Le 13/01/11 14:41, Charlie Clark a écrit :
> Am 13.01.2011, 14:01 Uhr, schrieb Godefroid Chapelle<got...@bubblenet.be>:
>>> Aren't you risking double encoding now? That patch looks like it makes
>>> things worse, not better.
>> I am not sure I understand what you mean.
> Hi Godefroid,
> the unpatched code checks to see whether a self.Title() or
> self.Description() are unicode in which case they can be encoded.
I think what you call the unpatched code is my first attempt at fixing
the bug (iow revision 119560).
My first message to the list might not be clear enough to express that I
am actually asking review for both revision 119560 and 119561.
> Your patch will force encoding even on strings, ie. double-encoding.
If I understand well, you tell me that my first attempt could be correct
while the second revision takes the risk to double encode.
Tell me if you agree with patch in 119560.
> Regarding the procedure: changes to the CMF should be tested on CMF
> buildouts, with CMF specific tests.
This is what I did ;-)
I even added the missing buildout (copied from trunk - see revision 119553).
> You should probably run the
> exportimport with pdb to see what self.Title() is returning but, yes, I
> suspect the problem you are experiencing is Plone and not CMF related.
I am still not convinced.
Godefroid Chapelle (aka __gotcha) http://bubblenet.be
Zope-CMF maillist - Zope-CMF@zope.org
See https://bugs.launchpad.net/zope-cmf/ for bug reports and feature requests