platform/upstream/coreutils.git
20 years ago(re_protect): Use `XSTAT (x, ...)' in place of `*(x->xstat) (...)'.
Jim Meyering [Sat, 7 Feb 2004 16:01:20 +0000 (16:01 +0000)]
(re_protect): Use `XSTAT (x, ...)' in place of `*(x->xstat) (...)'.
(do_copy): Declare/use local xstat rather than x->xstat.
(main): Remove code that set x.xstat.

20 years agoRemove xstat function pointer member. The way it was used was not
Jim Meyering [Sat, 7 Feb 2004 16:01:02 +0000 (16:01 +0000)]
Remove xstat function pointer member.  The way it was used was not
portable, since some systems (OSF V5.1, Solaris 2.5.1) provide static
inline `stat' and `lstat' functions, thus making the tests of
`xstat == lstat' in copy.c always fail.

(struct cp_options) [xstat]: Remove member.
(XSTAT): New macro.

20 years agoRemove xstat function pointer member. The way it was used was not
Jim Meyering [Sat, 7 Feb 2004 16:00:28 +0000 (16:00 +0000)]
Remove xstat function pointer member.  The way it was used was not
portable, since some systems (OSF V5.1, Solaris 2.5.1) provide static
inline `stat' and `lstat' functions, thus making the tests of
`xstat == lstat' in copy.c always fail.

(copy_dir): Set `.dereference' member, not .xstat.
(copy_internal): Use `XSTAT (x, ...)' in place of `*(x->xstat) (...)'.
Use `x->dereference == DEREF_NEVER' in place of `x->xstat == lstat'.
(valid_options): Remove now-obsolete FIXME comments.

20 years ago.
Jim Meyering [Sat, 7 Feb 2004 15:32:55 +0000 (15:32 +0000)]
.

20 years ago.
Jim Meyering [Sat, 7 Feb 2004 14:04:23 +0000 (14:04 +0000)]
.

20 years ago.
Jim Meyering [Sat, 7 Feb 2004 13:49:13 +0000 (13:49 +0000)]
.

20 years ago(gnu_ftp_host-alpha, etc.): Un-factor .gnu.org suffix,
Jim Meyering [Sat, 7 Feb 2004 13:49:03 +0000 (13:49 +0000)]
(gnu_ftp_host-alpha, etc.): Un-factor .gnu.org suffix,
so that emit_upload_commands can use these variables, too.

20 years ago(emit_upload_commands): new target
Jim Meyering [Sat, 7 Feb 2004 13:47:41 +0000 (13:47 +0000)]
(emit_upload_commands): new target
(alpha beta major): `$(MAKE)' it.

20 years ago.
Jim Meyering [Sat, 7 Feb 2004 12:04:24 +0000 (12:04 +0000)]
.

20 years ago.
Jim Meyering [Sat, 7 Feb 2004 11:47:01 +0000 (11:47 +0000)]
.

20 years agoadd missing quote
Jim Meyering [Sat, 7 Feb 2004 11:46:00 +0000 (11:46 +0000)]
add missing quote

20 years ago.
Jim Meyering [Sat, 7 Feb 2004 10:05:30 +0000 (10:05 +0000)]
.

20 years ago*** empty log message ***
Jim Meyering [Sat, 7 Feb 2004 09:49:54 +0000 (09:49 +0000)]
*** empty log message ***

20 years ago*** empty log message ***
Jim Meyering [Fri, 6 Feb 2004 16:47:52 +0000 (16:47 +0000)]
*** empty log message ***

20 years agoRemove `du' stack space test.
Jim Meyering [Fri, 6 Feb 2004 16:47:07 +0000 (16:47 +0000)]
Remove `du' stack space test.
Apparently, `ulimit -s N' isn't portable enough.
This test will be restored (with a guard against losing ulimit)
in its own file later.

20 years agoremove stray `done' diagnostic
Jim Meyering [Fri, 6 Feb 2004 16:45:09 +0000 (16:45 +0000)]
remove stray `done' diagnostic

20 years ago*** empty log message ***
Jim Meyering [Fri, 6 Feb 2004 15:26:15 +0000 (15:26 +0000)]
*** empty log message ***

20 years ago(deep): Remove progress-style diagnostics,
Jim Meyering [Fri, 6 Feb 2004 15:26:12 +0000 (15:26 +0000)]
(deep): Remove progress-style diagnostics,
since this test doesn't take long enough to merit them.
Run du on $tmp (the containing dir), not $deep, the full path to leaf.

20 years ago*** empty log message ***
Jim Meyering [Fri, 6 Feb 2004 10:39:01 +0000 (10:39 +0000)]
*** empty log message ***

20 years ago(signatures): Remove definition.
Jim Meyering [Fri, 6 Feb 2004 10:38:57 +0000 (10:38 +0000)]
(signatures): Remove definition.
Now, automake's gnupload handles this.
(%.sig: %): Remove now-unused rule.
(rel-files): Use automake's $(DIST_ARCHIVES), rather than
`$(distdir).tar.bz2 $(distdir).tar.gz'.
(emit-upload-commands): Adjust to use gnupload.

20 years ago(human_time): Correct merge problem.
Jim Meyering [Thu, 5 Feb 2004 13:46:41 +0000 (13:46 +0000)]
(human_time): Correct merge problem.

20 years ago*** empty log message ***
Jim Meyering [Thu, 5 Feb 2004 09:52:30 +0000 (09:52 +0000)]
*** empty log message ***

20 years ago*** empty log message ***
Jim Meyering [Thu, 5 Feb 2004 09:51:54 +0000 (09:51 +0000)]
*** empty log message ***

20 years agoDon't dump core if localtime returns NULL (possible on
Jim Meyering [Thu, 5 Feb 2004 09:51:49 +0000 (09:51 +0000)]
Don't dump core if localtime returns NULL (possible on
hosts with 64-bit time_t and 32-bit int).

Print "??" if the current clock can't
be converted by localtime.  This won't happen until the year
2*31 + 1900, but we don't want to dump core even if the current
clock has the wrong value.

20 years agoDon't dump core if localtime returns NULL (possible on
Jim Meyering [Thu, 5 Feb 2004 09:50:18 +0000 (09:50 +0000)]
Don't dump core if localtime returns NULL (possible on
hosts with 64-bit time_t and 32-bit int).

Include "inttostr.h".
(batch_convert, main):
If time conversion fails, exit with nonzero status.
(show_date): Return int to report conversion failure.
Print the time as an int if localtime fails.

20 years agoInclude "inttostr.h".
Jim Meyering [Thu, 5 Feb 2004 09:47:01 +0000 (09:47 +0000)]
Include "inttostr.h".
(human_time): Print the date/time as a number of seconds since the
epoch if it can't be converted by localtime.  This is better than
just saying "invalid", and is consistent with what "ls" does.
Don't dump core if the year has more than 48 digits; this isn't
possible on any contemporary host, but we might as well do it right.

20 years ago*** empty log message ***
Jim Meyering [Thu, 5 Feb 2004 09:35:02 +0000 (09:35 +0000)]
*** empty log message ***

20 years ago(human_time): Accept time rather than pointer-to-const-time parameter,
Jim Meyering [Thu, 5 Feb 2004 09:34:33 +0000 (09:34 +0000)]
(human_time): Accept time rather than pointer-to-const-time parameter,
for clarity.  All callers changed.

20 years ago(ST_TIME_CMP_NS, ST_TIME_CMP): Define.
Jim Meyering [Thu, 5 Feb 2004 09:28:22 +0000 (09:28 +0000)]
(ST_TIME_CMP_NS, ST_TIME_CMP): Define.
(ATIME_CMP, CTIME_CMP, MTIME_CMP, TIMESPEC_NS): Likewise.
From coreutils' system.h.

20 years ago*** empty log message ***
Jim Meyering [Thu, 5 Feb 2004 09:26:56 +0000 (09:26 +0000)]
*** empty log message ***

20 years ago(ST_TIME_CMP_NS, ST_TIME_CMP): Remove definitions.
Jim Meyering [Thu, 5 Feb 2004 09:26:52 +0000 (09:26 +0000)]
(ST_TIME_CMP_NS, ST_TIME_CMP): Remove definitions.
(ATIME_CMP, CTIME_CMP, MTIME_CMP, TIMESPEC_NS): Likewise.
Now, those are all defined in timespec.h.
Include timespec.h.

20 years agoDon't include timespec.h, now that system.h does it.
Jim Meyering [Thu, 5 Feb 2004 09:24:05 +0000 (09:24 +0000)]
Don't include timespec.h, now that system.h does it.

20 years ago*** empty log message ***
Jim Meyering [Thu, 5 Feb 2004 09:08:52 +0000 (09:08 +0000)]
*** empty log message ***

20 years ago*** empty log message ***
Jim Meyering [Wed, 4 Feb 2004 09:46:10 +0000 (09:46 +0000)]
*** empty log message ***

20 years ago*** empty log message ***
Jim Meyering [Tue, 3 Feb 2004 20:57:40 +0000 (20:57 +0000)]
*** empty log message ***

20 years ago*** empty log message ***
Jim Meyering [Tue, 3 Feb 2004 18:40:30 +0000 (18:40 +0000)]
*** empty log message ***

20 years ago(print_numbers): Use 'double' for loop index, not
Jim Meyering [Tue, 3 Feb 2004 18:40:25 +0000 (18:40 +0000)]
(print_numbers): Use 'double' for loop index, not
'int', to avoid problems with integer overflow.  On almost all
machines 'double' works in every case where 'int' works, and
it works on other cases besides.

20 years ago*** empty log message ***
Jim Meyering [Mon, 2 Feb 2004 17:41:52 +0000 (17:41 +0000)]
*** empty log message ***

20 years ago*** empty log message ***
Jim Meyering [Mon, 2 Feb 2004 17:35:56 +0000 (17:35 +0000)]
*** empty log message ***

20 years ago*** empty log message ***
Jim Meyering [Mon, 2 Feb 2004 17:33:35 +0000 (17:33 +0000)]
*** empty log message ***

20 years ago(do_stat): Remove extra trailing newline from default formats.
Jim Meyering [Mon, 2 Feb 2004 17:33:11 +0000 (17:33 +0000)]
(do_stat): Remove extra trailing newline from default formats.

20 years ago.
Jim Meyering [Mon, 2 Feb 2004 15:00:32 +0000 (15:00 +0000)]
.

20 years ago*** empty log message ***
Jim Meyering [Mon, 2 Feb 2004 13:20:55 +0000 (13:20 +0000)]
*** empty log message ***

20 years ago(nice invocation): Add examples.
Jim Meyering [Mon, 2 Feb 2004 13:20:52 +0000 (13:20 +0000)]
(nice invocation): Add examples.
Prompted by suggestion from Dan Jacobson.
(factor invocation): Add an example.
Update timing numbers for a more modern CPU.

20 years ago*** empty log message ***
Jim Meyering [Mon, 2 Feb 2004 13:12:51 +0000 (13:12 +0000)]
*** empty log message ***

20 years ago*** empty log message ***
Jim Meyering [Mon, 2 Feb 2004 08:08:09 +0000 (08:08 +0000)]
*** empty log message ***

20 years ago*** empty log message ***
Jim Meyering [Mon, 2 Feb 2004 08:05:36 +0000 (08:05 +0000)]
*** empty log message ***

20 years agoPrint actual fractional seconds in time stamps, not just `.00000000'.
Jim Meyering [Mon, 2 Feb 2004 07:59:58 +0000 (07:59 +0000)]
Print actual fractional seconds in time stamps, not just `.00000000'.

(human_time): Add and use new parameter, t_ns.
(print_stat): Update callers.

20 years ago(TIMESPEC_NS): Define here, instead, now that stat.c also uses this macro.
Jim Meyering [Mon, 2 Feb 2004 07:59:35 +0000 (07:59 +0000)]
(TIMESPEC_NS): Define here, instead, now that stat.c also uses this macro.

20 years ago(TIMESPEC_NS): Remove definition.
Jim Meyering [Mon, 2 Feb 2004 07:59:23 +0000 (07:59 +0000)]
(TIMESPEC_NS): Remove definition.

20 years ago.
Jim Meyering [Thu, 29 Jan 2004 20:42:25 +0000 (20:42 +0000)]
.

20 years ago*** empty log message ***
Jim Meyering [Thu, 29 Jan 2004 20:41:22 +0000 (20:41 +0000)]
*** empty log message ***

20 years ago[HAVE_SYS_UCRED_H]: Include grp.h before sys/ucred.h.
Jim Meyering [Thu, 29 Jan 2004 20:41:16 +0000 (20:41 +0000)]
[HAVE_SYS_UCRED_H]: Include grp.h before sys/ucred.h.

20 years ago*** empty log message ***
Jim Meyering [Thu, 29 Jan 2004 20:38:56 +0000 (20:38 +0000)]
*** empty log message ***

20 years agoAvoid a configure-time warning about sys/ucred.h on OSF V4.0.
Jim Meyering [Thu, 29 Jan 2004 20:38:52 +0000 (20:38 +0000)]
Avoid a configure-time warning about sys/ucred.h on OSF V4.0.

Test for sys/ucred.h separately, since
on some systems one must include <grp.h> before it.

20 years agoupdate from automake
Jim Meyering [Wed, 28 Jan 2004 17:16:15 +0000 (17:16 +0000)]
update from automake

20 years ago.
Jim Meyering [Tue, 27 Jan 2004 15:38:00 +0000 (15:38 +0000)]
.

20 years ago*** empty log message ***
Jim Meyering [Tue, 27 Jan 2004 13:09:20 +0000 (13:09 +0000)]
*** empty log message ***

20 years agoReword so as not to exclude the possibility that INCREMENT be zero.
Jim Meyering [Tue, 27 Jan 2004 13:09:16 +0000 (13:09 +0000)]
Reword so as not to exclude the possibility that INCREMENT be zero.

20 years ago*** empty log message ***
Jim Meyering [Tue, 27 Jan 2004 12:53:29 +0000 (12:53 +0000)]
*** empty log message ***

20 years ago Separate `Synopses' section into three examples.
Jim Meyering [Tue, 27 Jan 2004 12:53:24 +0000 (12:53 +0000)]
Separate `Synopses' section into three examples.
Clarify first paragraph.  @w{}-protect an expression.1
Use @option{--option}, rather than @code{--option}.

20 years ago*** empty log message ***
Jim Meyering [Tue, 27 Jan 2004 12:30:46 +0000 (12:30 +0000)]
*** empty log message ***

20 years ago(seq invocation): Remove `@dots{}' at end of synopsis.
Jim Meyering [Tue, 27 Jan 2004 12:30:36 +0000 (12:30 +0000)]
(seq invocation): Remove `@dots{}' at end of synopsis.

20 years ago*** empty log message ***
Jim Meyering [Sun, 25 Jan 2004 22:58:00 +0000 (22:58 +0000)]
*** empty log message ***

20 years agobump version to 5.1.3
Jim Meyering [Sun, 25 Jan 2004 22:57:23 +0000 (22:57 +0000)]
bump version to 5.1.3

20 years agocorrect example
Jim Meyering [Sun, 25 Jan 2004 13:40:09 +0000 (13:40 +0000)]
correct example

20 years ago.
Jim Meyering [Sat, 24 Jan 2004 23:32:11 +0000 (23:32 +0000)]
.

20 years ago*** empty log message *** COREUTILS-5_1_2 v5.1.2
Jim Meyering [Sat, 24 Jan 2004 23:16:21 +0000 (23:16 +0000)]
*** empty log message ***

20 years ago(signatures): Comment out definition.
Jim Meyering [Sat, 24 Jan 2004 23:16:18 +0000 (23:16 +0000)]
(signatures): Comment out definition.

20 years agominor tweaks
Jim Meyering [Fri, 23 Jan 2004 21:55:56 +0000 (21:55 +0000)]
minor tweaks

20 years ago.
Jim Meyering [Fri, 23 Jan 2004 18:48:03 +0000 (18:48 +0000)]
.

20 years ago*** empty log message ***
Jim Meyering [Fri, 23 Jan 2004 15:55:39 +0000 (15:55 +0000)]
*** empty log message ***

20 years ago(header_regexp): Add exitfail.
Jim Meyering [Fri, 23 Jan 2004 15:55:34 +0000 (15:55 +0000)]
(header_regexp): Add exitfail.

20 years ago.
Jim Meyering [Fri, 23 Jan 2004 15:55:13 +0000 (15:55 +0000)]
.

20 years ago*** empty log message ***
Jim Meyering [Fri, 23 Jan 2004 15:55:01 +0000 (15:55 +0000)]
*** empty log message ***

20 years ago(EXTRA_DIST): Add help2man.
Jim Meyering [Fri, 23 Jan 2004 15:54:35 +0000 (15:54 +0000)]
(EXTRA_DIST): Add help2man.

20 years ago*** empty log message ***
Jim Meyering [Fri, 23 Jan 2004 15:43:27 +0000 (15:43 +0000)]
*** empty log message ***

20 years ago*** empty log message ***
Jim Meyering [Fri, 23 Jan 2004 15:41:05 +0000 (15:41 +0000)]
*** empty log message ***

20 years ago(.x.1): Prefix help2man invocation with `$(PERL) --'
Jim Meyering [Fri, 23 Jan 2004 15:41:01 +0000 (15:41 +0000)]
(.x.1): Prefix help2man invocation with `$(PERL) --'
so it works on systems with Perl installed somewhere other than in /usr/bin.

20 years ago(paste_parallel): Declare local, chr, to be of type
Jim Meyering [Fri, 23 Jan 2004 09:26:24 +0000 (09:26 +0000)]
(paste_parallel): Declare local, chr, to be of type
`int', not `char', since it must hold EOF.  This bug would make
paste infloop on some systems.

20 years ago*** empty log message ***
Jim Meyering [Fri, 23 Jan 2004 09:26:12 +0000 (09:26 +0000)]
*** empty log message ***

20 years ago*** empty log message ***
Jim Meyering [Fri, 23 Jan 2004 09:15:58 +0000 (09:15 +0000)]
*** empty log message ***

20 years ago.
Jim Meyering [Thu, 22 Jan 2004 22:19:28 +0000 (22:19 +0000)]
.

20 years ago*** empty log message ***
Jim Meyering [Thu, 22 Jan 2004 22:19:13 +0000 (22:19 +0000)]
*** empty log message ***

20 years ago(expected_failure_status_date): Remove, as 'date' is now normal.
Jim Meyering [Thu, 22 Jan 2004 22:19:04 +0000 (22:19 +0000)]
(expected_failure_status_date): Remove, as 'date' is now normal.
(expected_failure_status_nohup): New var.

20 years ago.
Jim Meyering [Thu, 22 Jan 2004 22:18:36 +0000 (22:18 +0000)]
.

20 years ago(TESTS): Add fail-perm.
Jim Meyering [Thu, 22 Jan 2004 22:18:21 +0000 (22:18 +0000)]
(TESTS): Add fail-perm.

20 years ago*** empty log message ***
Jim Meyering [Thu, 22 Jan 2004 22:18:16 +0000 (22:18 +0000)]
*** empty log message ***

20 years agoTest for just-fixed rmdir bug.
Jim Meyering [Thu, 22 Jan 2004 22:18:12 +0000 (22:18 +0000)]
Test for just-fixed rmdir bug.

20 years ago*** empty log message ***
Jim Meyering [Thu, 22 Jan 2004 22:13:33 +0000 (22:13 +0000)]
*** empty log message ***

20 years ago(Exit status): Document that ordinary failure
Jim Meyering [Thu, 22 Jan 2004 22:13:27 +0000 (22:13 +0000)]
(Exit status): Document that ordinary failure
might not exit with status 1 on unusual platforms.
Mention chroot, env, nice, and su as having unusual exit
status patterns.  Don't bother to mention true and false
since their exit status patterns are actually normal.
(sort invocation, su invocation): Mention its unusual exit
status pattern.
(chroot invocation): Simplify description of exit status 1.
Remove duplicate description of status 127.
(env invocation): Use consistent tenses; simplifiy description
of status 1.
(nice invocation): Likewise.

20 years ago.
Jim Meyering [Thu, 22 Jan 2004 22:08:18 +0000 (22:08 +0000)]
.

20 years ago(usage): Use EXIT_SUCCESS, not 0, for clarity.
Jim Meyering [Thu, 22 Jan 2004 21:09:33 +0000 (21:09 +0000)]
(usage): Use EXIT_SUCCESS, not 0, for clarity.
(main): Use initialize_exit_failure rather than
setting exit_failure directly; this optimizes away redundant assignments.
(TTY_FAILURE, TTY_WRITE_ERROR): New enum values;
substitute them for the corresponding integer constants.

20 years ago(usage): Use EXIT_SUCCESS, not 0, for clarity.
Jim Meyering [Thu, 22 Jan 2004 21:07:40 +0000 (21:07 +0000)]
(usage): Use EXIT_SUCCESS, not 0, for clarity.
(usage): Don't bother normalizing exit status
since the arg is already the correct exit status now.
(FATAL_ERROR, main): Exit with status EXIT_FAILURE, not 2, on errors.

20 years ago(usage): Use EXIT_SUCCESS, not 0, for clarity.
Jim Meyering [Thu, 22 Jan 2004 21:04:26 +0000 (21:04 +0000)]
(usage): Use EXIT_SUCCESS, not 0, for clarity.

(err): Now a boolean int, not a counter,
so that we don't have to worry about int overflow.  All uses changed.

20 years ago(usage): Use EXIT_SUCCESS, not 0, for clarity.
Jim Meyering [Thu, 22 Jan 2004 21:03:13 +0000 (21:03 +0000)]
(usage): Use EXIT_SUCCESS, not 0, for clarity.
(main): Use initialize_exit_failure rather than
setting exit_failure directly; this optimizes away redundant
assignments.

20 years ago(initialize_exit_failure): New inline function.
Jim Meyering [Thu, 22 Jan 2004 21:02:22 +0000 (21:02 +0000)]
(initialize_exit_failure): New inline function.
Include exitfail.h here, since we refer to exit_failure.
All callers changed to not include exitfail.h.

(EXIT_FAIL, EXIT_CANNOT_INVOKE, EXIT_ENOENT): New enum values.

20 years ago(usage): Use EXIT_SUCCESS, not 0, for clarity.
Jim Meyering [Thu, 22 Jan 2004 21:00:25 +0000 (21:00 +0000)]
(usage): Use EXIT_SUCCESS, not 0, for clarity.
(main): Initialize exit_failure to EXIT_FAIL.

(main): Use EXIT_ENOENT and EXIT_CANNOT_INVOKE
rather than roll-your-own symbols or integers.

20 years ago(usage): Use EXIT_SUCCESS, not 0, for clarity.
Jim Meyering [Thu, 22 Jan 2004 20:58:59 +0000 (20:58 +0000)]
(usage): Use EXIT_SUCCESS, not 0, for clarity.
(main): Exit with status 1, not 2, on errors detected by stty proper.

20 years ago(usage): Use EXIT_SUCCESS, not 0, for clarity.
Jim Meyering [Thu, 22 Jan 2004 20:55:26 +0000 (20:55 +0000)]
(usage): Use EXIT_SUCCESS, not 0, for clarity.

(main): Use initialize_exit_failure rather than
setting exit_failure directly; this optimizes away redundant
assignments.

Don't include <assert.h>.
(SORT_OUT_OF_ORDER,  SORT_FAILURE): Now enums, not macros.
(usage): Don't use 'assert'.
(main): Remove redundant assignment to exit_failure.