1999-12-05  John Harper  <john@dcs.warwick.ac.uk>
	* rend.c (Imlib_render): stop GC caching leaking memory

1999-10-16  Kjartan Maraas  <kmaraas@online.no>

	* cache.c: (clean_caches): Remove unused variable.
	* colors.c: (PaletteLUTSet): Removed unused variables.
	* utils.c: (Imlib_bevel_image): Removed unused variables.
		
Tue Jun 15 13:08:28 1999  Owen Taylor  <otaylor@redhat.com>

	* misc.c (Imlib_copy_mask): create gc's without
	GraphicsExpose events, as elsewhere in imlib.

1999-02-11  Owen Taylor  <otaylor@redhat.com>

	* misc.c: Don't overwrite existing locale settings, just 
	 temporarily set LC_NUMERIC to "C", then restore it.

1999-02-11  Tomas Ogren  <stric@ing.umu.se>

	* utils.c: Since when did sizeof() be the same as strlen() on char*'s?
	Also changed some malloc() + strcpy() -> strdup().

1999-02-11  Tomas Ogren  <stric@ing.umu.se>

	* Recommend 1.0.3 instead of 1.0.1 (if broken 1.0.2)

Sat Apr 11 12:30:47 1998  George Lebl  <jirka@5z.com>

	* utils.c: fixed yet one more SIGFPE on alpha

Fri Feb 27 09:41:57 1998  Tom Tromey  <tromey@cygnus.com>

	* load.c (Imlib_load_image): Typo fix from Robert Wilhelm.

1998-02-25  Raja R Harinath  <harinath@cs.umn.edu>

	* Makefile.am (DEFS): Define `SYSTEM_IMRC' too.

1998-02-25  Federico Mena Quintero  <federico@nuclecu.unam.mx>

	* misc.c (Imlib_copy_image): The drawable passed to XCreateGC
	should be the dest drawable, not image pixmap.  This caused
	BadMatches all over the place (especially in copy_mask).
	(Imlib_copy_mask): Likewise.

1998-02-24  Raja R Harinath  <harinath@cs.umn.edu>

	* Makefile.am (DEFS): New var.  Moved stuff from CFLAGS.
	(INCLUDES): New var.

1998-02-24  Mark Galassi  <rosalia@cygnus.com>

	* Makefile.am: converted the Makefile.in to a Makefile.am.
	(lib_LTLIBRARIES): 
	(libImlib_la_SOURCES): also converted it to use libtool and
	generate shared libraries.

	* load.c (Imlib_load_image): changed JPEG_PATH to DJPEG_PROG and
	removed %s/djpeg, sine DJPEG_PROG is the full program path.

1998-02-16  Federico Mena Quintero  <federico@nuclecu.unam.mx>

	* misc.c (Imlib_init):  I create a colormap specific to the visual
	the base_window will be using.  This fixes BadMatch errors on
	machines with multiple visual/depth combinations.


1998-02-17 The Rasterman <raster@redhat.com>

        * misc.c (Imlib_init) Fixed visual stuff slightly, and Colormap 
	creation (so it only does so if the visual for the default and the 
	chosen visual mismatch), added function calls to retrieve Imlib's 
	visual and Colormap, plus endianess fixes for network displaying.

1998-02-17 The Rasterman <raster@redhat.com>

	added system imrc config retunr function

1998-02-18 The Rasterman <raster@redhat.com>

        Fixed load.c - missed a not (!) in an if clause
	
1998-03-17 Mandrake <mandrake@mandrake.net>
	* misc.c - boy I hate reading "Imlib Init" all the time. the fprintf
	to stderr is commented out.

1999-03-22  Pavel Machek  <pavel@artax.karlin.mff.cuni.cz>

	* utils.c (Imlib_create_image_from_xpm_data): robustness: fail if
 	id->x.disp is not initialized

1999-03-23  Michael Jennings <mej@eterm.org>

	* utils.c (Imlib_create_image_from_xpm_data): Please keep the gdk
	macros in the gdk version.  g_return_val_if_fail() is not valid
	in Imlib.

1999-05-11  Michael Jennings <mej@eterm.org>

	Added support for specifying a colormap with the ImlibInitParams.

1999-06-16  Michael Jennings <mej@eterm.org>

	Speedups.  Don't recalculate map tables or dirty pixmaps if nothing
	has changed.

1999-06-17  Michael Jennings <mej@eterm.org>

	New function:
	Imlib_bevel_image(ImlibData *, ImlibImage *, ImlibBorder *, unsigned char)

	Draws a bevel onto an image before rendering.  Widths of the bevel edges
	are taken from the ImlibBorder parameter.  The final parameter is numerical,
	true to bevel up, false to bevel down.

1999-06-18  Michael Jennings <mej@eterm.org>

	Another new function, related to yesterday's:
	Imlib_bevel_pixmap(ImlibData *, Pixmap, int, int, ImlibBorder *, unsigned char)

	Draws a bevel onto a pixmap after rendering.  Widths of the bevel edges are
	taken from the ImlibBorder parameter.  The final parameter is numerical,
	true to bevel up, false to bevel down.  The two int parameters are the width
	and height, respectively, of the pixmap (in pixels, of course).

	I haven't tested this in anything but 8 bpp, so I'm hoping someone will tell me
	if it's broken.

