summaryrefslogtreecommitdiff
AgeCommit message (Collapse)Author
2007-01-13Put CU_TEST_NAME in the environment for each test run by "make check".Jim Meyering
* Transform all Makefile.am files so that when running "make check", CU_TEST_NAME is set to the name of the test. This is so that when I run valgrind-enabled (--log-file-qualifier=CU_TEST_NAME) "make check" on the entire package it is more convenient to map a leak or error found in a valgrind log file back to the offending test. Use this command: (echo tests/Makefile.am.in; find tests -name Makefile.am) \ |xargs perl -pi -e '/^(\s*)PATH=...VG_PATH_PREFIX/ and ' \ -e 'print $1,q|CU_TEST_NAME=`basename $(abs_srcdir)`,$$tst |,"\\\n"'
2007-01-13Regenerate .cvsignore and .gitignore files.Jim Meyering
2007-01-11Avoid a leak in expr's implementation of the ":" (match) operator.Jim Meyering
* src/expr.c (docolon): Free the regexp buffer using regfree, rather than doing it manually, being careful to set fastmap to NULL first. Free any re_regs.start and .end members, if necessary.
2007-01-11* tests/misc/test-diag: Work also when libc's error functionJim Meyering
reports the entire program name ("../../src/test"), rather than just the final component.
2007-01-11Merge branch 'master' of /cuJim Meyering
2007-01-11trivial mergeJim Meyering
2007-01-10Don't use fts_statp uninitialized for "chown -RLh --preserve-root ...".Jim Meyering
* src/chown-core.c (FTSENT_IS_DIRECTORY): New macro. (change_file_owner): Perform the ROOT_DEV_INO_CHECK only for a directory. Non-directory entries lack fts_statp data when using the FTS_NOSTAT option.
2007-01-07* tests/sample-test: Update copyright date to 2007.Jim Meyering
* Makefile.maint (copyright-check): Also check the copyright date in tests/sample-test.
2007-01-06* tests/fmt/basic (pfx-only, pfx-of-pfx): New tests,Jim Meyering
based on examples from G.P. Halkes in <http://article.gmane.org/gmane.comp.gnu.core-utils.bugs/9388>.
2007-01-06* src/fmt.c (copy_rest): Correct prefix handling.G.P. Halkes
Don't elide a line with the prefix followed by only white space. (get_line): Move EOF-check to loop-termination condition. * tests/fmt/basic (pfx-1): Adjust test to expect desired result. (pfx-2): Remove test; its premise was contrary to the documentation.
2007-01-05Avoid a used-uninitialized bug for invalid input, i.e., when the sizeJim Meyering
of the input, not counting newlines, is 1 % 4. * gl/lib/base64.c (base64_decode): Don't hard-code inlen==4. It may be smaller when flushing.
2007-01-05* src/dircolors.hin: Add a TERM directive for cons25.Mike Frysinger
2007-01-04Use the release year, not the current year.Jim Meyering
* src/groups.sh (version): Use a better name: @RELEASE_YEAR@. * src/Makefile.am (RELEASE_YEAR): Define it. (.sh): Use it. Thanks to a prod from Eric Blake.
2007-01-04Ensure that "group --version" always prints the current year.Jim Meyering
* src/groups.sh (version): Use @CURRENT_YEAR@, rather than 2006. * src/Makefile.am (.sh): Also substitute for @CURRENT_YEAR@. Suggestion from Eric Blake.
2007-01-04* Makefile.cfg (local-checks-to-skip): Skip strftime-check, inJim Meyering
case you don't have convenient access to glibc info documentation.
2007-01-04.cvsignore, .gitignore: Add fchownat.c and wctype-related files.Jim Meyering
2007-01-04[ChangeLog]Jim Meyering
When decoding, always allow newlines in input, with almost no performance impact. * src/base64.c (do_decode): Initialize decode context. Call base64_decode one more time, after all input is processed. (usage): When decoding, newlines are always accepted. * tests/misc/base64: Add a bunch of tests, for the above. * gl/lib/base64.c: Include <string.h>. (base64_decode_ctx_init, get_4, decode_4): New functions. (base64_decode): Efficiently handle interspersed newlines. (base64_decode_alloc): Update signature. * gl/lib/base64.h (struct base64_decode_context): Define. (base64_decode_ctx_init): Add prototype. (base64_decode, base64_decode_alloc): Update prototypes. [doc/ChangeLog] * coreutils.texi (base64 invocation): When decoding, newlines are always accepted.
2007-01-04* gl/lib/base64.c: Copied from gnulib.Jim Meyering
* gl/lib/base64.h: Likewise.
2007-01-03[oops] Document what the ".0" in e.g., "-k 2,3.0" means, and...Evan Hunt
* coreutils.texi (sort invocation): ... that it can be applied to the field-end spec, but not the field-start one. Patch from Evan Hunt. Signed-off-by: Jim Meyering <jim@meyering.net>
2007-01-03Document what the ".0" in e.g., "-k 2,3.0" means, and...Evan Hunt
* coreutils.texi (sort invocation): ... that it can be applied to the field-end spec, but not the field-start one. Patch from Evan Hunt. Signed-off-by: Jim Meyering <jim@meyering.net>
2007-01-03Avoid spurious test failures on MacOS X 10.3.9, in a German locale.Bruno Haible
* tests/chown/deref: Apply lang-default. * tests/misc/split-a: Likewise. * tests/mv/reply-no: Likewise. Signed-off-by: Jim Meyering <jim@meyering.net>
2007-01-03* src/copy.c (copy_internal): Use mkfifo as a fallback if mknod fails.Bruno Haible
Needed on MacOS X. Signed-off-by: Jim Meyering <jim@meyering.net>
2007-01-02Now, "ls -FRL" always follows symbolic links on Linux.Paul Eggert
* NEWS: Mention this bug fix. * src/ls.c (gobble_file): Fix bug reported by Nobuyuki Tsuchimura in http://lists.gnu.org/archive/html/bug-coreutils/2006-12/msg00152.html where "ls -FRL" didn't follow a symbolic link in some cases on Linux. * tests/ls/follow-slink: Add a test for this case. Signed-off-by: Jim Meyering <jim@meyering.net>
2007-01-01* tests/rm/fail-eperm: Instead, simply transform the actual output diagnostic.Jim Meyering
Test failure reported by Bob Proulx.
2007-01-01* tests/rm/fail-eperm: Revert last change. The PATH=... settingJim Meyering
is not honored at least on HP-UX 11.23 systems.
2006-12-30* bootstrap (gnulib_extra_files): Remove announce-gen.Jim Meyering
* bootstrap.conf (gnulib_modules): Add it here instead, now that it's a module.
2006-12-30* tests/misc/base64: Factor a long, repetitive string.Jim Meyering
2006-12-30* src/c99-to-c89.diff: Adjust remove.c offsets.Jim Meyering
2006-12-30Clean up after the change of 2006-12-28.Jim Meyering
* src/remove.c (AD_pop_and_chdir): Change **DIRP parameter to *DIRP, now that this function never modifies the pointer. Adjust comments and code accordingly. (remove_dir): Set "dirp" to NULL right after AD_pop_and_chdir call, now that AD_pop_and_chdir no longer does that.
2006-12-30* tests/rm/fail-eperm: Avoid spurious differences (the error functionJim Meyering
from latest glibc no longer prints the full program_name): so don't invoke rm via ../../src/rm. Instead, invoke it via "PATH=../../src rm".
2006-12-30* tests/mv/acl (skip): Skip this test also if the destinationJim Meyering
directory, which is on a different file system, lacks ACL support.
2006-12-30* src/copy.c (copy_reg): Rewrite a comment that was renderedJim Meyering
inaccurate by the 2006-10-18 change.
2006-12-28When moving "up" the hierarchy, be careful to remove a just-emptiedJim Meyering
directory before opening ".", to avoid trouble with file system implementations that cache readdir results at opendir-time. * src/remove.c (AD_pop_and_chdir): Add a file descriptor parameter. Don't update **DIRP. Don't call fdopendir here. (remove_dir): Call fdopendir here instead. Report and patch from Mikulas Patocka: <http://lists.gnu.org/archive/html/bug-coreutils/2006-12/msg00170.html>
2006-12-28Remove 2006-12-24 entry (the corresponding change was never committed).Jim Meyering
The underlying problem was fixed on 2006-12-26, by requiring a new-enough version of autoconf.
2006-12-27* src/tail.c (usage): Mention +N for --bytes and --lines.Jim Meyering
Suggestion from Evan Hunt.
2006-12-27* configure.ac: Require autoconf-2.61 and automake-1.10.Jim Meyering
Without the former (even with autoconf-2.60), "make distcheck" would fail (without the 2006-09-26 autoconf AC_CHECK_DECL fix), due to an inttypes.h generated with CFLAGS including -pedantic. With the old decl check, @HAVE_DECL_STRTOUMAX@ would be 0.
2006-12-26* Makefile.maint (VC-tag): Define, so as to gpg-sign each releaseJim Meyering
tag, using the release version number as the message. (vc-dist): Use $(VC-tag), rather than "$(VC) tag". 2006-12-24 Jim Meyering <jim@meyering.net> * Makefile.maint (my-distcheck): Don't *configure* with restrictive CFLAGS ('-Werror -ansi -pedantic -Wno-long-long'), since it ends up provoking errors in generation of inttypes.h on an Ubuntu edgy-based system. That leads to compilation failure due to e.g., imaxabs redefinition. Instead, configure with default CFLAGS, and compile the package with restrictive CFLAGS.
2006-12-22* NEWS: dd bs= operands now silently override later ibs= and obs=,Paul Eggert
as POSIX requires. * src/dd.c (scanargs): Implement it. * tests/dd/misc (outbytes): Test it. * doc/coreutils.texi (dd invocation): Specify that bs=N overrides later ibs and obs, undoing part of the previous change. (The behavior was wrong.)
2006-12-21* coreutils.texi (dd invocation): Specify that bs=N overrides only anyJim Meyering
_preceding_ ibs and obs settings. Spotted by Andreas Schwab.
2006-12-21* Makefile.maint (patch-check): Fix shell quoting.Jim Meyering
2006-12-21* coreutils.texi (dd invocation): Add to the description of cbs.Jim Meyering
2006-12-21* coreutils.texi (dd invocation): Improve the documentationJim Meyering
for bs, ibs, obs, and cbs. Suggestion from Dan Jacobson. Patch by Olivier Delhomme.
2006-12-21"rm -rf /etc/motd" (run by non-root) now prints a diagnostic.Jim Meyering
* src/remove.c (remove_entry): Handle EACCES for a non-directory, too. Karl Berry reported that a cross-partition "mv /etc/issue ~" failed with the um,... suboptimal diagnostic, "mv: cannot remove `/etc/issue': Not a directory". * tests/rm/Makefile.am (TESTS): Add fail-eacces. * tests/rm/fail-eacces: New file. * NEWS: Mention that both mv and rm are affected.
2006-12-20"cut -f 2- A B" no longer triggers a double-free bugJim Meyering
* src/cut.c (cut_fields): Set file-scoped global to NULL after freeing it. This avoids a double-free (and core dump on some systems) for this usage: "echo 1>a; echo 2>b; cut -f2- a b". Reported by James Hunt in <http://bugzilla.redhat.com/220312>. * NEWS: List this bug fix. * THANKS: Mention him. * tests/misc/cut: New file. * tests/misc/Makefile.am (TESTS): Add cut.
2006-12-15* tests/cp/open-perm-race: Correct the gdb-existence check.Jim Meyering
Don't run either subsequent gdb command in a sub-shell. Reported by Thomas Schwinge. * THANKS: bring up to date.
2006-12-15Make sure cp -p isn't too generous with file permissions.Paul Eggert
* tests/cp/Makefile.am (TESTS): Add file-perm-race. * tests/cp/file-perm-race: New file. Signed-off-by: Jim Meyering <jim@meyering.net>
2006-12-15Ensure cp -pR --parents isn't too generous with parent permissions.Paul Eggert
* tests/cp/Makefile.am (TESTS): Add parent-perm-race. * tests/cp/parent-perm-race: New file. Signed-off-by: Jim Meyering <jim@meyering.net>
2006-12-14* tests/chgrp/default-no-deref: Don't assume that files are createdJim Meyering
with the primary group by default. That's not true in a directory with the set-GID bit set.
2006-12-14Don't hang when there's no input tty.Jim Meyering
* tests/cp/open-perm-race: Skip this test if there is no controlling input `terminal'.
2006-12-14Test for a hard-to-detect race fix, using gdb.Jim Meyering
* tests/cp/open-perm-race: New file, to test for the cp --preserve=ownership fix of 2006-12-06. * tests/cp/Makefile.am (TESTS_ENVIRONMENT): Define abs_top_builddir. (TESTS): Add open-perm-race.