Re: Bug in gtkgamma.h

[Tim Janik <timj gtk org>]

On Fri, 17 Sep 1999, Tor Lillqvist wrote:

> Tomi Ollila writes:
>  > Definition GTK_GAMMA_CURVE_CLASS has `gtk_gamma_curve_get_type' instead
>  > of `gtk_gamma_curve_get_type()'.
> 
>  > Header files: 
>  > gtkcolorsel.h gtkcombo.h gtkdrawingarea.h gtkgamma.h gtkhandlebox.h
>  > gtkhbbox.h gtkhpaned.h gtkhruler.h gtkplug.h gtkpreview.h gtkruler.h
>  > gtksocket.h gtkstatusbar.h gtktoolbar.h gtkvbbox.h gtkvpaned.h gtkvruler.h
>  > 
>  > Doesn't declare gtk_.*_get_type() function to return type GtkType
>  > but just `guint'. Not so consistent...
> 
> I committed a fix for these, and some other inconsistencies. It
> shouldn't have any effect on binary compatibility.
> 

which you did with:
> 
> CVSROOT:	/cvs/gnome
> Module name:	gtk+
> Changes by:	tml	99/09/17 14:17:20
> 
> Modified files:
> 	.              : ChangeLog 
> 	gdk            : gdk.c gdk.h 
> 	gdk/win32      : gdk.c gdk.def 
> 	gtk            : gtkclist.c gtkcolorsel.c gtkcolorsel.h 
> 	                 gtkcombo.c gtkcombo.h gtkdrawingarea.c 
> 	                 gtkdrawingarea.h gtkgamma.c gtkgamma.h 
> 	                 gtkhandlebox.c gtkhandlebox.h gtkhpaned.c 
> 	                 gtkhpaned.h gtkhruler.c gtkhruler.h gtkmain.c 
> 	                 gtkmenufactory.c gtknotebook.c gtkplug.c 
> 	                 gtkplug.h gtkpreview.c gtkpreview.h gtkruler.c 
> 	                 gtkruler.h gtksocket.c gtksocket.h 
> 	                 gtkstatusbar.c gtkstatusbar.h gtktoolbar.c 
> 	                 gtktoolbar.h gtkvbbox.c gtkvbbox.h gtkvpaned.c 
> 	                 gtkvpaned.h gtkvruler.c gtkvruler.h gtkwidget.c 
> 	                 makefile.msc 
> 
> Log message:
> 1999-09-17  Tor Lillqvist  <tml@iki.fi>
> 
> * gdk/win32/gdk.def: Add gdk_threads_mutex.
> 
> * gtk/makefile.msc: Correct path to libintl.
> 
> Thanks to Tomi Ollila and Bernd Herd: Fix some inconsistencies in
> use of gint vs. int, and gint vs. gboolean in headers and
> definitions. Use GtkType for the *_get_type functions. Note that
> these changes preserve binary compatibility.
> 
> * gdk/gdk.c
> * gdk/win32/gdk.c
> * gdk/gdk.h: Fix inconsistencies: gint vs int.
> 
> * gtk/gtkmain.c
> * gtk/gtkclist.c
> * gtk/gtkmenufactory.c
> * gtk/gtknotebook.c
> * gtk/gtkwidget.c: Fix inconsistencies, also gint
> vs. gboolean.
> 
> * gtk/gtkcolorsel.[ch]
> * gtk/gtkcombo.[ch]
> * gtk/gtkdrawingarea.[ch]
> * gtk/gtkgamma.[ch]
> * gtk/gtkhandlebox.[ch]
> * gtk/gtkhpaned.[ch]
> * gtk/gtkhruler.[ch]
> * gtk/gtkplug.[ch]
> * gtk/gtkpreview.[ch]
> * gtk/gtkruler.[ch]
> * gtk/gtksocket.[ch]
> * gtk/gtkstatusbar.[ch]
> * gtk/gtktoolbar.[ch]
> * gtk/gtkvbbox.[ch]
> * gtk/gtkvpaned.[ch]
> * gtk/gtkvruler.[ch]: Always use type GtkType for the *_get_type
> functions.
> 
> * gtk/gtkgamma.h: Fix bug, missing () in call of
> gtk_gamma_curve_get_type() in GTK_GAMMA_CURVE_CLASS.
> 



tor, you comitted this stuff to HEAD, while it actually belongs
into the gtk-1-2 branch (no one cares about binary compatibility
in HEAD currently).
changing this kind of stuff in HEAD only does two things:
- duplication of work since we have to commit stuff to -r gtk-1-2
  seperatedly.
- hardens merging of the trees because diffs get cluttered with
  unimportant cruft and CVS is confused because the HEAD version
  already contains unmerged changes from a branch.



> --tml
> 
> 

---
ciaoTJ