gtg-contributors team mailing list archive

[Lionel Dricot <ploum@xxxxxxxxx>]

Thanks a lot Luca. It's very helpful.

I want to clarify : I've changed the way signals are working in
liblarch.

Before, signals were : added/modified/deleted(node_id, list_of_paths)

now, I've replaced list_of_paths by an individual path. It makes signals
closer of the gtk.TreeModel and of what is happening really in the tree.

It means that, if you remove a node with 3 paths, we will now send 3
signals.

Also, signal modified is used only if the path doesn't change. If path
changes, we send a deleted then a add signal.


-s standard is now working perfectly for me on liblarch_rebased. Not one
single error.

-s bryce still gives me plenty of errors but they are all liblarch
errors. I would be really interested if someone could translate those
errors into tests.


Lionel


Le samedi 14 août 2010 à 02:13 +0200, Luca Invernizzi a écrit :
> On Sat, Aug 14, 2010 at 1:31 AM, Lionel Dricot <ploum@xxxxxxxxx> wrote:
> > This new FT pass all the tests currently (Luca, can you have a look ?
> > There are some crashes in the tests although the report says it's fine).
> > It's fun to rewrite code with test, you feel a lot more confident ;-)
> >
> Copying-n-pasting the commit message
> """
> Testing was giving signal errors, now they're fixed.
>   The errors were caused by the testing function that was waiting for
> more signals
>   than those that were actually emitted (apparently, we're sending
> less signals now).
>   However, the test were passing notwithstanding that, because among the signals
>   that were received there were the ones we were looking for.
> 
>   To be clearer, now, if a signal fails, it also prints out the row of
> the testing
>   function that failed. This way, it's easy to find out which signal is
>   misbehaving.
> """