HTML example:

@example
print gettext <<EOF;
<h1>My Homepage</h1>
<script language="JavaScript"><!--
for (i = 0; i < 100; ++i) @{
    alert ("Thank you so much for visiting my homepage!");
@}
//--></script>
EOF
@end example

The parser will extract the entire here document, and it will appear
entirely in the resulting PO file, including the JavaScript snippet
embedded in the HTML code.  If you exaggerate with constructs like 
the above, you will run the risk that the translators of your package 
will look out for a less challenging project.  You should consider an 
alternative expression here:

@example
print <<EOF;
<h1>$gettext@{"My Homepage"@}</h1>
<script language="JavaScript"><!--
for (i = 0; i < 100; ++i) @{
    alert ("$gettext@{'Thank you so much for visiting my homepage!'@}");
@}
//--></script>
EOF
@end example

Only the translatable portions of the code will be extracted here, and
the resulting PO file will begrudgingly improve in terms of readability.

You can interpolate hash lookups in all strings or quote-like
expressions that are subject to interpolation (see the manual page
@samp{man perlop} for details).  Double interpolation is invalid, however:

@example
# TRANSLATORS: Replace "the earth" with the name of your planet.
print gettext qq@{Welcome to $gettext->@{"the earth"@}@};
@end example

The @code{qq}-quoted string is recognized as an argument to @code{xgettext} in
the first place, and checked for invalid variable interpolation.  The
dollar sign of hash-dereferencing will therefore terminate the parser 
with an ``invalid interpolation'' error.

It is valid to interpolate hash lookups in regular expressions:

@example
if ($var =~ /$gettext@{"the earth"@}/) @{
   print gettext "Match!\n";
@}
s/$gettext@{"U. S. A."@}/$gettext@{"U. S. A."@} $gettext@{"(dial +0)"@}/g;
@end example

@node Parentheses, Long Lines, Interpolation II, Perl
@subsubsection When To Use Parentheses
@cindex Perl parentheses

In Perl, parentheses around function arguments are mostly optional.
@code{xgettext} will always assume that all
recognized keywords (except for hashes and hash references) are names
of properly prototyped functions, and will (hopefully) only require
parentheses where Perl itself requires them.  All constructs in the
following example are therefore ok to use:

@example
@group
print gettext ("Hello World!\n");
print gettext "Hello World!\n";
print dgettext ($package => "Hello World!\n");
print dgettext $package, "Hello World!\n";

# The "fat comma" => turns the left-hand side argument into a
# single-quoted string!
print dgettext smellovision => "Hello World!\n";

# The following assignment only works with prototyped functions.
# Otherwise, the functions will act as "greedy" list operators and
# eat up all following arguments.
my $anonymous_hash = @{
   planet => gettext "earth",
   cakes => ngettext "one cake", "several cakes", $n,
   still => $works,
@};
# The same without fat comma:
my $other_hash = @{
   'planet', gettext "earth",
   'cakes', ngettext "one cake", "several cakes", $n,
   'still', $works,
@};

# Parentheses are only significant for the first argument.
print dngettext 'package', ("one cake", "several cakes", $n), $discarded;
@end group
@end example

@node Long Lines, Perl Pitfalls, Parentheses, Perl
@subsubsection How To Grok with Long Lines
@cindex Perl long lines

The necessity of long messages can often lead to a cumbersome or
unreadable coding style.  Perl has several options that may prevent
you from writing unreadable code, and
@code{xgettext} does its best to do likewise.  This is where the dot
operator (the string concatenation operator) may come in handy:

@example
@group
print gettext ("This is a very long"
               . " message that is still"
               . " readable, because"
               . " it is split into"
               . " multiple lines.\n");
@end group
@end example

Perl is smart enough to concatenate these constant string fragments
into one long string at compile time, and so is
@code{xgettext}.  You will only find one long message in the resulting
POT file.

Note that the future Perl 6 will probably use the underscore
(@samp{_}) as the string concatenation operator, and the dot 
(@samp{.}) for dereferencing.  This new syntax is not yet supported by
@code{xgettext}.

If embedded newline characters are not an issue, or even desired, you
may also insert newline characters inside quoted strings wherever you
feel like it:

@example
@group
print gettext ("<em>In HTML output
embedded newlines are generally no
problem, since adjacent whitespace
is always rendered into a single
space character.</em>");
@end group
@end example

You may also consider to use here documents:

@example
@group
print gettext <<EOF;
<em>In HTML output
embedded newlines are generally no
problem, since adjacent whitespace
is always rendered into a single
space character.</em>
EOF
@end group
@end example

Please do not forget that the line breaks are real, i.e.@: they
translate into newline characters that will consequently show up in
the resulting POT file.

@node Perl Pitfalls,  , Long Lines, Perl
@subsubsection Bugs, Pitfalls, And Things That Do Not Work
@cindex Perl pitfalls

The foregoing sections should have proven that
@code{xgettext} is quite smart in extracting translatable strings from
Perl sources.  Yet, some more or less exotic constructs that could be
expected to work, actually do not work.  

One of the more relevant limitations can be found in the
implementation of variable interpolation inside quoted strings.  Only
simple hash lookups can be used there:

@example
print <<EOF;
$gettext@{"The dot operator"
          . " does not work"
          . "here!"@}
Likewise, you cannot @@@{[ gettext ("interpolate function calls") ]@}
inside quoted strings or quote-like expressions.
EOF
@end example

This is valid Perl code and will actually trigger invocations of the
@code{gettext} function at runtime.  Yet, the Perl parser in
@code{xgettext} will fail to recognize the strings.  A less obvious
example can be found in the interpolation of regular expressions:

@example
s/<!--START_OF_WEEK-->/gettext ("Sunday")/e;
@end example

The modifier @code{e} will cause the substitution to be interpreted as
an evaluable statement.  Consequently, at runtime the function
@code{gettext()} is called, but again, the parser fails to extract the
string ``Sunday''.  Use a temporary variable as a simple workaround if
you really happen to need this feature:

@example
my $sunday = gettext "Sunday";
s/<!--START_OF_WEEK-->/$sunday/;
@end example

Hash slices would also be handy but are not recognized:

@example
my @@weekdays = @@gettext@{'Sunday', 'Monday', 'Tuesday', 'Wednesday',
                        'Thursday', 'Friday', 'Saturday'@};
# Or even:
@@weekdays = @@gettext@{qw (Sunday Monday Tuesday Wednesday Thursday
                         Friday Saturday) @};
@end example

This is perfectly valid usage of the tied hash @code{%gettext} but the
strings are not recognized and therefore will not be extracted.

Another caveat of the current version is its rudimentary support for
non-ASCII characters in identifiers.  You may encounter serious
problems if you use identifiers with characters outside the range of
'A'-'Z', 'a'-'z', '0'-'9' and the underscore '_'.

Maybe some of these missing features will be implemented in future
versions, but since you can always make do without them at minimal effort,
these todos have very low priority.

A nasty problem are brace format strings that already contain braces
as part of the normal text, for example the usage strings typically
encountered in programs:

@example
die "usage: $0 @{OPTIONS@} FILENAME...\n";
@end example

If you want to internationalize this code with Perl brace format strings,
you will run into a problem:

@example
die __x ("usage: @{program@} @{OPTIONS@} FILENAME...\n", program => $0);
@end example

Whereas @samp{@{program@}} is a placeholder, @samp{@{OPTIONS@}}
is not and should probably be translated. Yet, there is no way to teach
the Perl parser in @code{xgettext} to recognize the first one, and leave
the other one alone.

There are two possible work-arounds for this problem.  If you are
sure that your program will run under Perl 5.8.0 or newer (these
Perl versions handle positional parameters in @code{printf()}) or
if you are sure that the translator will not have to reorder the arguments
in her translation -- for example if you have only one brace placeholder
in your string, or if it describes a syntax, like in this one --, you can
mark the string as @code{no-perl-brace-format} and use @code{printf()}:

@example
# xgettext: no-perl-brace-format
die sprintf ("usage: %s @{OPTIONS@} FILENAME...\n", $0);
@end example

If you want to use the more portable Perl brace format, you will have to do
put placeholders in place of the literal braces:

@example
die __x ("usage: @{program@} @{[@}OPTIONS@{]@} FILENAME...\n",
         program => $0, '[' => '@{', ']' => '@}');
@end example

Perl brace format strings know no escaping mechanism.  No matter how this
escaping mechanism looked like, it would either give the programmer a
hard time, make translating Perl brace format strings heavy-going, or
result in a performance penalty at runtime, when the format directives
get executed.  Most of the time you will happily get along with
@code{printf()} for this special case.

@node PHP, Pike, Perl, List of Programming Languages
@subsection PHP Hypertext Preprocessor
@cindex PHP

@table @asis
@item RPMs
mod_php4, mod_php4-core, phpdoc

@item File extension
@code{php}, @code{php3}, @code{php4}

@item String syntax
@code{"abc"}, @code{'abc'}

@item gettext shorthand
@code{_("abc")}

@item gettext/ngettext functions
@code{gettext}, @code{dgettext}, @code{dcgettext}; starting with PHP 4.2.0
also @code{ngettext}, @code{dngettext}, @code{dcngettext}

@item textdomain
@code{textdomain} function

@item bindtextdomain
@code{bindtextdomain} function

@item setlocale
Programmer must call @code{setlocale (LC_ALL, "")}

@item Prerequisite
---

@item Use or emulate GNU gettext
use

@item Extractor
@code{xgettext}

@item Formatting with positions
@code{printf "%2\$d %1\$d"}

@item Portability
On platforms without gettext, the functions are not available.

@item po-mode marking
---
@end table

An example is available in the @file{examples} directory: @code{hello-php}.

@node Pike, GCC-source, PHP, List of Programming Languages
@subsection Pike
@cindex Pike

@table @asis
@item RPMs
roxen

@item File extension
@code{pike}

@item String syntax
@code{"abc"}

@item gettext shorthand
---

@item gettext/ngettext functions
@code{gettext}, @code{dgettext}, @code{dcgettext}

@item textdomain
@code{textdomain} function

@item bindtextdomain
@code{bindtextdomain} function

@item setlocale
@code{setlocale} function

@item Prerequisite
@code{import Locale.Gettext;}

@item Use or emulate GNU gettext
use

@item Extractor
---

@item Formatting with positions
---

@item Portability
On platforms without gettext, the functions are not available.

@item po-mode marking
---
@end table

@node GCC-source, Lua, Pike, List of Programming Languages
@subsection GNU Compiler Collection sources
@cindex GCC-source

@table @asis
@item RPMs
gcc

@item File extension
@code{c}, @code{h}.

@item String syntax
@code{"abc"}

@item gettext shorthand
@code{_("abc")}

@item gettext/ngettext functions
@code{gettext}, @code{dgettext}, @code{dcgettext}, @code{ngettext},
@code{dngettext}, @code{dcngettext}

@item textdomain
@code{textdomain} function

@item bindtextdomain
@code{bindtextdomain} function

@item setlocale
Programmer must call @code{setlocale (LC_ALL, "")}

@item Prerequisite
@code{#include "intl.h"}

@item Use or emulate GNU gettext
Use

@item Extractor
@code{xgettext -k_}

@item Formatting with positions
---

@item Portability
Uses autoconf macros

@item po-mode marking
yes
@end table

@node Lua, JavaScript, GCC-source, List of Programming Languages
@subsection Lua

@table @asis
@item RPMs
lua

@item File extension
@code{lua}

@item String syntax
@itemize @bullet

@item @code{"abc"}

@item @code{'abc'}

@item @code{[[abc]]}

@item @code{[=[abc]=]}

@item @code{[==[abc]==]}

@item ...

@end itemize

@item gettext shorthand
@code{_("abc")}

@item gettext/ngettext functions
@code{gettext.gettext}, @code{gettext.dgettext}, @code{gettext.dcgettext},
@code{gettext.ngettext}, @code{gettext.dngettext}, @code{gettext.dcngettext}

@item textdomain
@code{textdomain} function

@item bindtextdomain
@code{bindtextdomain} function

@item setlocale
automatic

@item Prerequisite
@code{require 'gettext'} or running lua interpreter with @code{-l gettext} option

@item Use or emulate GNU gettext
use

@item Extractor
@code{xgettext}

@item Formatting with positions
---

@item Portability
On platforms without gettext, the functions are not available.

@item po-mode marking
---
@end table

@node JavaScript, Vala, Lua, List of Programming Languages
@subsection JavaScript

@table @asis
@item RPMs
js

@item File extension
@code{js}

@item String syntax
@itemize @bullet

@item @code{"abc"}

@item @code{'abc'}

@end itemize

@item gettext shorthand
@code{_("abc")}

@item gettext/ngettext functions
@code{gettext}, @code{dgettext}, @code{dcgettext}, @code{ngettext},
@code{dngettext}

@item textdomain
@code{textdomain} function

@item bindtextdomain
@code{bindtextdomain} function

@item setlocale
automatic

@item Prerequisite
---

@item Use or emulate GNU gettext
use, or emulate

@item Extractor
@code{xgettext}

@item Formatting with positions
---

@item Portability
On platforms without gettext, the functions are not available.

@item po-mode marking
---
@end table

@node Vala,  , JavaScript, List of Programming Languages
@subsection Vala

@table @asis
@item RPMs
vala

@item File extension
@code{vala}

@item String syntax
@itemize @bullet

@item @code{"abc"}

@item @code{"""abc"""}

@end itemize

@item gettext shorthand
@code{_("abc")}

@item gettext/ngettext functions
@code{gettext}, @code{dgettext}, @code{dcgettext}, @code{ngettext},
@code{dngettext}, @code{dpgettext}, @code{dpgettext2}

@item textdomain
@code{textdomain} function, defined under the @code{Intl} namespace

@item bindtextdomain
@code{bindtextdomain} function, defined under the @code{Intl} namespace

@item setlocale
Programmer must call @code{Intl.setlocale (LocaleCategory.ALL, "")}

@item Prerequisite
---

@item Use or emulate GNU gettext
Use

@item Extractor
@code{xgettext}

@item Formatting with positions
Same as for the C language.

@item Portability
autoconf (gettext.m4) and #if ENABLE_NLS

@item po-mode marking
yes
@end table

@c This is the template for new languages.
@ignore

@ node
@ subsection 

@table @asis
@item RPMs

@item File extension

@item String syntax

@item gettext shorthand

@item gettext/ngettext functions

@item textdomain

@item bindtextdomain

@item setlocale

@item Prerequisite

@item Use or emulate GNU gettext

@item Extractor

@item Formatting with positions

@item Portability

@item po-mode marking
@end table

@end ignore

@node List of Data Formats,  , List of Programming Languages, Programming Languages
@section Internationalizable Data

Here is a list of other data formats which can be internationalized
using GNU gettext.

@menu
* POT::                         POT - Portable Object Template
* RST::                         Resource String Table
* Glade::                       Glade - GNOME user interface description
* GSettings::                   GSettings - GNOME user configuration schema
* AppData::                     AppData - freedesktop.org application description
* Preparing ITS Rules::  Preparing Rules for XML Internationalization
@end menu

@node POT, RST, List of Data Formats, List of Data Formats
@subsection POT - Portable Object Template

@table @asis
@item RPMs
gettext

@item File extension
@code{pot}, @code{po}

@item Extractor
@code{xgettext}
@end table

@node RST, Glade, POT, List of Data Formats
@subsection Resource String Table
@cindex RST

@table @asis
@item RPMs
fpk

@item File extension
@code{rst}

@item Extractor
@code{xgettext}, @code{rstconv}
@end table

@node Glade, GSettings, RST, List of Data Formats
@subsection Glade - GNOME user interface description

@table @asis
@item RPMs
glade, libglade, glade2, libglade2, intltool

@item File extension
@code{glade}, @code{glade2}, @code{ui}

@item Extractor
@code{xgettext}, @code{libglade-xgettext}, @code{xml-i18n-extract}, @code{intltool-extract}
@end table

@node GSettings, AppData, Glade, List of Data Formats
@subsection GSettings - GNOME user configuration schema

@table @asis
@item RPMs
glib2

@item File extension
@code{gschema.xml}

@item Extractor
@code{xgettext}, @code{intltool-extract}
@end table

@node AppData, Preparing ITS Rules, GSettings, List of Data Formats
@subsection AppData - freedesktop.org application description

@table @asis
@item RPMs
appdata-tools, appstream, libappstream-glib, libappstream-glib-builder

@item File extension
@code{appdata.xml}

@item Extractor
@code{xgettext}, @code{intltool-extract}, @code{itstool}
@end table

@menu
@end menu

@node Preparing ITS Rules,  , AppData, List of Data Formats
@subsection Preparing Rules for XML Internationalization
@cindex preparing rules for XML translation

Marking translatable strings in an XML file is done through a separate
"rule" file, making use of the Internationalization Tag Set standard
(ITS, @uref{http://www.w3.org/TR/its20/}).  The currently supported ITS
data categories are: @samp{Translate}, @samp{Localization Note},
@samp{Elements Within Text}, and @samp{Preserve Space}.  In addition to
them, @code{xgettext} also recognizes the following extended data
categories:

@table @samp
@item Context

This data category associates @code{msgctxt} to the extracted text.  In
the global rule, the @code{contextRule} element contains the following:

@itemize
@item
A required @code{selector} attribute.  It contains an absolute selector
that selects the nodes to which this rule applies.

@item
A required @code{contextPointer} attribute that contains a relative
selector pointing to a node that holds the @code{msgctxt} value.

@item
An optional @code{textPointer} attribute that contains a relative
selector pointing to a node that holds the @code{msgid} value.
@end itemize

@item Escape Special Characters

This data category indicates whether the special XML characters
(@code{<}, @code{>}, @code{&}, @code{"}) are escaped with entity
reference.  In the global rule, the @code{escapeRule} element contains
the following:

@itemize
@item
A required @code{selector} attribute.  It contains an absolute selector
that selects the nodes to which this rule applies.

@item
A required @code{escape} attribute with the value @code{yes} or @code{no}.
@end itemize

@item Extended Preserve Space

This data category extends the standard @samp{Preserve Space} data
category with the additional value @samp{trim}.  The value means to
remove the leading and trailing whitespaces of the content, but not to
normalize whitespaces in the middle.  In the global rule, the
@code{preserveSpaceRule} element contains the following:

@itemize
@item
A required @code{selector} attribute.  It contains an absolute selector
that selects the nodes to which this rule applies.

@item
A required @code{space} attribute with the value @code{default},
@code{preserve}, or @code{trim}.
@end itemize

@end table

All those extended data categories can only be expressed with global
rules, and the rule elements have to have the
@code{https://www.gnu.org/s/gettext/ns/its/extensions/1.0} namespace.

Given the following XML document in a file @file{messages.xml}:

@example
<?xml version="1.0"?>
<messages>
  <message>
    <p>A translatable string</p>
  </message>
  <message>
    <p translatable="no">A non-translatable string</p>
  </message>
</messages>
@end example

To extract the first text content ("A translatable string"), but not the
second ("A non-translatable string"), the following ITS rules can be used:

@example
<?xml version="1.0"?>
<its:rules xmlns:its="http://www.w3.org/2005/11/its" version="1.0">
  <its:translateRule selector="/messages" translate="no"/>
  <its:translateRule selector="//message/p" translate="yes"/>

  <!-- If 'p' has an attribute 'translatable' with the value 'no', then
       the content is not translatable.  -->
  <its:translateRule selector="//message/p[@@translatable = 'no']"
    translate="no"/>
</its:rules>
@end example

@samp{xgettext} needs another file called "locating rule" to associate
an ITS rule with an XML file.  If the above ITS file is saved as
@file{messages.its}, the locating rule would look like:

@example
<?xml version="1.0"?>
<locatingRules>
  <locatingRule name="Messages" pattern="*.xml">
    <documentRule localName="messages" target="messages.its"/>
  </locatingRule>
  <locatingRule name="Messages" pattern="*.msg" target="messages.its"/>
</locatingRules>
@end example

The @code{locatingRule} element must have a @code{pattern} attribute,
which denotes either a literal file name or a wildcard pattern of the
XML file.  The @code{locatingRule} element can have child
@code{documentRule} element, which adds checks on the content of the XML
file.

The first rule matches any file with the @file{.xml} file extension, but
it only applies to XML files whose root element is @samp{<messages>}.

The second rule indicates that the same ITS rule file are also
applicable to any file with the @file{.msg} file extension.  The
optional @code{name} attribute of @code{locatingRule} allows to choose
rules by name, typically with @code{xgettext}'s @code{-L} option.

The associated ITS rule file is indicated by the @code{target} attribute
of @code{locatingRule} or @code{documentRule}.  If it is specified in a
@code{documentRule} element, the parent @code{locatingRule} shouldn't
have the @code{target} attribute.

Locating rule files must have the @file{.loc} file extension.  Both ITS
rule files and locating rule files must be installed in the
@file{$prefix/share/gettext/its} directory.  Once those files are
properly installed, @code{xgettext} can extract translatable strings
from the matching XML files.

@subsubsection Two Use-cases of Translated Strings in XML

For XML, there are two use-cases of translated strings.  One is the case
where the translated strings are directly consumed by programs, and the
other is the case where the translated strings are merged back to the
original XML document.  In the former case, special characters in the
extracted strings shouldn't be escaped, while they should in the latter
case.  To control wheter to escape special characters, the @samp{Escape
Special Characters} data category can be used.

To merge the translations, the @samp{msgfmt} program can be used with
the option @code{--xml}.  @xref{msgfmt Invocation}, for more details
about how one calls the @samp{msgfmt} program.  @samp{msgfmt}'s
@code{--xml} option doesn't perform character escaping, so translated
strings can have arbitrary XML constructs, such as elements for markup.

@c This is the template for new data formats.
@ignore

@ node
@ subsection 

@table @asis
@item RPMs

@item File extension

@item Extractor
@end table

@end ignore

@node Conclusion, Language Codes, Programming Languages, Top
@chapter Concluding Remarks

We would like to conclude this GNU @code{gettext} manual by presenting
an history of the Translation Project so far.  We finally give
a few pointers for those who want to do further research or readings
about Native Language Support matters.

@menu
* History::                     History of GNU @code{gettext}
* References::                  Related Readings
@end menu

@node History, References, Conclusion, Conclusion
@section History of GNU @code{gettext}
@cindex history of GNU @code{gettext}

Internationalization concerns and algorithms have been informally
and casually discussed for years in GNU, sometimes around GNU
@code{libc}, maybe around the incoming @code{Hurd}, or otherwise
(nobody clearly remembers).  And even then, when the work started for
real, this was somewhat independently of these previous discussions.

This all began in July 1994, when Patrick D'Cruze had the idea and
initiative of internationalizing version 3.9.2 of GNU @code{fileutils}.
He then asked Jim Meyering, the maintainer, how to get those changes
folded into an official release.  That first draft was full of
@code{#ifdef}s and somewhat disconcerting, and Jim wanted to find
nicer ways.  Patrick and Jim shared some tries and experimentations
in this area.  Then, feeling that this might eventually have a deeper
impact on GNU, Jim wanted to know what standards were, and contacted
Richard Stallman, who very quickly and verbally described an overall
design for what was meant to become @code{glocale}, at that time.

Jim implemented @code{glocale} and got a lot of exhausting feedback
from Patrick and Richard, of course, but also from Mitchum DSouza
(who wrote a @code{catgets}-like package), Roland McGrath, maybe David
MacKenzie, Fran@,{c}ois Pinard, and Paul Eggert, all pushing and
pulling in various directions, not always compatible, to the extent
that after a couple of test releases, @code{glocale} was torn apart.
In particular, Paul Eggert -- always keeping an eye on developments
in Solaris -- advocated the use of the @code{gettext} API over
@code{glocale}'s @code{catgets}-based API.

While Jim took some distance and time and became dad for a second
time, Roland wanted to get GNU @code{libc} internationalized, and
got Ulrich Drepper involved in that project.  Instead of starting
from @code{glocale}, Ulrich rewrote something from scratch, but
more conforming to the set of guidelines who emerged out of the
@code{glocale} effort.  Then, Ulrich got people from the previous
forum to involve themselves into this new project, and the switch
from @code{glocale} to what was first named @code{msgutils}, renamed
@code{nlsutils}, and later @code{gettext}, became officially accepted
by Richard in May 1995 or so.

Let's summarize by saying that Ulrich Drepper wrote GNU @code{gettext}
in April 1995.  The first official release of the package, including
PO mode, occurred in July 1995, and was numbered 0.7.  Other people
contributed to the effort by providing a discussion forum around
Ulrich, writing little pieces of code, or testing.  These are quoted
in the @code{THANKS} file which comes with the GNU @code{gettext}
distribution.

While this was being done, Fran@,{c}ois adapted half a dozen of
GNU packages to @code{glocale} first, then later to @code{gettext},
putting them in pretest, so providing along the way an effective
user environment for fine tuning the evolving tools.  He also took
the responsibility of organizing and coordinating the Translation
Project.  After nearly a year of informal exchanges between people from
many countries, translator teams started to exist in May 1995, through
the creation and support by Patrick D'Cruze of twenty unmoderated
mailing lists for that many native languages, and two moderated
lists: one for reaching all teams at once, the other for reaching
all willing maintainers of internationalized free software packages.

Fran@,{c}ois also wrote PO mode in June 1995 with the collaboration
of Greg McGary, as a kind of contribution to Ulrich's package.
He also gave a hand with the GNU @code{gettext} Texinfo manual.

In 1997, Ulrich Drepper released the GNU libc 2.0, which included the
@code{gettext}, @code{textdomain} and @code{bindtextdomain} functions.

In 2000, Ulrich Drepper added plural form handling (the @code{ngettext}
function) to GNU libc.  Later, in 2001, he released GNU libc 2.2.x,
which is the first free C library with full internationalization support.

Ulrich being quite busy in his role of General Maintainer of GNU libc,
he handed over the GNU @code{gettext} maintenance to Bruno Haible in
2000.  Bruno added the plural form handling to the tools as well, added
support for UTF-8 and CJK locales, and wrote a few new tools for
manipulating PO files.

@node References,  , History, Conclusion
@section Related Readings
@cindex related reading
@cindex bibliography

@strong{ NOTE: } This documentation section is outdated and needs to be
revised.

Eugene H. Dorr (@file{dorre@@well.com}) maintains an interesting
bibliography on internationalization matters, called
@cite{Internationalization Reference List}, which is available as:
@example
ftp://ftp.ora.com/pub/examples/nutshell/ujip/doc/i18n-books.txt
@end example

Michael Gschwind (@file{mike@@vlsivie.tuwien.ac.at}) maintains a
Frequently Asked Questions (FAQ) list, entitled @cite{Programming for
Internationalisation}.  This FAQ discusses writing programs which
can handle different language conventions, character sets, etc.;
and is applicable to all character set encodings, with particular
emphasis on @w{ISO 8859-1}.  It is regularly published in Usenet
groups @file{comp.unix.questions}, @file{comp.std.internat},
@file{comp.software.international}, @file{comp.lang.c},
@file{comp.windows.x}, @file{comp.std.c}, @file{comp.answers}
and @file{news.answers}.  The home location of this document is:
@example
ftp://ftp.vlsivie.tuwien.ac.at/pub/8bit/ISO-programming
@end example

Patrick D'Cruze (@file{pdcruze@@li.org}) wrote a tutorial about NLS
matters, and Jochen Hein (@file{Hein@@student.tu-clausthal.de}) took
over the responsibility of maintaining it.  It may be found as:
@example
ftp://sunsite.unc.edu/pub/Linux/utils/nls/catalogs/Incoming/...
     ...locale-tutorial-0.8.txt.gz
@end example
@noindent
This site is mirrored in:
@example
ftp://ftp.ibp.fr/pub/linux/sunsite/
@end example

A French version of the same tutorial should be findable at:
@example
ftp://ftp.ibp.fr/pub/linux/french/docs/
@end example
@noindent
together with French translations of many Linux-related documents.

@node Language Codes, Country Codes, Conclusion, Top
@appendix Language Codes
@cindex language codes
@cindex ISO 639

The @w{ISO 639} standard defines two-letter codes for many languages, and
three-letter codes for more rarely used languages.
All abbreviations for languages used in the Translation Project should
come from this standard.

@menu
* Usual Language Codes::        Two-letter ISO 639 language codes
* Rare Language Codes::         Three-letter ISO 639 language codes
@end menu

@node Usual Language Codes, Rare Language Codes, Language Codes, Language Codes
@appendixsec Usual Language Codes

For the commonly used languages, the @w{ISO 639-1} standard defines two-letter
codes.

@table @samp
@include iso-639.texi
@end table

@node Rare Language Codes,  , Usual Language Codes, Language Codes
@appendixsec Rare Language Codes

For rarely used languages, the @w{ISO 639-2} standard defines three-letter
codes.  Here is the current list, reduced to only living languages with at least
one million of speakers.

@table @samp
@include iso-639-2.texi
@end table

@node Country Codes, Licenses, Language Codes, Top
@appendix Country Codes
@cindex country codes
@cindex ISO 3166

The @w{ISO 3166} standard defines two character codes for many countries
and territories.  All abbreviations for countries used in the Translation
Project should come from this standard.

@table @samp
@include iso-3166.texi
@end table

@node Licenses, Program Index, Country Codes, Top
@appendix Licenses
@cindex Licenses

The files of this package are covered by the licenses indicated in each
particular file or directory.  Here is a summary:

@itemize @bullet
@item
The @code{libintl} and @code{libasprintf} libraries are covered by the
GNU Lesser General Public License (LGPL).  
A copy of the license is included in @ref{GNU LGPL}.

@item
The executable programs of this package and the @code{libgettextpo} library
are covered by the GNU General Public License (GPL).
A copy of the license is included in @ref{GNU GPL}.

@item
This manual is free documentation.  It is dually licensed under the
GNU FDL and the GNU GPL.  This means that you can redistribute this
manual under either of these two licenses, at your choice.
@*
This manual is covered by the GNU FDL.  Permission is granted to copy,
distribute and/or modify this document under the terms of the
GNU Free Documentation License (FDL), either version 1.2 of the
License, or (at your option) any later version published by the
Free Software Foundation (FSF); with no Invariant Sections, with no
Front-Cover Text, and with no Back-Cover Texts.
A copy of the license is included in @ref{GNU FDL}.
@*
This manual is covered by the GNU GPL.  You can redistribute it and/or
modify it under the terms of the GNU General Public License (GPL), either
version 2 of the License, or (at your option) any later version published
by the Free Software Foundation (FSF).
A copy of the license is included in @ref{GNU GPL}.
@end itemize

@menu
* GNU GPL::                     GNU General Public License
* GNU LGPL::                    GNU Lesser General Public License
* GNU FDL::                     GNU Free Documentation License
@end menu

@page
@node GNU GPL, GNU LGPL, Licenses, Licenses
@appendixsec GNU GENERAL PUBLIC LICENSE
@cindex GPL, GNU General Public License
@cindex License, GNU GPL
@include gpl.texi
@page
@node GNU LGPL, GNU FDL, GNU GPL, Licenses
@appendixsec GNU LESSER GENERAL PUBLIC LICENSE
@cindex LGPL, GNU Lesser General Public License
@cindex License, GNU LGPL
@include lgpl.texi
@page
@node GNU FDL,  , GNU LGPL, Licenses
@appendixsec GNU Free Documentation License
@cindex FDL, GNU Free Documentation License
@cindex License, GNU FDL
@include fdl.texi

@node Program Index, Option Index, Licenses, Top
@unnumbered Program Index

@printindex pg

@node Option Index, Variable Index, Program Index, Top
@unnumbered Option Index

@printindex op

@node Variable Index, PO Mode Index, Option Index, Top
@unnumbered Variable Index

@printindex vr

@node PO Mode Index, Autoconf Macro Index, Variable Index, Top
@unnumbered PO Mode Index

@printindex em

@node Autoconf Macro Index, Index, PO Mode Index, Top
@unnumbered Autoconf Macro Index

@printindex am

@node Index,  , Autoconf Macro Index, Top
@unnumbered General Index

@printindex cp

@bye

@c Local variables:
@c texinfo-column-for-description: 32
@c End: