summaryrefslogtreecommitdiff
path: root/lib
AgeCommit message (Collapse)Author
2005-09-23*** empty log message ***Jim Meyering
2005-09-23Include <config.h>. Otherwise builds with strictJim Meyering
C89 (e.g., Sun's /opt/SUNWspro/bin/c89) would fail due to the use of `inline'.
2005-09-23* lib/getaddrinfo.c [HAVE_NETINET_IN_H]: Include <netinet/in.h>.Paul Eggert
Problem reported by Eric Blake. (getaddrinfo): Initialize se so that it's not garbage. Redo internal storage allocation so that it doesn't make unportable assumptions about alignment. Fix a memory leak.
2005-09-22.Jim Meyering
2005-09-22Regenerate.Paul Eggert
2005-09-22Sync from gnulib.Paul Eggert
2005-09-22(libcoreutils_a_SOURCES): Add getaddrinfo.h,Paul Eggert
mbchar.c, mbchar.h, mbuiter.h, strcase.h, strlen1.c, strnlen1.h, strstr.h.
2005-09-22Nontrivial sync from gnulib.Paul Eggert
2005-09-22Add getline.h, getpass.cPaul Eggert
2005-09-22(GL_CONCAT): Remove.Paul Eggert
Include "verify.h" instead. (ARGMATCH_CONSTRAINT, ARGMATCH_ASSERT): Remove. (ARGMATCH_VERIFY): Rewrite in terms of new verify macros.
2005-09-22(VERIFY_EXPR, X2NREALLOC, X2REALLOC): Remove.Paul Eggert
The latter two macros are moving to ../src/system.h for now. This reverts this file to the gnulib version.
2005-09-22Remove; no longer used.Paul Eggert
2005-09-22New file, from gnulib.Paul Eggert
2005-09-22Add getaddrinfo.h, getdelim.h, mbchar.h,Paul Eggert
mbuiter.h, strcase.h, strnlen.h, strnlen1.h.
2005-09-22Sync from gnulib.Paul Eggert
2005-09-22Sort.Paul Eggert
2005-09-22(ENOSYS): Define to EEXIST if not defined.Paul Eggert
(make_dir_parents): Treat ENOSYS like EEXIST.
2005-09-21.Jim Meyering
2005-09-20*** empty log message ***Jim Meyering
2005-09-20(fdopendir): Be sure to close the suppliedJim Meyering
file descriptor before returning. This makes our replacement implementation a little closer to Solaris's, where fdopendir ties the file descriptor to the returned DIR* pointer.
2005-09-19*** empty log message ***Jim Meyering
2005-09-19(unlinkat): New function.Jim Meyering
2005-09-19(unlinkat): Add prototype.Jim Meyering
2005-09-17(set_stat_atime, set_stat_ctime, set_stat_mtime): Remove; they werePaul Eggert
buggy in the HAVE_STRUCT_STAT_ST_SPARE1 case, and they were dubious anyway. I rewrote 'tar' to not use them any more.
2005-09-16Regenerate.Paul Eggert
2005-09-16Import file time stamp resolution improvements from gnulib.Paul Eggert
2005-09-16Include stat-time.h.Paul Eggert
(SYSCALL_RESOLUTION): Depend on whether various struct stat members exist, not on the obsolescent ST_MTIM_NSEC. (utimecmp): Use the new stat-time functions rater than TIMESPEC_NS.
2005-09-16(ST_TIME_CMP_NS, ST_TIME_CMP, ATIME_CMP, CTIME_CMP):Paul Eggert
(MTIME_CMP, TIMESPEC_NS): Remove. Now done by stat-time.h, in a different way. (timespec_cmp): New function.
2005-09-16New file.Paul Eggert
2005-09-16Add stat-time.h.Paul Eggert
2005-09-16*** empty log message ***Jim Meyering
2005-09-16(libcoreutils_a_SOURCES): Add fprintftime.c and fprintftime.h.Jim Meyering
2005-09-16*** empty log message ***Jim Meyering
2005-09-16[FPRINTFTIME] (fprintftime): Provide a new interface:Jim Meyering
size_t fprintftime (FILE *fp, char const *fmt, struct tm const *tm, int utc, int nanoseconds); Background: date should not have to allocate a megabyte of virtual memory to handle a format argument like +%1048575T. When implemented with strftime, it must allocate such a buffer, use strftime to fill it in, print it, then free it. With fprintftime, it simply prints everything and exits. With no need for memory allocation, that's one fewer way to fail.
2005-09-16*** empty log message ***Jim Meyering
2005-09-15Rewrite previous strftime change slightly.Paul Eggert
2005-09-15(my_strftime): Rewrite the previous change slightly,Paul Eggert
to make it a bit faster and (I hope) clearer.
2005-09-14*** empty log message ***Jim Meyering
2005-09-14(my_strftime): Be sure to use L_('x') for literals.Jim Meyering
2005-09-14*** empty log message ***Jim Meyering
2005-09-14(my_strftime): Parse the colons of %:::z *after* theJim Meyering
optional field width, not before, so we accept %9:z, not %:9z.
2005-09-14(my_strftime): Fix typo in octal number introduced in last change.Paul Eggert
2005-09-13Merge md5 from libc, and clean up some sha1 glitches.Paul Eggert
2005-09-13(SWAP): Renamed from NOTSWAP, to avoid a horrible misnaming.Paul Eggert
All uses changed. Remove the old SWAP. Remove obvious comment about BLOCKSIZE. (rol): New macro, moved here from md5.h. (sha1_process_block): Remove an incoherent FIXME comment.
2005-09-13(struct sha1_ctx): Mark buffer as being aligned.Paul Eggert
2005-09-13(CYCLIC): New macro. All uses of rol changed to use CYCLIC.Paul Eggert
Redo comment to minimize changes from libc.
2005-09-13(__GNUC_PREREQ, __THROW, __attribute__): New macros.Paul Eggert
(__md5_buffer): Renamed from md5_buffer. Add a macro undoing this if _LIBC is not defined. Add __THROW. (__md5_finish_ctx, __md5_init_ctx, __md5_process_block): (__md5_process_bytes, __md5_read_ctx, __md5_stream): Likewise. (struct md5_ctx): Mark buffer as being aligned. (rol): Remove; all uses changed to CYCLIC.
2005-09-13Add date --rfc-3339.Paul Eggert
2005-09-13(my_strftime): Add support for %:z, %::z, %:::z.Paul Eggert
Fix bug in formats like %2N.
2005-09-13tweak commentJim Meyering