Re: Running -Wall -Werror on glib



Maciej Piechotka wrote:

> For example the g_ptr_array_set_size (array, -1) is perfectly legal call
> although the exact behaviour is undefined as I believe. Changing the
> gint to guint would:
> - Change nothing  for the 'non-purist' as the for (int i = ...) ... will
> still work (the cast will be performed. May be a warning be generated -
> but they are 'non-purist').
> - It will help debuging for 'purist' as some errors may be detected.

Changing API like this is unacceptable.

> I have 2 questions:
> - should I work on some cleaning of the glib compilation? I.e. if such
> type of patches will be commited at all or is it a waste of my time.

Such changes are committed regularly.  Please file bugs in bugzilla and attach
small patches.

> - should I 'fix' also a API in such patches, the API is too
> 'standarised' to do it now (if I submit around 2.0 it would be accepted)
> or there are other reson of doing such parts in this specific way?

Yeah, no touching the API.

behdad

> Regards
> PS. Why git repo is  not updated since some time?
> PPS. I'm attaching a small patch for some of the issues:



[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]