David Steinbrunner [Tue, 21 May 2013 10:59:07 +0000 (06:59 -0400)]
typo fix for File::Find
Bump $VERSION for File::Find.
Chris 'BinGOs' Williams [Fri, 24 May 2013 20:46:26 +0000 (21:46 +0100)]
Update HTTP-Tiny to CPAN version 0.029
[DELTA]
0.029 2013-04-17 13:49:07 America/New_York
[FIXED]
- Checks for new enough OpenSSL library before using SNI (otherwise
IO::Socket::SSL throws warnings)
0.028 2013-03-05 14:11:57 America/New_York
[SUPPORT]
- Fix repository/issue links to reflect proper repo name
0.027 2013-03-05 12:02:58 America/New_York
[SUPPORT]
- Changed metadata to point to the chansen github repository
for code and issues
[DOCUMENTATION]
- Added hyperlink for HTTP::CookieJar
0.026 2013-03-04 22:53:39 America/New_York
[ADDED]
- Added cookie support if an HTTP::CookieJar object is provided in the
'cookie_jar' attribute [Edward Zborowski]
Chris 'BinGOs' Williams [Fri, 24 May 2013 20:43:09 +0000 (21:43 +0100)]
Update Module-Build to CPAN version 0.4005
[DELTA]
0.4005 - Thu Apr 25 15:10:14 CEST 2013
[ENHANCEMENTS]
- Added --pureperl-only support
[BUG FIXES]
- #72176: pod2html will fail with an empty podpath in some cases. [Phillip Moore]
- Fix links between modules in HTML docs output [Michael Wild, Leon Timmermans]
0.4004 - Fri Mar 29 15:05:00 CET 2013
[BUG FIXES]
- Minor VMS fix for @INC [Craig Berry]
[ENHANCEMENTS]
- test_requires support has been added [Matsuno Tokuhiro]
Chris 'BinGOs' Williams [Fri, 24 May 2013 19:45:36 +0000 (20:45 +0100)]
Update Pod-Perldoc to CPAN version 3.20
[DELTA]
3.20 - Sat Apr 27 05:25:30 UTC 2013
* Make default formatter "ToTerm" instead of "ToMan" for
better UTF-8 support
* When dynamically building POD from perlfunc (perlvar)
make sure we open filehandles with utf8 and add an
=encoding (RT #80527) (RT #80640)
Chris 'BinGOs' Williams [Fri, 24 May 2013 19:22:20 +0000 (20:22 +0100)]
Update Pod-Usage to CPAN version 1.62
[DELTA]
1.62 (marekr)
Patches provided by rjbs - many thanks!
- These commits update Pod::Usage to only use Pod::Simple, so that any
reliance on Pod::Parser is removed, making it easier to remove Pod::Parser
from core.
- [PATCH 1/4] update Makefile: strict, INSTALLDIR, 5.6
- [PATCH 2/4] we will start to require perl 5.6 from here on out
- [PATCH 3/4] always use Pod::Text as default base class
- [PATCH 4/4] eliminate the branch in which Pod::Parser would be used
This commit is intended entirely to free Pod::Usage from any reliance on
Pod::Parser.
* Usage.pm now defaults to using Pod::Text, rather than checking $] to pick.
* $Pod::Select::MAX_HEADING_LEVEL is replaced with a local var
* &Pod::Select::_compile_section_spec is copied into this module
This isn't the most elegant fix, but it's a bit of a ball of mud. The code
is written to let you alter @ISA at runtime to something that is derived
either from Pod::Parser or Pod::Simple. This should probably be more
explicitly limited to Pod::Simple in future releases.
- CPAN#84031: eliminate branches using Pod::Parser
removed test suite dependencies on Pod::Parser
Nicholas Clark [Thu, 23 May 2013 11:46:05 +0000 (13:46 +0200)]
Ensure that the IV in struct pmop (for ithreads) is aligned properly.
The members of struct pmop vary between ithreads and non-ithreads builds.
For ithreads, an IV is used for op_pmoffset. For 32 bit platforms built
with -Duse64bitint, IVs are 64 bits, but pointers (and all other members of
all OP structures) are 32 bits, hence on 32 bit platforms OP structure sizes
can be an odd multiple of 4 bytes. Since the swap to the slab allocator,
this is causing problems on architectures where 8 byte IVs have to be 8
byte aligned, notably sparcs. The long-term fix is to replace the IV with
an appropriate 32 bit type, as it is being used for an array index. However,
for maint-5.18 we need a work around which doesn't change the public headers
and ideally doesn't impose a large speed or memory penalty.
Hence for now, simply avoid using the slab allocator when allocating a
struct pmop. Only do this for 32 bit platforms build with 8 byte IVs and
ithreads.
This is the interim fix for RT #118055.
Chris 'BinGOs' Williams [Fri, 24 May 2013 18:39:54 +0000 (19:39 +0100)]
Update to CPAN-Meta means META.* need regenerating
Chris 'BinGOs' Williams [Fri, 24 May 2013 18:26:56 +0000 (19:26 +0100)]
Update File-Temp to CPAN version 0.2301
[DELTA]
0.2301 2013-04-11 16:30:05 Europe/London
* dist.ini: Managed with Dist::Zilla now; generates Makefile.PL
to avoid circular dependency when using Build.PL
Chris 'BinGOs' Williams [Fri, 24 May 2013 18:22:44 +0000 (19:22 +0100)]
Update CPAN-Meta to CPAN version 2.130880
[DELTA]
2.130880 2013-03-29 09:51:00 America/New_York
[BUGFIX]
- Fix provides 'version' bug introduced in the 2.130870 fix
2.130870 2013-03-28 11:08:50 America/New_York
[BUGFIX]
- Don't add 'version' to 'provides' during conversion if one didn't
exist in the original [Reported by Miyagawa; fixed by David Golden]
Chris 'BinGOs' Williams [Fri, 24 May 2013 18:19:17 +0000 (19:19 +0100)]
Updated Module-Metadata to CPAN version 1.000014
[DELTA]
1.000014 - 2014-05-09
- Remove "now installs to 'site' for perl 5.12+" from last version
1.000013 - 2013-05-08
- Fix reliance on recent Test::Builder
- Make tests perl 5.6 compatible
- now installs to 'site' for perl 5.12+, as per p5p recommendation
1.000012 - 2013-05-04
- improved package detection heuristics (thanks, Edward Zborowski!)
- fix ->contains_pod (RT#84932, Tokuhiro Matsuno)
- fix detection of pod after __END__ (RT79656, Tokuhiro Matsuno)
Chris 'BinGOs' Williams [Fri, 24 May 2013 18:17:43 +0000 (19:17 +0100)]
Update File-Fetch to CPAN version 0.42
[DELTA]
Changes for 0.42 Fri Apr 12 15:28:34 2013
=================================================
* Skip slurp tests for git://
Changes for 0.40 Fri Apr 12 11:18:52 2013
=================================================
* Added git:// url support
Chris 'BinGOs' Williams [Fri, 24 May 2013 18:15:58 +0000 (19:15 +0100)]
Update perlfaq to CPAN version 5.0150043
[DELTA]
5.0150043 Sun 12 May 2013 19:40:17 +0100
* Various typos and URL fixes (dsteinbrunner)
* Typo (argrath)
Chris 'BinGOs' Williams [Fri, 24 May 2013 18:14:08 +0000 (19:14 +0100)]
Update Encode to CPAN version 2.51
[DELTA]
$Revision: 2.51 $ $Date: 2013/04/29 22:19:11 $
! Encode.xs
Addressed: Encode.xs doesn't compile with Microsoft C compiler
https://rt.cpan.org/Public/Bug/Display.html?id=84920
! MANIFEST
Addressed: t/taint.t missing
https://rt.cpan.org/Public/Bug/Display.html?id=84919
2.50 2013/04/26 18:30:46
! Encode.xs Unicode/Unicode.xs
lib/Encode/Unicode/UTF7.pm lib/CN/HZ.pm lib/Encode/GSM0338.pm
t/taint.t
Addressed: Encode::encode and Encode::decode
gratuitously launders tainted data
Taintedness now propagates as it should.
https://rt.cpan.org/Ticket/Display.html?id=84879
! encoding.pm
Addressed: 5.18 deprecation
https://rt.cpan.org/Ticket/Display.html?id=84709
! bin/piconv
Applied: Update piconv documentation
https://rt.cpan.org/Ticket/Display.html?id=84695
Chris 'BinGOs' Williams [Fri, 24 May 2013 18:12:47 +0000 (19:12 +0100)]
Update Config-Perl-V version in Maintainers.pl
H.Merijn Brand [Fri, 24 May 2013 15:33:07 +0000 (17:33 +0200)]
Update Config-Perl-V to version 0.18
H.Merijn Brand [Fri, 24 May 2013 14:36:55 +0000 (16:36 +0200)]
Add NO_TAINT_SUPPORT to S_Internals_V ()
David Steinbrunner [Tue, 21 May 2013 10:56:08 +0000 (06:56 -0400)]
typo fixes for porting scripts
David Steinbrunner [Tue, 21 May 2013 10:57:44 +0000 (06:57 -0400)]
typo fixes for version pod
David Golden [Thu, 23 May 2013 23:11:23 +0000 (19:11 -0400)]
Try to avoid nesting F<$F<...>> in generated Config.pm
David Golden [Thu, 23 May 2013 22:08:48 +0000 (18:08 -0400)]
fix various podcheck nits
David Golden [Thu, 23 May 2013 20:58:20 +0000 (16:58 -0400)]
fix various Pod line length warnings
David Golden [Thu, 23 May 2013 22:08:24 +0000 (18:08 -0400)]
help t/podcheck.t skip duplicate pod in utils
David Golden [Thu, 23 May 2013 22:07:58 +0000 (18:07 -0400)]
stop t/podcheck.t from flagging blockquotes
David Golden [Thu, 23 May 2013 20:33:09 +0000 (16:33 -0400)]
Make t/podcheck.t less sensitive
The max line length is increased to 100. This eliminates dozens of
known issues.
The warnings about using F<> and L<> instead of C<> have been removed
on the rationale that they are too opinionated. They flag stylistic
issues rather than syntactic or structural issues and have minimal
impact on the effectiveness of the resulting documentation.
David Steinbrunner [Tue, 21 May 2013 10:53:18 +0000 (06:53 -0400)]
typo fixes for porting modules
Bump $VERSION for Porting/Maintainers.pm.
David Steinbrunner [Tue, 21 May 2013 10:52:43 +0000 (06:52 -0400)]
typo fixes for porting pod note optimiser
Change to optimizer to be consistent with other uses of optimizer in todo.
Clear up confusion in regex optimizer entry.
David Steinbrunner [Tue, 21 May 2013 10:50:14 +0000 (06:50 -0400)]
making DOS stand out in dosify
David Steinbrunner [Tue, 21 May 2013 10:49:13 +0000 (06:49 -0400)]
typo fix for os2 module
Bump $VERSION for os2/OS2/OS2-Process/Process.pm.
David Steinbrunner [Tue, 21 May 2013 10:48:37 +0000 (06:48 -0400)]
typo fix for util script
David Steinbrunner [Tue, 21 May 2013 10:48:17 +0000 (06:48 -0400)]
typo fixes for win32 scripts
Chris 'BinGOs' Williams [Thu, 23 May 2013 20:52:08 +0000 (21:52 +0100)]
Sync Module-CoreList version in Maintainers.pl
Nicholas Clark [Tue, 17 Apr 2012 13:59:44 +0000 (15:59 +0200)]
Refactor autodoc.pl to read MANIFEST line by line, instead of slurping it.
Nicholas Clark [Tue, 17 Apr 2012 13:47:13 +0000 (15:47 +0200)]
autodoc.pl now uses regen/embed_lib.pl to parse embed.fnc
Previously it had its own custom parsing code for embed.fnc.
Nicholas Clark [Sun, 8 Apr 2012 17:44:11 +0000 (19:44 +0200)]
Refactor autodoc.pl to use regen_lib.pl to generate file headers & footers.
This reduces duplication of code that generates '-*- buffer-read-only: t -*-',
'ex: set ro:' etc.
Nicholas Clark [Sun, 8 Apr 2012 17:36:49 +0000 (19:36 +0200)]
In regen/regen_lib.pl, add 'Pod' as a third supported 'language'.
Pod needs a commenting style distinct from C and Perl. (ie the empty string)
Nicholas Clark [Sun, 8 Apr 2012 11:28:06 +0000 (13:28 +0200)]
t/porting/diag.t now uses regen/embed_lib.pl to parse embed.fnc
Previously it had its own custom parsing code for embed.fnc.
Nicholas Clark [Sun, 8 Apr 2012 11:08:28 +0000 (13:08 +0200)]
In t/porting/diag.t, use TestInit to chdir and set up @INC.
Also eliminate the $|=1, as it's really not necessary for tests that neither
fork nor shell out to other processes.
Nicholas Clark [Sun, 8 Apr 2012 17:56:47 +0000 (19:56 +0200)]
Avoid VMS running autodoc.pl twice for no reason.
Previous descrip_mms.template carried separate rules to build perlintern.pod
and perlapi.pod, each of which caused autodoc.pl to run. autdoc.pl builds
both files when it runs, so this was duplicate work.
Nicholas Clark [Wed, 22 May 2013 09:48:03 +0000 (11:48 +0200)]
Refactor the "check categories and severity" logic in diag.t
Move the "todo" bailout as early as possible, before any work is done.
Replace C<use 5.01> [sic] with C<use feature 'state'>, as this makes the
intent much clearer.
Use parenthesis for like() and is() as this is consistent with the rest of the
test, and permits a simplification of the BEGIN handling.
James E Keenan [Thu, 23 May 2013 01:26:55 +0000 (03:26 +0200)]
Provide alternate email address for jkeenan.
James Keenan [Thu, 23 May 2013 01:06:54 +0000 (03:06 +0200)]
Correct known_pod_issues.dat.
David Steinbrunner [Tue, 21 May 2013 10:47:33 +0000 (06:47 -0400)]
typo fix for vms script
David Steinbrunner [Tue, 21 May 2013 10:47:05 +0000 (06:47 -0400)]
typo fixes for regen scripts
David Steinbrunner [Tue, 21 May 2013 10:46:18 +0000 (06:46 -0400)]
typo fixes for root level scripts
Add David Steinbrunner to AUTHORS.
Update pod issues database.
Karl Williamson [Wed, 22 May 2013 19:28:07 +0000 (13:28 -0600)]
perldiag: Describe \g syntax error
Commit
779fedd7c3021f013726c8f53cb9e66c54637ebf added a warning,
which is nearly identical to an existing one, but didn't change the
description of the existing one to encompass both. This rectifies that.
Karl Williamson [Wed, 20 Mar 2013 17:42:18 +0000 (11:42 -0600)]
regcomp.c: Actually emit proper warning
Before this commit, /\g/ raised the wrong warning
Reference to invalid group 0
This rearranges the code so that the proper warning is emitted.
Unterminated \g... pattern
Karl Williamson [Wed, 22 May 2013 02:34:47 +0000 (20:34 -0600)]
Make Unicode::UCD::search_invlist() available
This commit documents this function, removing the initial underscore
from its name. (And it hardens input checking.)
Karl Williamson [Wed, 22 May 2013 01:41:44 +0000 (19:41 -0600)]
Unicode::UCD Clarifications in pod
There are no "missing" values in inversion maps; there is a default
value returned for each one. So change the example variables' names.
Plus another sentence rewording for clarity.
Karl Williamson [Mon, 20 May 2013 18:13:22 +0000 (12:13 -0600)]
Unicode::UCD: Move function in file.
This is in preparation for making this function public, and it should be
listed in the pod later than it otherwise would be.
Karl Williamson [Mon, 20 May 2013 16:41:01 +0000 (10:41 -0600)]
Unicode::UCD: Correct wrong pod info
This was erroneous. Extra clarifications are also added.
Ricardo Signes [Wed, 22 May 2013 13:33:16 +0000 (09:33 -0400)]
remove a spurious space
Nicholas Clark [Wed, 22 May 2013 11:56:50 +0000 (13:56 +0200)]
Some more comments describing the slab allocator.
These are based on the description of commit
8be227ab5eaa23f2, but provide
explanations near to the relevant C code.
Nicholas Clark [Mon, 8 Apr 2013 14:30:18 +0000 (16:30 +0200)]
Add comments describing the operation of S_scan_ident().
Nicholas Clark [Mon, 8 Apr 2013 14:27:19 +0000 (16:27 +0200)]
More tests for the "Ambiguous use of ... resolved ..." warning S_scan_ident().
In particular, test that the special case for reporting $# as @ works.
Max Maischein [Tue, 21 May 2013 19:40:21 +0000 (21:40 +0200)]
Fix typo in Porting/sync-with-cpan
I will run `make test_porting` on the correct branch before submitting a patch.
I will run `make test_porting` on the correct branch before submitting a patch.
I will run `make test_porting` on the correct branch before submitting a patch.
Max Maischein [Tue, 21 May 2013 17:09:47 +0000 (19:09 +0200)]
Eliminate external call to tar executable in favour of Archive::Tar
Archive::Tar has ->extract_archive, which does Just That. This means
we lose immediate support for bzip2 files, but not all
incarnations of `tar` support it either.
Signed-off-by: David Golden <dagolden@cpan.org>
Chris 'BinGOs' Williams [Tue, 21 May 2013 13:13:52 +0000 (14:13 +0100)]
Add new Module::CoreList::Utils to MANIFEST
Chris 'BinGOs' Williams [Tue, 21 May 2013 12:32:53 +0000 (13:32 +0100)]
Merge branch 'smoke-me/coreutils' into blead
Nicholas Clark [Wed, 24 Apr 2013 15:01:43 +0000 (17:01 +0200)]
Remove 2 superfluous use lines from ExtUtils::CBuilder.
use IO::File was added with the upgrade to 0.23 (commit
c3fb68a339256eb3 in
April 2008), but the use of IO::File in the code was eliminated by the
upgrade to 0.2800 (commit
06e8058f27e4269b in Dec 2010), which replaced
the code in question with use of File::Temp. The latter refactoring also
added the use Data::Dumper; line, but did not add any code which uses
Data::Dumper.
Nicholas Clark [Wed, 10 Apr 2013 11:44:07 +0000 (13:44 +0200)]
Refactor 3 tests so that they avoid clearing %:: via list assignment.
For all 3, clearing %:: was intended to expose another bug, not to directly
test the effects of clearing %::
Tested by building the 3 revisions that added the tests, and confirming that
the revised test also triggers the bug that each fixed.
Nicholas Clark [Fri, 5 Apr 2013 10:12:53 +0000 (12:12 +0200)]
Test that caller does not SEGV when the current package is undefined.
Prior to commit
d4d03940c58a0177 this code would SEGV under ithreads.
Test suggested by Brian Fraser.
Nicholas Clark [Mon, 20 May 2013 15:55:46 +0000 (17:55 +0200)]
Remove fakethr.h and eliminate all references to it and FAKE_THREADS
fakethr.h and FAKE_THREADS were for a "green" threads implementation of
5005threads. 5005threads itself is long gone, and it's not clear that
-DFAKE_THREADS *ever* built correctly. Certainly it did not work for the
5.005 release, and it did not work at the time of the commits for the initial
checkin. The closest that it seems to have been to working is around commit
c6ee37c52f2ca9e5 (Dec 1997), where the headers no longer contained errors,
but perl.c failed to compile.
Nicholas Clark [Tue, 21 May 2013 05:49:29 +0000 (07:49 +0200)]
Merge the refactoring of x2p's Makefile to blead.
Nicholas Clark [Wed, 15 May 2013 19:48:12 +0000 (21:48 +0200)]
Simplify x2p's Makefile.
Eliminate plexe and plc, which relate to tests for the compiler and were
added as part of commit
52cebf5efc9883c7 in May 1988. The compiler was
removed by commit
de1254415ffeb03b in Sept 2006.
The files a2p.man and s2p.man were removed with Perl 5.000.
The macros 'public', 'private' and 'util' mirror the structure of the top
level Makefile.SH. However, x2p only ever used 'public', so simplify things
by inlining the value of 'public' in the only place that uses it.
'addedbyconf' now only contains the value of 'plextract', so use the latter
directly in the one place that had used 'addedbyconf'.
Nicholas Clark [Wed, 15 May 2013 19:34:21 +0000 (21:34 +0200)]
Remove rules to delete *.loadmap and *.prelmap files, generated on VM/ESA
These rules to clean up properly on VM/ESA were added in Oct 1998 by
perforce changes 1961, 1962 and 1963 in the confperl branch. VM/ESA support
was removed by commit
043fec90e88a2e23 in Aug 2012.
Nicholas Clark [Wed, 15 May 2013 19:26:59 +0000 (21:26 +0200)]
Add x2p/Makefile.SH to the *.SH files processed by the top level Makefile.
This makes the top level Makefile responsible for deleting x2p/Makefile.
With this, we can eliminate the 'sh' and 'shextract' from x2p's Makefile.
Nicholas Clark [Wed, 15 May 2013 19:17:31 +0000 (21:17 +0200)]
Remove the now unused 'shlist' Makefile targets.
Nicholas Clark [Wed, 15 May 2013 18:57:37 +0000 (20:57 +0200)]
Programmatically generate dependencies for all *.SH files.
Previously dependencies for Makefile, config.h and makedepend were
explicitly coded into verbatim sections of Makefile.SH, with most of the
others being generated by makedepend.SH
Nicholas Clark [Wed, 15 May 2013 18:25:47 +0000 (20:25 +0200)]
Add runtests.SH to the code that generates the targets sh and shextract.
The generated Makefile is functionally identical.
Nicholas Clark [Wed, 15 May 2013 15:14:50 +0000 (17:14 +0200)]
Eliminate x2p/cflags.SH. x2p's Makefile can use ../cflags
x2p/Makefile is only used on *nix, so this should not present a problem.
Nicholas Clark [Wed, 15 May 2013 17:40:46 +0000 (19:40 +0200)]
Pass in to makedepend the name(s) of files that all objects depend on.
Nicholas Clark [Wed, 15 May 2013 14:54:40 +0000 (16:54 +0200)]
In the Makefile, generate the value of shextract programmatically.
shextract lists the files generated by shell scripts named *.SH which need
deleting. Instead of maintaining it by hand, generate it in Makefile.SH from
the list of *.SH files.
This indirectly eliminates the last reference to F<makedir>. makedir.SH
was eliminated by commit
ebca06e95b02d1a8 and the reference to makedir.SH in
Makefile.SH eliminated by commit
847471027aa42997 (both Sept 2009), but
that commit missed that the reference to makedir a few lines later was
related.
Nicholas Clark [Wed, 15 May 2013 11:01:22 +0000 (13:01 +0200)]
Defer running makedepend in x2p/ until it's needed.
We can run makedepend in x2p/ as part of the regular build process, which
means that it can run as part of a parallel make. This gets a slight
speedup (0.2s on this system), and increases flexibility on how x2p/Makefile
is generated.
Nicholas Clark [Wed, 15 May 2013 10:49:44 +0000 (12:49 +0200)]
Add x2p/makefile to the dependencies for targets within x2p/
This will give us more flexibility in when and how we generate it.
[x2p/makefile is x2p/GNUmakefile on OS X]
Nicholas Clark [Wed, 15 May 2013 10:33:05 +0000 (12:33 +0200)]
Remove vestigial *.SH code which can link the extracted file into ../
4 of the *.SH files have code which links the extracted file into the parent
directory if the current directory ends SH. This code dates from Perl 1.
It seems that the intent was that the *.SH files could live in a SH/
directory, but that feature was neither used or mentioned in the
documentation in Perl 1, and hasn't been used since.
Nicholas Clark [Mon, 20 May 2013 19:42:33 +0000 (21:42 +0200)]
Merge the pack refactoring to blead.
Nicholas Clark [Mon, 20 May 2013 19:40:27 +0000 (21:40 +0200)]
Notes in perldelta for the tiny XS-visible changes of the pack refactoring.
Nicholas Clark [Wed, 8 May 2013 08:40:07 +0000 (10:40 +0200)]
Annotate the "cold" paths in pp_pack.c with UNLIKELY().
The usual case is nice regular bytes in the host's nice regular order.
Nicholas Clark [Tue, 7 May 2013 16:25:53 +0000 (18:25 +0200)]
Eliminate the implementations of [hv]to[vh][ls] for mixed-endian systems.
As pp_pack.c has had mixed-endian support removed, there is little point in
keeping code in perl.h and util.c only needed for architectures that cannot
be built.
Nicholas Clark [Tue, 7 May 2013 16:17:39 +0000 (18:17 +0200)]
Eliminate Perl_my_swabn(), as it is now unused.
It is not marked as part of the API, and no code on CPAN is using it.
Nicholas Clark [Tue, 7 May 2013 15:39:42 +0000 (17:39 +0200)]
When endian-swapping in pack, simply copy the bytes in reverse order.
This should restore support for big endian Crays. It doesn't support
mixed-endian systems.
Nicholas Clark [Tue, 7 May 2013 14:22:08 +0000 (16:22 +0200)]
Pass 'needs_swap' as a parameter to SHIFT_BYTES.
It feels wrong to have it as an implicit parameter sucked in via the textual
expansion of the macro. Whilst it can be derived from the parameter
'datumtype', it seems that the C compiler generates a lot less efficient code
that way.
Nicholas Clark [Tue, 7 May 2013 13:15:54 +0000 (15:15 +0200)]
When endian-swapping in unpack, simply copy the bytes in reverse order.
It is considerably simpler to re-order the bytes before reading them into
the variable of the desired type, than to read into the variable and then
need a specialised "reverse this integer" function for each size of integer.
This should restore support for big endian Crays. It doesn't support
mixed-endian systems. Support for mixed-endian systems can be restored (if
needed) by re-ordering the bytes correctly at the locations which currently
only know how to reverse the bytes.
Nicholas Clark [Tue, 7 May 2013 10:23:00 +0000 (12:23 +0200)]
Fold the implementation of DO_BO_UNPACK into SHIFT_BYTES(), its only user.
Nicholas Clark [Tue, 7 May 2013 09:53:17 +0000 (11:53 +0200)]
Re-implement SHIFT16 and SHIFT32 as wrappers to SHIFT_BYTES.
Nicholas Clark [Tue, 7 May 2013 08:08:53 +0000 (10:08 +0200)]
Refactor the two uses of SHIFT{16,32} not followed by DO_BO_UNPACK().
There is one place where SHIFT16() is not immediately followed by
DO_BO_UNPACK() ('n!' and 'v!'), and one place where SHIFT32() is not
immediately followed by DO_BO_UNPACK() ('N!' and 'V!'). Every other use of
either macro is followed by DO_BO_UNPACK(). For the two locations that don't
have DO_BO_UNPACK(), the modifiers '<' and '>' are illegal, hence
DO_BO_UNPACK() will always be a no-op. Hence add DO_BO_UNPACK() to both
(which permits refactoring to simplify things), and add assert()s that the
assumptions are true.
Nicholas Clark [Tue, 7 May 2013 07:25:39 +0000 (09:25 +0200)]
Remove the now unused 'type' argument from the macros DO_BO_(UN)?PACK.
Nicholas Clark [Mon, 6 May 2013 18:40:34 +0000 (20:40 +0200)]
As DO_BO_(UN)?PACK are now implemented with my_swabn(), simplify 'J' and 'j'.
The conditionally compiled code to figure out the correct named integer
size for UVs and IVs is no longer needed, as my_swabn() will swap UVs and
IVs, whatever their length.
Nicholas Clark [Mon, 6 May 2013 17:58:10 +0000 (19:58 +0200)]
Merge the identical definitions of DO_BO_UNPACK and DO_BO_PACK.
Delete the now-unused mixed-endian definitions of DO_BO_UNPACK and
DO_BO_PACK, and the macro BO_CANT_DOIT, as they were the last two things
using it.
Nicholas Clark [Mon, 6 May 2013 17:41:10 +0000 (19:41 +0200)]
Swap byte order in DO_BO_(UN)?PACK based on a variable needs_swap.
Add the macro NEEDS_SWAP to initialise needs_swap based on
TYPE_ENDIANNESS(datumtype). This makes the two definitions of DO_BO_UNPACK
identical, and the two definitions of DO_BO_PACK identical.
This also makes building pp_pack.c on a mixed endian byteorder architecture
a compile time error. The commit adds pointers on where to add code to
re-instate support for such architectures.
Nicholas Clark [Mon, 6 May 2013 16:43:26 +0000 (18:43 +0200)]
DO_BO_{UN,}PACK can use my_swabn() directly.
This removes the only users of my_{be,le,h}to{be,le,h}n, so remove the
definitions for those macros.
Nicholas Clark [Mon, 6 May 2013 16:05:24 +0000 (18:05 +0200)]
Use DO_BO_{UN,}PACK in place of DO_BO_UNPACK_{N,PC}
Also, eliminate the intermediate macros DO_BO_PACK_PTR and DO_BO_UNPACK_PTR.
Nicholas Clark [Mon, 6 May 2013 15:16:18 +0000 (17:16 +0200)]
Eliminate my_{hto[bl]e,[bl]etoh}{16,32,64,s,i,l} as nothing now uses them.
Nicholas Clark [Mon, 6 May 2013 14:54:12 +0000 (16:54 +0200)]
In pp_pack.c, refactor DO_BO_(UN)?PACK to use my_letohn etc
Previously they were using a my_letoh* function for the appropriate size.
This commit probably breaks fixed 16 and 32 bit templates on big endian
Crays where sizeof(short) and sizeof(int) are 8. A future commit will
restore support. (Although it's not clear whether such Crays can still
build blead cleanly, as we've not had any feedback since Boeing
decommissioned their Cray.)
Nicholas Clark [Mon, 6 May 2013 14:36:09 +0000 (16:36 +0200)]
In pp_pack.c, refactor DO_BO_(UN)?PACK_PTR to use my_letohn etc
Previously they were casting the pointer to an integer value, calling
a my_letoh* function for the appropriate size, and casting the return value
back to a pointer.
Nicholas Clark [Mon, 6 May 2013 14:22:01 +0000 (16:22 +0200)]
In pp_pack.c, refactor DO_BO_(UN)?PACK_N to use sizeof() on the variable.
Previously they were using sizeof() on the type argument passed to the
macro. This change makes their implementation more like DO_BO_(UN)?PACK
and DO_BO_(UN)?PACK_PTR.
Nicholas Clark [Mon, 6 May 2013 13:46:02 +0000 (15:46 +0200)]
Refactor macros in pp_pack.c, removing support for mixed-endian byte orders.
The byte-order handling can be simplified considerably if we don't have to
support mixed-endian systems. It's not clear whether Perl 5.000 even
compiled on PDP-11s, let alone more recent versions.
Support probably can be added back at the end of the refactoring, if someone
has time and a PDP-11 to test on, but for the intermediate stages it is a
lot easier not to need to think about such platforms.
Nicholas Clark [Mon, 6 May 2013 12:40:04 +0000 (14:40 +0200)]
Eliminate the conditionally-compiled fallback functions for htonl etc.
These are now only being used for mixed-endian platforms which do not
provide their own htnol (etc) functions. Given that the fallbacks have been
buggy since they were added in Perl 3.0, it's safe to conclude that no
mixed-endian platforms were ever using these functions.
It's also unclear why these functions were ever marked as 'A', part of the
API. XS code can't call them directly, as it can't rely on them being
compiled. Unsurprisingly, no code on CPAN references them.
Nicholas Clark [Mon, 6 May 2013 11:28:20 +0000 (13:28 +0200)]
Remove buggy loop-based byte swapping code.
The irony is that the union-based code special-cased for little endian
systems actually works everywhere, even on mixed-endian systems.
Nicholas Clark [Mon, 6 May 2013 06:20:57 +0000 (08:20 +0200)]
Update the Win32 miniperl conditional compilation, for the htonl etc changes.
As of commit
19253ae62cd13079 (Oct 2012), miniperl on Win32 avoids using
Winsock. The win32_* wrappers for htonl etc had used the pre-processor
macro MYSWAP to conditionally compile in the correct code. However, MYSWAP
was defined as a side effect of using the htonl etc wrappers in util.c,
which are no longer needed. Hence use the WIN32_NO_SOCKETS macro directly
in win32sck.c for the correct conditional compilation.