1 comm: add an option, --output-delimiter=STR
2 Files to change: src/comm.c, ChangeLog, NEWS, doc/coreutils.texi,
3 Add a new file, tests/misc/comm (use another file in that directory as
4 a template), to exercise the new option. Suggestion from Dan Jacobson.
7 Now that gnulib supports *printf("%a"), import one of the
8 *printf-posix modules so that printf(1) will support %a even on
9 platforms where the native *printf(3) is deficient.
10 Suggestion form Eric Blake.
12 strip: add an option to specify the program used to strip binaries.
13 suggestion from Karl Berry
16 Address this comment: FIXME: mv's behavior in this case is system-dependent
17 Better still: fix the code so it's *not* system-dependent.
19 ls: add --format=FORMAT option that controls how each line is printed.
21 cp --no-preserve=X should not attempt to preserve attribute X
22 reported by Andreas Schwab
24 copy.c: Address the FIXME-maybe comment in copy_internal.
25 And once that's done, add an exclusion so that `cp --link'
26 no longer incurs the overhead of saving src. dev/ino and dest. filename
29 See if we can be consistent about where --verbose sends its output:
30 These all send --verbose output to stdout:
31 head, tail, rm, cp, mv, ln, chmod, chown, chgrp, install, ln
32 These send it to stderr:
36 Write an autoconf test to work around build failure in HPUX's 64-bit mode.
37 See notes in README -- and remove them once there's a work-around.
39 Integrate use of sendfile, suggested here:
40 http://mail.gnu.org/archive/html/bug-fileutils/2003-03/msg00030.html
41 I don't plan to do that, since a few tests demonstrate no significant benefit.
43 Should printf '\0123' print "\n3"?
44 per report from TAKAI Kousuke on Mar 27
45 http://mail.gnu.org/archive/html/bug-coreutils/2003-03/index.html
47 printf: consider adapting builtins/printf.def from bash
49 df: add `--total' option, suggested here http://bugs.debian.org/186007
51 seq: give better diagnostics for invalid formats:
52 e.g. no or too many % directives
53 seq: consider allowing format string to contain no %-directives
55 tail: don't use xlseek; it *exits*.
56 Instead, maybe use a macro and return nonzero.
58 add mktemp? Suggested by Nelson Beebe
60 tr: support nontrivial equivalence classes, e.g. [=e=] with LC_COLLATE=fr_FR
62 lib/strftime.c: Since %N is the only format that we need but that
63 glibc's strftime doesn't support, consider using a wrapper that
64 would expand /%(-_)?\d*N/ to the desired string and then pass the
65 resulting string to glibc's strftime.
67 unexpand: [http://www.opengroup.org/onlinepubs/007908799/xcu/unexpand.html]
68 printf 'x\t \t y\n'|unexpand -t 8,9 should print its input, unmodified.
69 printf 'x\t \t y\n'|unexpand -t 5,8 should print "x\ty\n"
71 Let GNU su use the `wheel' group if appropriate.
72 (there are a couple patches, already)
74 sort: Investigate better sorting algorithms; see Knuth vol. 3.
76 We tried list merge sort, but it was about 50% slower than the
77 recursive algorithm currently used by sortlines, and it used more
78 comparisons. We're not sure why this was, as the theory suggests it
79 should do fewer comparisons, so perhaps this should be revisited.
80 List merge sort was implemented in the style of Knuth algorithm
81 5.2.4L, with the optimization suggested by exercise 5.2.4-22. The
82 test case was 140,213,394 bytes, 426,4424 lines, text taken from the
83 GCC 3.3 distribution, sort.c compiled with GCC 2.95.4 and running on
84 Debian 3.0r1 GNU/Linux, 2.4GHz Pentium 4, single pass with no
85 temporary files and plenty of RAM.
87 Since comparisons seem to be the bottleneck, perhaps the best
88 algorithm to try next should be merge insertion. See Knuth section
89 5.3.1, who credits Lester Ford, Jr. and Selmer Johnson, American
90 Mathematical Monthly 66 (1959), 387-389.
92 cp --recursive: perform dir traversals in source and dest hierarchy rather
93 than forming full file names. The latter (current) approach fails
94 unnecessarily when the names become very long.
96 Remove suspicious uses of alloca (ones that may allocate more than
99 Adapt these contribution guidelines for coreutils:
100 http://sources.redhat.com/automake/contribute.html
103 Changes expected to go in, someday.
104 ======================================
106 dd patch from Olivier Delhomme
108 Andreas Gruenbacher's xattr changes
110 Apply Bruno Haible's hostname changes
112 test/mv/*: clean up $other_partition_tmpdir in all cases
114 ls: when both -l and --dereference-command-line-symlink-to-dir are
115 specified, consider whether to let the latter select whether to
116 dereference command line symlinks to directories. Since -l has
117 an implicit --NO-dereference-command-line-symlink-to-dir meaning.
118 Pointed out by Karl Berry.
120 A more efficient version of factor, and possibly one that
121 accepts inputs of size 2^64 and larger.
123 dd: consider adding an option to suppress `bytes/block read/written'
124 output to stderr. Suggested here:
125 http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=165045
127 Pending copyright papers:
128 ------------------------
129 ls --color: Ed Avis' patch to suppress escape sequences for
130 non-highlighted files
132 getpwnam from Bruce Korb
134 pb (progress bar) from Miika Pekkarinen
136 ------------------------------
138 Remove long-deprecated options. Search case-insensitive for
139 `deprecated' and `remove in '. Automate this.
141 Add a distcheck-time test to ensure that every distributed
142 file is either read-only(indicating generated) or is
143 version-controlled and up to date.
145 Implement Ulrich Drepper's suggestion to use getgrouplist rather
146 than getugroups. This affects only `id', but makes a big difference
147 on systems with many users and/or groups, and makes id usable once
148 again on systems where access restrictions make getugroups fail.
149 But first we'll need a run-test (either in an autoconf macro or at
150 run time) to avoid the segfault bug in libc-2.3.2's getgrouplist.
151 In that case, we'd revert to using a new (to-be-written) getgrouplist
152 module that does most of what `id' already does. Or just avoid the
153 buggy use of getgrouplist by never passing it a buffer of length zero.
154 See https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=200327
156 remove `%s' notation:
157 grep -E "\`%.{,4}s'" src/*.c
159 remove or adjust chown's --changes option, since it
160 can't always do what it currently says it does.
162 Adapt tools like wc, tr, fmt, etc. (most of the textutils) to be
163 multibyte aware. The problem is that I want to avoid duplicating
164 significant blocks of logic, yet I also want to incur only minimal
165 (preferably `no') cost when operating in single-byte mode.
167 Remove all uses of the `register' keyword
169 pr's use of nstrftime can make it malloc a very large (up to SIZE_MAX) buffer
171 ls.c: use gettime rather than clock_gettime, gettimeofday, time
175 Copyright (C) 2002-2007 Free Software Foundation, Inc.
177 This program is free software: you can redistribute it and/or modify
178 it under the terms of the GNU General Public License as published by
179 the Free Software Foundation, either version 3 of the License, or
180 (at your option) any later version.
182 This program is distributed in the hope that it will be useful,
183 but WITHOUT ANY WARRANTY; without even the implied warranty of
184 MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
185 GNU General Public License for more details.
187 You should have received a copy of the GNU General Public License
188 along with this program. If not, see <http://www.gnu.org/licenses/>.