]> git.ipfire.org Git - thirdparty/postgresql.git/commit
projects / thirdparty / postgresql.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: f517590) | patch
Okay, I've had it with mktime() bugs. While chasing Torello Querci's
authorTom Lane <tgl@sss.pgh.pa.us>
Sat, 13 Sep 2003 21:12:38 +0000 (21:12 +0000)
committerTom Lane <tgl@sss.pgh.pa.us>
Sat, 13 Sep 2003 21:12:38 +0000 (21:12 +0000)
commit94a13b8a117c134a18ae37764c5524874b479db2
tree0473fba0cbde9fc4bdb4fc5fe6e691bcee2073b8tree | snapshot
parentf51759064c78bcc7468ccc1b4d87f68f98bd43a9commit | diff
Okay, I've had it with mktime() bugs.  While chasing Torello Querci's
recent gripe, I discovered not one but two undocumented, undesirable
behaviors of glibc's mktime.  So, stop using it entirely, and always
rely on inversion of localtime() to determine the local time zone.
It's not even very much slower, as it turns out that mktime (at least
in the glibc implementation) also does repeated reverse-conversions.
src/backend/utils/adt/datetime.c diff | blob | blame | history
Mirror of https://git.postgresql.org/git/postgresql.git