CGIHTTPServer, revisited (sorry)

Fred Pacquier fredp at mygale.org.nospam
Fri Apr 12 13:20:28 EDT 2002


Fred Pacquier <fredp at mygale.org.nospam> said :

Of course, just after posting this, I found one of those Google groups
posts I'd overlooked :

http://groups.google.fr/groups?hl=fr&threadm=67abb823.0105301417.6be429ae%40posting.google.com&rnum=28&prev=/groups%3Fq%3DCGIHTTPServer%26hl%3Dfr%26scoring%3Dd%26as_drrb%3Db%26as_mind%3D12%26as_minm%3D4%26as_miny%3D2000%26as_maxd%3D12%26as_maxm%3D4%26as_maxy%3D2002%26start%3D20%26sa%3DN

This describes the exact symptoms (although they are said to be MSIE-
specific, where I also have problems with Netscape 4). There were two
related bugs open of SourceForge, one of them by Kevin Altis :

http://sourceforge.net/tracker/?func=detail&atid=105470&aid=427345&group_id=5470

All of this is almost a year old (May 2001), and still stands in 2.2.1
(the topic seems to irritate Guido a lot, although it does make a few
people like me miserable :-)

BTW, I have tried out the one-line patch to socket.py suggested in one
of the bug follow-ups, and behold, everything now works perfectly (AND
much faster) in both MSIE _and_ Netscape 4... Probably tricky to distribute
an app with a specific socket.py though ? :)
(there is also that 'reading an extra 2 bytes from the rfile before sending 
your response' suggestion, but it's not clear to me where/how to apply it...)

Anybody knows if Kevin's modified CGIHTTPServer.py is available somewhere ?

Thanks,
fp

-- 
YAFAP : http://www.multimania.com/fredp/



More information about the Python-list mailing list