Commit Graph
26230 Commits
Author SHA1 Message Date
Jim Meyering e080c8a1b0 tests: begin using init.sh
* tests/init.sh: New file.
* tests/Makefile.am (EXTRA_DIST): Add it here.
2010-05-10 11:33:31 +02:00
Jim Meyering 722173444a tests: update help-version
* tests/misc/help-version: Use init.sh, rather than test-lib.sh.
Add idutils' setup.
2010-05-09 21:22:21 +02:00
Jim Meyering 74f07c98d4 tests: move sc_prohibit_always-defined_macros to gnulib's maint.mk
* cfg.mk (sc_prohibit_always-defined_macros): Remove it from here.
* gnulib: Update to latest, for a maint.mk that includes that rule.
2010-05-09 18:54:33 +02:00
Jim Meyering 0f741381c4 tests: rename a syntax-check
* cfg.mk (sc_prohibit_always-defined_macros): Rename from
sc_always_defined_macros.
* .x-sc_prohibit_always-defined_macros: New file, renamed from...
* .x-sc_always_defined_macros: ...removed.
* Makefile.am (syntax_check_exceptions): Rename here, too.
2010-05-09 18:47:34 +02:00
Jim Meyering 6ac165922c tests: loosen/tighten the always_defined_macros check
* cfg.mk (.re-defmac): Generate better regexps: allow white space
before the '#', and append a word-boundary requirement.
Without the latter, #define NULL_DEV ... would evoke a false-positive.
2010-05-09 18:29:22 +02:00
Jim Meyering cab2119542 tests: improve the always_defined_macros check
* cfg.mk (sc_always_defined_macros): Adjust its helpers not to depend
on the existence of ./lib.  Instead, extract symbols directly from
gnulib/lib/*.in.h files.
2010-05-09 12:08:21 +02:00
Jim Meyering 9e42ecceff doc: factor hard-coded project-specific bits from README-release, ...
using the new --mail-headers option to gnulib's announce-gen, and
the updated maint.mk rules to connect the pieces.
* README-release: Remove hard-coded To:, Cc: etc. parts, now
that they're emitted automatically into the announcement template.
* cfg.mk (announcement_Cc_): Override the default.
* gnulib: Update to latest, to get newer announce-gen and maint.mk.
2010-05-03 22:21:31 +02:00
Jim Meyering ca41dcff1d doc: update release procedure
* README-release: Rearrange slightly: post the announcement to
Savannah first, so you can include a link to that post in the email.
2010-05-03 20:37:13 +02:00
Jim Meyering 7d501cdd9c maint: factor trap-related code out of two syntax-check rules
* cfg.mk (gl_trap_): Define, using a loop and eval'd trap,
rather than repeated "trap" uses.  Also handle "13", SIGPIPE.
(sc_always_defined_macros): Use it.
(sc_system_h_headers): Likewise.
2010-05-03 15:49:38 +02:00
Jim Meyering ecf6d16dc7 maint: extend the always_defined_macros syntax-check
* cfg.mk (gl_generated_headers_): Define.
(headers_with_interesting_macro_defs): Remove headers covered
by the above.
(.re-defmac): Extract symbol names from many more files.
(sc_always_defined_macros): Use VC_LIST_EXCEPT, not VC_LIST, so
that we can use the usual exception mechanism.
Test for $(gnulib_dir), not system.h.
* .x-sc_always_defined_macros: New file.  Exempt src/seq.c.
* Makefile.am (syntax_check_exceptions): Add it here.
2010-05-03 13:24:04 +02:00
Jim Meyering c0e784b923 maint: remove now-redundant definitions provided by signal.h
* src/dd.c (SA_NODEFER, SA_RESETHAND): Remove definitions,
now that gnulib guarantees they are defined in <signal.h>.
* src/ls.c (SA_RESTART): Likewise.
2010-05-03 12:26:20 +02:00
Jim Meyering a9c542e416 maint: remove now-redundant definitions provided by sys/wait.h
* src/timeout.c (WIFSIGNALED, WTERMSIG): Remove definitions,
now that gnulib guarantees they are defined in <sys/wait.h>.
* src/operand2sig.c: Likewise.
* src/kill.c: Likewise.
2010-05-03 12:26:00 +02:00
Paul Eggert d861519d2a sort: use long doubles only when effective
* src/sort.c (general_numcompare): Don't use long double if strtold
is not available, as it may introduce needless overhead.
2010-05-01 23:25:39 +01:00
Jim Meyering 5e9a7ae46b build: update gnulib submodule to latest; bootstrap, too 2010-04-30 18:27:19 +02:00
Pádraig Brady 8b5087d4e6 sort: use long doubles for general numeric mode
* src/sort.c (general_numcompare): Use long doubles unconditionally,
and strtold when available, to convert numbers with greater range and
precision.  Performance was seen to be on par with standard doubles.
* doc/coreutils.texi (sort invocation): Amend the -g description to
mention long double rather than double, and strtold rather than strtod.
* src/getlimits.c (main): Output floating point limits for use in tests.
* tests/misc/sort-float: A new test to ensure sort is using long
doubles when possible, and that locale specific floats are handled.
* tests/Makefile.am: Reference the new test.
* tests/test-lib.sh (getlimits_): Normalize indenting.
* NEWS: Mention the new behaviour.
Reported by Nelson Beebe.
2010-04-29 18:55:57 +01:00
Jim Meyering 7905d6d34d build: remove now-duplicate use of AC_SYS_LARGEFILE
* m4/jm-macros.m4 (gl_CHECK_ALL_TYPES): Don't require AC_SYS_LARGEFILE,
now that gnulib does it.
2010-04-29 16:41:22 +02:00
Jim Meyering 518a8664a2 build: remove now-useless configure-time header checks
* m4/jm-macros.m4 (gl_CHECK_ALL_HEADERS): Don't test for headers
that gnulib now provides: sys/ioctl.h, sys/time.h, sys/wait.h.
2010-04-29 16:41:22 +02:00
Jim Meyering 83e4f0ca02 doc: tweak factor-describing wording
* doc/coreutils.texi (factor invocation): Don't say that "factoring
large prime numbers is hard".  A pedant might ding you, since it's
trivial to factor a number that is known to be prime.  Instead, say
that "factoring large numbers... is hard".  Reported by Andreas Eder.
2010-04-25 10:35:51 +02:00
Jim Meyering 2191fe8f8d maint: remove now-unnecessary #if HAVE_header_H tests.
* .x-sc_prohibit_always_true_header_tests: New file.
* Makefile.am (syntax_check_exceptions): Add it.
* src/cat.c: Remove #if HAVE_SYS_IOCTL_H test.
* src/copy.c: Likewise.
* src/ls.c: Likewise.
* src/stty.c: Likewise.
* src/install.c: Remove #if HAVE_SYS_WAIT_H test.
* src/kill.c: Likewise.
* src/operand2sig.c: Likewise.
* src/timeout.c: Likewise.
* src/pathchk.c: Remove #if HAVE_WCHAR_H test.
* src/stat.c: Remove #if HAVE_NETINET_IN_H test.
2010-04-24 17:44:58 +02:00
Jim Meyering a4ab8d9c55 build: enable gnulib modules for more replacement headers
* bootstrap.conf (gnulib_modules): Add the following:
netinet_in, sys_ioctl, sys_wait, so that we can eliminate
the #if HAVE_<header>_H tests guarding their header inclusions.
2010-04-24 17:16:56 +02:00
Jim Meyering 301c74865e tests: clean up also upon SIGQUIT
* tests/test-lib.sh: Also trap on SIGQUIT.
Spotted by Dmitry V. Levin.
2010-04-24 15:50:41 +02:00
Jim Meyering 765d67410c post-release administrivia
* NEWS: Add header line for next release.
* .prev-version: Record previous version.
* cfg.mk (old_NEWS_hash): Auto-update.
2010-04-23 18:36:32 +02:00
Jim Meyering 1d10eb8b1e version 8.5
* NEWS: Record release date.
v8.5
2010-04-23 17:18:16 +02:00
Dmitry V. Levin c0a121c9bc tests: fix exit status of signal handlers in shell scripts
The value of `$?' on entrance to signal handlers in shell scripts
cannot be relied upon, so set the exit code explicitly.

* cfg.mk (sc_always_defined_macros, sc_system_h_headers): Set
the exit code in signal handler explicitly to 128 + SIG<SIGNAL>.
* src/Makefile.am (sc_tight_scope): Likewise.
* tests/test-lib.sh: Likewise.
2010-04-23 15:44:19 +02:00
Eric Blake c9e4ea6ee2 base64: always treat input in binary mode
Necessary for cygwin.  Technically, this patch is not correct,
in that it clobbers O_APPEND, but it is no different than any
other use of xfreopen to force binary mode, so all such uses
should be fixed at once in a later patch.

* src/base64.c (main): Open input in binary mode.
* THANKS: Update.
Reported by Yutaka Amanai.
2010-04-22 08:42:54 -06:00
Jim Meyering 819265b8be build: update gnulib submodule to latest 2010-04-21 20:05:52 +02:00
Eric Blake a9b904a282 docs: document transformation of obsolete sort syntax
* doc/coreutils.texi (sort invocation): Mention the conversion.
2010-04-21 07:59:03 -06:00
Pádraig Brady 7dc398a30f maint: update a couple of NEWS items for the pending release
* NEWS: Mention that cp and mv from the previous release did
not support preserving extended attributes (fixed in e489fd04).
Improve the grammar for the "cp capabilities" item.
2010-04-21 07:55:38 +01:00
Pádraig Brady 8fc12909f6 sort: fix parsing of end field in obsolescent key formats
This regression was introduced in commit 224a69b5, 2009-02-24,
"sort: Fix two bugs with determining the end of field".
The specific regression being that we include 1 field too many when
an end field is specified using obsolescent key syntax (+POS -POS).

* src/sort.c (struct keyfield): Clarify the description of the eword
member, as suggested by Alan Curry.
(main): When processing obsolescent format key specifications,
normalize eword to a zero based count when no specific end char is given
for an end field. This matches what's done when keys are specified with -k.
* tests/misc/sort: Add a few more tests for the obsolescent key formats,
with test 07i being the particular failure addressed by this change.
* THANKS: Add Alan Curry who precisely identified the issue.
* NEWS: Mention the fix.
Reported by Santiago Rodríguez
2010-04-20 22:37:18 +01:00
Pádraig Brady 1777d0dfe3 cp: preserve "capabilities" when also preserving file ownership
* src/copy.c (copy_reg): Copy xattrs _after_ setting file ownership
so that capabilities are not cleared when setting ownership.
* tests/cp/capability: A new root test.
* tests/Makefile.am (root_tests): Reference the new test.
* NEWS: Mention the fix.
2010-04-16 23:02:02 +01:00
Jim Meyering 56fce1aed7 * HACKING (Add tests): Change example name, "newtest" to "new-test".
Not that anyone would confuse with something newt-related, but just
because it is more readable that way.

doc: tweak HACKING
2010-04-16 08:42:47 +02:00
Jim Meyering 36cc6ac787 doc: tweak HACKING
* HACKING (Curly braces): Tweak a sentence.  Filter a few
paragraphs through "fmt".
2010-04-16 08:21:32 +02:00
Pádraig Brady 62dd4b63c5 maint: fix build on platforms that replace strsignal
* src/Makefile.am (kill_LDADD): Add $(LIBTHREAD) so that
we link with the appropriate libraries to provide Thread Local Storage
on platforms that replace strsignal (like AIX for example).

Tested-by: Daniel Richard G. <danielg@teragram.com>
2010-04-16 00:55:43 +01:00
Pádraig Brady 3a1595a3fb tests: avoid spurious failure of ls/color-norm test
* tests/ls/color-norm: Use the "time" output by `ls -l`
to check normal style.  Previously we used the size from `ls -s`,
but the size of "empty" files can vary depending on whether
SELinux is enabled for example.
2010-04-16 00:26:21 +01:00
Jim Meyering f8291d0ec4 doc: document our code formatting policy regarding curly braces
* HACKING (Curly braces: use judiciously): New section.
2010-04-15 18:52:41 +02:00
Jim Meyering 05bee6f116 tests: avoid spurious failure of root-only ls/capability test
* tests/ls/capability: Adjust this test not to expect the no-op escape
sequence that was removed from all other tests by 2010-01-30 commit
5d43617e, "ls --color: don't emit a final no-op escape sequence".
2010-04-14 15:48:31 +02:00
Pádraig Brady 584e38d8b3 cp: treat selinux warnings consistently
* src/copy.c (copy_reg): Suppress SELinux ENOTSUP warnings consistently
between the destination being present or not.  Previously we did
not suppress ENOTSUP messages when the destination was present.
(copy_internal): Use the same ENOTSUP supression method as
copy_reg() even though the issue was not seen in this case.
* tests/cp/cp-a-selinux: Add a test case for the issue and
group the other test cases in the file more coherently.
* tests/cp/cp-mv-enotsup-xattr: Do the same check for xattr
warnings, even though they did not have the issue.
2010-04-13 13:09:50 +01:00
Pádraig Brady c420cfef77 doc: clarify when cp and mv output xattr warnings
The 2010-03-26 commit, 4c38625e, "doc: fix info on cp --preserve..."
was not entirely correct as cp --preserve=all does produce some
xattr warnings.

* src/copy.h: Update and clarify the comments for reduce_diagnostics
and require_preserve_{xattr,context}.
* doc/coreutils.texi (cp invocation): Update the -a and
--preserve=xattr,context options to say when and which
xattr warnings are output.
(mv invocation): Mention that some warnings are output
when preserving xattrs.
2010-04-12 10:06:11 +01:00
Pádraig Brady be9a42e7ac doc: mention that "capabilities" are preserved by cp/mv
* doc/coreutils.texi (cp invocation): Mention that
"capabilities" are preserved when implemented using
extended attributes.
(mv invocation): Mention ACLs etc. are maintained
due to xattrs being copied.
2010-04-11 16:37:40 +01:00
Jie Liu 20a5a60c61 build: tell ./bootstrap to check for xz up-front
* bootstrap.conf (buildreq): Add xz to the list.
2010-04-11 17:09:35 +02:00
Jim Meyering 0ba945b050 tests: more syntax-checks
* gnulib: update to latest
* cfg.mk (sc_prohibit_empty_lines_at_EOF): Remove, now that it's
in gnulib's maint.mk.
(_hv_file): Override the default.
2010-04-11 11:29:00 +02:00
Jim Meyering 8c4b6ba09b maint: new syntax-check rule: prohibit empty lines at EOF
* cfg.mk (detect_empty_lines_at_EOF_): Define.
(sc_prohibit_empty_lines_at_EOF): New rule.
* .x-sc_prohibit_empty_lines_at_EOF: New file.  Exempt pr test inputs.
* Makefile.am (syntax_check_exceptions): Add it.
Pádraig Brady suggested to parse the output of tail -n1.
2010-04-10 20:26:39 +02:00
Mike Frysinger ac6feeb9f2 dircolors: add rxvt-256color and rxvt-unicode256
* src/dircolors.hin: Add them.
2010-04-09 16:28:46 +02:00
Jim Meyering ef5758c328 maint: ftruncate is always available, even without gnulib
Now that even MinGW provides ftruncate, we know that all
reasonable portability targets provide this function.
Remove the workaround code.  We nearly removed the gnulib
module three years ago:
  http://thread.gmane.org/gmane.comp.lib.gnulib.bugs/9203
and it is now officially "obsolete".
* bootstrap.conf (gnulib_modules): Remove ftruncate.
* src/copy.c (copy_reg): Remove use of HAVE_FTRUNCATE and its
no-longer-used workaround code.
* src/truncate.c: Remove a comment about handling missing ftruncate.
2010-04-09 10:50:43 +02:00
James Youngman 0a9302e7f7 doc: make wc --help say how it defines a 'word'
* src/wc.c (usage): Add wc's definition of "word".
2010-04-08 23:53:12 +02:00
Jim Meyering 9e66806e28 doc: adjust a header in announcement email template
* README-release: Use Mail-Followup-To: rather than Reply-To:.
The former works more reliably, at least with Gnus.
2010-04-08 08:45:23 +02:00
Jim Meyering f6b2f46714 tests: add a PATH-sanity-check to help-version
* tests/misc/help-version: Sync from gzip's version.
* tests/check.mk (TESTS_ENVIRONMENT): Export VERSION, as required
for new help-version test.
2010-04-07 11:47:28 +02:00
Jim Meyering f267926f2d maint: fix a masked syntax-check violation
* m4/jm-macros.m4 (ARGMATCH_DIE): Use usage(EXIT_FAILURE), not usage(1).
* .x-sc_prohibit_magic_number_exit: Remove *.m4 exemption that was
masking the above.
2010-04-06 16:02:27 +02:00
Jim Meyering 6f6f599867 build: update gnulib submodule to latest 2010-04-06 16:02:19 +02:00
Jim Meyering 2dc09bf809 build: update gnulib submodule to latest 2010-04-05 09:03:23 +02:00