Webware for Python

Jehiah Czebotar schrieb:
> It's a problem in ThreadedAppServer keeping track of the # of bytes
> remaining to be read from the socket, which is only apparent when
> there is some delay in receiving data. (which for me means when webkit
> and apache are on different servers). So that also explains why it was
> intermittent.

Thank you Jehiah, I can confirm your analysis and patch. It's already 
checked in as r7749. This bug slipped in when I added the SCGIAdapater, 
which had a similar problem. Unfortunately, such bugs are never found in 
the beta version, no matter how long the beta phase is :-( I think I'll 
make a bug fix release soon.

-- Christoph

Bugs item #2524767, was opened at 2009-01-20 22:34
Message generated for change (Comment added) made by cito
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=104866&aid=2524767&group_id=4866

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: WebKit
Group: None
Status: Open
Resolution: None
Priority: 5
Private: No
Submitted By: Jehiah Czebotar (jehiah)
Assigned to: Nobody/Anonymous (nobody)
Summary: bug in ThreadedAppServer handling data across multiple packe

Initial Comment:
ThreadedAppServer.py has a bug in the way it keeps track of the number of missing bytes when it reads from a socket. This bug is only apparent when using webkit via mod_webkit2 on a *different* host than apache. (ie: when on localhost all data appears to be available immediately, there needs to be some delay between when the packets arrive).

the count of # of bytes remaining to be read is decremented by all the bytes read previously instead of just the # of bytes read during the current loop

----------------------------------------------------------------------

>Comment By: Christoph Zwerschke (cito)
Date: 2009-01-21 12:19

Message:
Thank you. This has been fixed in r7749.

----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=104866&aid=2524767&group_id=4866