Hanno Schlichting wrote:
> On Sun, Mar 21, 2010 at 12:05 PM, Martin Aspeli
> <optilude+li...@gmail.com> wrote:
>> Hanno Schlichting wrote:
>>> Why would you need to have the event on the request, if all you want
>>> is to set headers? Why not make it an event with the response as the
>>> argument instead?
>> Mainly because that's what all the other IPub* events carry with them.
>> We discovered this omission implementing plone.caching, which only needs
>> the response, but it's not all that unlikely that something else may
>> need the request.
> At the point where you start out writing the actual response
> byte-stream, I think it's too late to do anything useful with the
> But consistency is good, so maybe the different nature of the event
> could be reflected in the event name more clearly.
>> Of course, "something else" could do the same getSite() trick if needed,
>> or use five.globalrequest.
> Yes, please. Unless there's a real need for such a hack, I'd rather
> not see it in Zope2.
Would you be happy with the approach outlined if the event carried just
the response instead?
Author of `Professional Plone Development`, a book for developers who
want to work with Plone. See http://martinaspeli.net/plone-book
Zope-Dev maillist - Zope-Dev@zope.org
** No cross posts or HTML encoding! **
(Related lists -