[Webware-discuss] Re: Session funkiness

[Steve P. <ste...@ya...>]

Chuck Esterbrook wrote:
> I'm having problems with a new regression test suite (for a site, not 
> specifically for WebKit) which ends up making the app server say "bad 
> session id" even though my print statements seem to show that it should be 
> valid.
> 
> At this point I suspect httpsession.py more than I do WebKit for various 
> reasons:
> 
> * I can do the test by hand with IE.
> * httpsession simulates a web browser and therefore keeps cookies and could 
> be passing funkiness back to the web server.
> * All 3 session stores result in the same problem.


Hi Chuck and WebWearers <wink>,

I just recently re-wrote httpsession from scratch. Key improvements include

  * Support for POST requests
  * Support for multipart/form-data POST requests
  * Standardised cookie handling using Tim O'Malley's Cookie module

I've started using the new module a little, but I'm not yet happy enough
with it that I would spend the time releasing or documenting it; I plan to
tidy up a few things in the coming couple of weeks.

When that's done, I will make a release with example code and documentation
as part of the yet-to-be-worked-on WebUnit project on SourceForge
(http://sourceforge.net/projects/webunit).

In the meantime, a preliminary use-at-your-own-risk version is available:

    http://webunit.sourceforge.net/httpsession.py

(Caveat: the above code might not be self-explanatory in its current state)

I'll make an announcement on this list when progress is made.

Best wishes to all,

-Steve

-- 
Steve Purcell, Pythangelist
Get testing at http://pyunit.sourceforge.net/
Any opinions expressed herein are my own and not necessarily those of Yahoo