Paul Meurer <[email protected]> writes:

> Hi,
>
>> I'm looking at it. It's puzzling -- this can only happen if
>> update-children was never called for the selector that the navigation
>> is derived from. This is something that should never happen.
>>
>> Paul -- any chance you could provide me with a more complete  
>> backtrace?
>> I'd really like to know who called selector-base-uri and when.
>
> I tried to reproduce the problem, but it magically disappeared! The  
> solution to the puzzle is that I had changed two things in the  
> meanwhile: I used Opera instead of Safari 4 Beta, and I had set  
> hunchentoot::*default-content-type* to "application/xhtml+xml;  
> charset=utf-8". Then it works in Opera, but not in Safari or Firefox.  
> If I set hunchentoot::*default-content-type* back to "text/html; ...",  
> it doesn't work. I get the following trivial-backtrace: (see below).

Thanks, this backtrace explains things very well. But I don't think this
has much to do with *default-content-type*.

The problem is that after you log in, the login widget gets replaced
with a navigation widget in an AJAX request. Then weblocks tries to
render the dirty widgets, which include the navigation widget. However,
the navigation widget does not know what its base-uri is, because it was
never accessed with an URI before. It suddenly got dropped in and tree
shakedown wasn't performed.

I don't know what the correct solution is here. It is not something I
expected when I wrote the new navigation code. I'll keep thinking, in
the meantime any suggestions are welcome.

--J.

--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups 
"weblocks" group.
To post to this group, send email to [email protected]
To unsubscribe from this group, send email to 
[email protected]
For more options, visit this group at 
http://groups.google.com/group/weblocks?hl=en
-~----------~----~----~----~------~----~------~--~---

Reply via email to