Re: [Zope] Problem with __before_publishing_traverse__
Wow! You are amazing! Yes, recreating the CookieCrumber just like it was before made it work. Quirky I must admit. Dieter Maurer wrote: Peter Bengtsson wrote at 2007-3-28 10:40 +0100: ... Removing the CookieCrumbler object obviously isn't a good enough solution because we need it for pretty logins. At the moment I don't understand how CookieCrumbler could have anything to do with it. It's stuff it does surrounding the REQUEST looks sane and I don't understand how it could effect my application. A "CookieCumbler" is setting up itself as a (one of the) "__before_publishing_traverse__" hook of its container. Thus, it may cause problems from a __before_publishing_traverse__" call. As you found out, it does not do so normally. Special (and rare) circumstances are necessary. It may be enough to recreate the "CookieCrumbler" instance. -- Peter Bengtsson, work www.fry-it.com home www.peterbe.com hobby www.issuetrackerproduct.com ___ Zope maillist - Zope@zope.org http://mail.zope.org/mailman/listinfo/zope ** No cross posts or HTML encoding! ** (Related lists - http://mail.zope.org/mailman/listinfo/zope-announce http://mail.zope.org/mailman/listinfo/zope-dev )
Re: [Zope] Problem with __before_publishing_traverse__
Peter Bengtsson wrote at 2007-3-28 10:40 +0100: > ... >Removing the CookieCrumbler object obviously isn't a good enough >solution because we need it for pretty logins. At the moment I don't >understand how CookieCrumbler could have anything to do with it. It's >stuff it does surrounding the REQUEST looks sane and I don't understand >how it could effect my application. A "CookieCumbler" is setting up itself as a (one of the) "__before_publishing_traverse__" hook of its container. Thus, it may cause problems from a __before_publishing_traverse__" call. As you found out, it does not do so normally. Special (and rare) circumstances are necessary. It may be enough to recreate the "CookieCrumbler" instance. -- Dieter ___ Zope maillist - Zope@zope.org http://mail.zope.org/mailman/listinfo/zope ** No cross posts or HTML encoding! ** (Related lists - http://mail.zope.org/mailman/listinfo/zope-announce http://mail.zope.org/mailman/listinfo/zope-dev )
Re: [Zope] Problem with __before_publishing_traverse__
Sorry for late reply.
There is no Five in this environment. The "solution" was to remove the
CookieCrumbler instance inside the MyApp instance. CookieCrumber (1.2)
doesn't have a __before_publishing_traverse__ method but it does have
this (CookieCrumbler.py:214)::
def __call__(self, container, req):
'''The __before_publishing_traverse__ hook.'''
resp = self.REQUEST['RESPONSE']
try:
attempt = self.modifyRequest(req, resp)
except CookieCrumblerDisabled:
return
if req.get('disable_cookie_login__', 0):
return
...more stuff I don't understand...
I tried setting up a similar environment (quickly) and a simple
application that uses __before_publishing_traverse__() but could get it
to stop working just because I put in a CookieCrumber instance :(
Removing the CookieCrumbler object obviously isn't a good enough
solution because we need it for pretty logins. At the moment I don't
understand how CookieCrumbler could have anything to do with it. It's
stuff it does surrounding the REQUEST looks sane and I don't understand
how it could effect my application.
Dieter Maurer wrote:
Peter Bengtsson wrote at 2007-3-23 13:29 +:
I have a product that looks like this::
class MyApp(Something):
meta_type = 'x'
def foobar(self):
""" doc str """
return "xxx"
def __before_publishing_traverse__(self, object, request=None):
print "HOUSTON"
...
When I view objects of this class I'd expect Zope to call the
__before_publishing_traverse__ hook and print HOUSTON, but it doesn't :(
...
if object.__class__.__name__ == 'MyApp':
print object.foobar()
print object.foobar
print object.__before_publishing_traverse__(object, self)
print bpth
path = request.path = request['TraversalRequestNameStack']
When I look at the output of this I get:
xxx
>
None
Maybe, instead of "foobar", you should print "__before_publishing_traverse__".
"Five" started to fiddle around with "__before_publishing_traverse__"
(to make Zope 3 traversal available) and maybe this introduced
the bug no longer to call the existing "__before_publishing_traverse__"?
--
Peter Bengtsson,
work www.fry-it.com
home www.peterbe.com
hobby www.issuetrackerproduct.com
___
Zope maillist - Zope@zope.org
http://mail.zope.org/mailman/listinfo/zope
** No cross posts or HTML encoding! **
(Related lists -
http://mail.zope.org/mailman/listinfo/zope-announce
http://mail.zope.org/mailman/listinfo/zope-dev )
Re: [Zope] Problem with __before_publishing_traverse__
Peter Bengtsson wrote at 2007-3-23 13:29 +: >I have a product that looks like this:: > >class MyApp(Something): > meta_type = 'x' > def foobar(self): > """ doc str """ > return "xxx" > def __before_publishing_traverse__(self, object, request=None): > print "HOUSTON" > ... >When I view objects of this class I'd expect Zope to call the >__before_publishing_traverse__ hook and print HOUSTON, but it doesn't :( > ... > if object.__class__.__name__ == 'MyApp': > print object.foobar() > print object.foobar > print object.__before_publishing_traverse__(object, self) > print bpth > path = request.path = request['TraversalRequestNameStack'] > > > >When I look at the output of this I get: > >xxx >> >None Maybe, instead of "foobar", you should print "__before_publishing_traverse__". "Five" started to fiddle around with "__before_publishing_traverse__" (to make Zope 3 traversal available) and maybe this introduced the bug no longer to call the existing "__before_publishing_traverse__"? -- Dieter ___ Zope maillist - Zope@zope.org http://mail.zope.org/mailman/listinfo/zope ** No cross posts or HTML encoding! ** (Related lists - http://mail.zope.org/mailman/listinfo/zope-announce http://mail.zope.org/mailman/listinfo/zope-dev )
[Zope] Problem with __before_publishing_traverse__
I have a product that looks like this:: class MyApp(Something): meta_type = 'x' def foobar(self): """ doc str """ return "xxx" def __before_publishing_traverse__(self, object, request=None): print "HOUSTON" index_html = PageTemplateFile(...) When I view objects of this class I'd expect Zope to call the __before_publishing_traverse__ hook and print HOUSTON, but it doesn't :( I started fiddling with /usr/lib/zope-2.8.8/lib/python/ZPublisher/BaseRequest.py to try to see what's going on. BEFORE (around line 269):: while 1: bpth = getattr(object, '__before_publishing_traverse__', None) if bpth is not None: bpth(object, self) path = request.path = request['TraversalRequestNameStack'] AFTER:: while 1: bpth = getattr(object, '__before_publishing_traverse__', None) if bpth is not None: bpth(object, self) if object.__class__.__name__ == 'MyApp': print object.foobar() print object.foobar print object.__before_publishing_traverse__(object, self) print bpth path = request.path = request['TraversalRequestNameStack'] When I look at the output of this I get: xxx > None This is on Zope 2.8.8 and I've never had problems with __before_publishing_traverse__ in Zope 2.8's before. Could it be that something in Zope 2.8.8 has changed? -- Peter Bengtsson, work www.fry-it.com home www.peterbe.com hobby www.issuetrackerproduct.com ___ Zope maillist - Zope@zope.org http://mail.zope.org/mailman/listinfo/zope ** No cross posts or HTML encoding! ** (Related lists - http://mail.zope.org/mailman/listinfo/zope-announce http://mail.zope.org/mailman/listinfo/zope-dev )
