Subject: Why TZ=EST5EDT fails
To: djgpp AT sun DOT soe DOT clarkson DOT edu
Date: Fri, 26 Aug 1994 11:39:37 +0200 (MET DST)
From: Henrik Storner <storner AT olicom DOT dk>
Cc: dj AT ctron DOT com

Well, thanks to gdb I finally managed to figure out why TZ=EST5EDT (or
any XXXNYYY setting) fails. Not a problem with the library itself, really:
Some support files are missing.

The tzset() function in libsrc/c/gen/ctime.c calls tzparse() which in
turn calls tzload(). tzload() is expected to load the timezone rules
file, from a file named /usr/share/zoneinfo/<timezone-name>. (The
directory is defined in include/tzfile.h, as TZDIR). Few people have a
file called /usr/share/zoneinfo/EST5EDT on a DOS system. so tzload()
fails.

tzparse() - which parses the TZ specification - does all the right things,
computing correct values for the standard offset and DST offset. It then
sees that tzload() failed, and aborts in line 773. Net result: All time
functions use localtime, as if TZ=GMT0.

I guess there isn't much that can be done about this. Distributing
zoneinfo files with DJGPP seems a bit much; picking a default rule to
apply when none is specified in the TZ setting is bound to fail on
some systems - the very reason for the zoneinfo files being that DST
changes are not universally coordinated.

Unfortunate, but I can live with having to use a POSIX TZ setting.
-- 
Henrik Storner        | "Man is the best computer we can put aboard a space-
(storner AT olicom DOT dk)   |  craft ... and the only one that can be mass produced 
System Engineering    |  with unskilled labor."
Olicom Denmark        |                                   Wernher von Braun