Hoooo, if right that's lame ! Probably one the side effects of the
SYSV/Bsd war, and the Streams vs. Socket :-(((
But anyway the effect is the same, the IT were hearing "POSIX compatibility"
and assumed "Unix compatibility on PC" and the poor smuck just got a platform
were select() wasn't working as sson as they needed it. Actually I would like
to see say Oracle source, and count the #ifdef PLATFORM_XYZ, then do a uniq -n
| sort , to get an estimate of the divergence to "Unix" per Os, I also bet
that the Window code is doesn't share much with the original, poor guys ...
One of the strange thing is that I remember the early time of Linux devel.
where most of the apps had to be compiled with the -DSYSV flags (before
autoconf, thanks again Cygnus :-) and Linus expressed the will to go to
full POSIX compatibility, which was believed to be reaching a far better
level of compatibility. I bet that at that time he didn't had bought yet
the POSIX spec :-) . I nether did ... spec ought to be freely available.
I remember all the Linux console login prompt showed POSIX until recently.
> In fact, is select even in POSIX.96? If so, how did they resolve the
Don't know, tried to find some information too on the opengroup pages
but the search for "select" didn't bring back any related pages ...
> possibly-modified-timeout issue?
Basically, it works on Linux but the man pages explicitely says
"Do not rely on it it's not portable", so there is still some crappy
Unices around and until they die we will have to keep that. Honnestly
Linux proves it's quality by being able to grow in the devastated land
left buy the Unix War. We will get rid of them ... too :-)
Daniel
-- Daniel.Veillard@w3.org | W3C MIT/LCS NE43-344 | Today's Bookmarks : Tel : +1 617 253 5884 | 545 Technology Square | Linux, WWW, rpm2html, Fax : +1 617 258 5999 | Cambridge, MA 02139 USA | badminton, Kaffe, http://www.w3.org/People/W3Cpeople.html#Veillard | HTTP-NG and Amaya.