Re: TreeModelFilter modify func

[muppet <scott asofyet org>]

On May 2, 2008, at 7:26 PM, Kevin Ryde wrote:

> Is there a trick to giving back the value in a set_modify_func  
> function
> in a TreeModelFilter?  I wondered if it was/is/meant-to-be a return
> value.  Maybe as per below?
>
> < 
> GtkTreeModelFilter 
> .t.set_modify.diff><GtkTreeModelFilter.xs.set_modify.diff>


Hrm.  Yeah, that looks suspect.  We pass in the value, but don't  
marshal it back.  So, while changing the behavior at this point could  
be considered an API break, nobody could really be using this  
successfully at present.

The patch looks okay, and it's only the API stability issues that  
worry me.

Of course, google code search turns up the unit test as the only use  
of set_modify_func() in web-accessible perl code...   http://google.com/codesearch?hl=en&lr=&q=set_modify_func+lang%3Aperl&sbtn=Search


--
I've been using emacs for 20+ years and have barely touched lisp.  I  
wouldn't know lambda calculus if it took all its clothes off and waved  
a placard that reads "I am lambda calculus" in blinking 48-point Comic  
Sans.
  -- Dave Hodgkinson, on london.pm