Re: [gtk-list] Re: [ron@microtronics.com.au (Ron)] gtk1.2.7 can break apps written for 1.2.6

[Havoc Pennington <hp redhat com>]

Drazen Kacar <dave@srce.hr> writes: 
> gint8 is a typedef for something which depends on the environment. In my
> case it's typedef for signed char, while gchar is typedef for char. Even
> in C, char and signed char are distinct types, so compilers will not grok
> the mix in function prototypes (some might, though).
> 

That could break the compilation of GTK itself if gdk_gc_set_dashes
was prototyped and defined differently, but would it ever cause more
than a warning while compiling another app? ("conversion from signed
to unsigned")?

Havoc