Wed Jan 5 21:35:14 GMT 2000 Andy Thomas <alt@gimp.org>
* PLUGIN_MAINTAINERS
Added my name to:-
gfig (yes I know its one large file & its UI needs work)
blinds
plugindetails
sel2path
tileit
2000-01-05 Garry R. Osgood <gosgood@idt.net>
*app/undo.c: GimpChannel->bounds_known field on the
image Selection Mask channel was never invalidated when restored
via undo_pop_channel_mod(). Caused PixelRegion iterators calculating
new selection boundaries to use cached image dimensions, thereby
trying to access tiles non-existent in the scaled-down image.
Closes#4840.
2000-01-05 Michael Natterer <mitch@gimp.org>
* app/gimphelp.[ch]: removed most functions. Show the help page in
an idle function to avoid confusion with calling the temporary
help browser procedure in the middle of the "gimp_help" PDB call.
(I beleive this should work, so this is maybe a workaround for
some other bug).
* app/gimpui.[ch]: removed the dialog functions.
* libgimp/Makefile.am
* libgimp/gimpdialog.[ch]
* libgimp/gimphelp.c
* libgimp/gimphelpui.[ch]: new files. Moved most of the help
functions and the dialog constructors to libgimp and libgimpui.
* libgimp/gimp.h: declaration of the "gimp_help*" functions.
* libgimp/gimpui.h: include "gimpdialog.h" and "gimphelpui.h".
* plug-ins/common/grid.c: use the dialog constructor. It's now
possible to show the plugin's help with "F1".
2000-01-05 Michael Natterer <mitch@gimp.org>
* PLUGIN_MAINTAINERS: I'm maintaining the helpbrowser (Sven, I
dared to add your name, too :-)
* app/context_manager.c
* app/flip_tool.[ch]
* app/perspective_tool.[ch]
* app/rotate_tool.[ch]
* app/scale_tool.[ch]
* app/shear_tool.[ch]
* app/transform_core.[ch]
* app/transform_tool.[ch]
* app/tools_cmds.c
* tools/pdbgen/pdb/tools.pdb
- Show the correct help pages in the transform tools' dialogs.
- The transform tool button of the toolbox is now always pressed
if a transform tool is active (not only for "rotate").
- Replaced the transform action (CREATING, HANDLE_1, ...) and the
transform state (INIT, MOTION, ...) #define's with typed enums.
- Changed the return type of the *_recalc functions to "void"
instead of "void *" and the return type of the *_transform
functions to "TileManager *" instead of "void *".
(I probably removed an artefact here because all *_recalc
functions returned "(void *) 1").
- Use gboolean instead of int where appropriate.
- Code cleanup, indentation.
Tue Jan 4 17:19:22 EST 2000 Adrian Likins <adrian@gimp.org>
* PLUGIN_MAINTAINERS: I claim mail.c as mine
* MAINTAINERS: update my entry while I am at it
2000-01-04 Tor Lillqvist <tml@iki.fi>
* libgimp/makefile.{cygwin.msc}: Add gimpcolorspace object.
* libgimp/gimp.def: Add functions from it.
Fixes from Hans Breuer:
* app/datafiles.c: redefine the executable flag for Win32
to _S_IREAD, to get _all_ files from the plug-in dirs as
executables (including scripts)
* app/main.c: Win32-specific changes to allow building Gimp as a
console application, with all its benefits (like inheriting the
console), but hide it if the user doesn't want it. Also, if stdout
goes to a console, give the user a chance to read the help or
version messages. (tml: I am not convinced that it is better to
build gimp as a console application, but let's try it this way for
a while.)
* app/makefile.{cygwin,msc}: Build as console application, and
link with shell32 library.
* app/paint_core.c (paint_core_motion): Pass the value of a call
to the function gimage_active_drawable() to the paint_func,
instead of just passing the address of gimage_active_drawable...
(tml: This code is only called when the TOOL_TRACES_ON_WINDOW flag
is on, and only the clone tool sets that, and the clone tool's
paint_func doesn't use the drawable argument, so this hasn't
caused any trouble.)
* app/plug_in.c: On Win32, to support scripts, use new function
xspawn() instead of _spawnv. Add some more code to properly kill
plug-ins.
* libgimp/color_display.h: Add G_MODULE_EXPORT declarations.
* libgimp/gimpcolorspace.c
* libgimp/gimpcolorspace.h: Prefixed all functions with gimp_
to avoid namespace collisions.
Changed the License in the header to LGPL. If you don't like this,
please remove those files! (But I would like them to stay since this
moving those functions into libgimp is something that should have
happened much earlier.) Nice work, Daniel!
* app/plug_in.c: Redid my latest change here, since it didn't solve
the problem. Now, registering with image_types == NULL means the same
as what "" used to mean: No image is required and the menu entry will
always be sensitive.
The rest are fixes to my last commit.
--Sven
* configure.in: bumped GTK req. to 1.2.6 -- I hadn't updated GTK
or glib in a while and the silent acceptance of out-of-date
versions was making for a lot of annoying bugs. This should be
updated whenever appropriate, not simply ignored until major
release time, people.
2000-01-02 Tor Lillqvist <tml@iki.fi>
* README.win32: Some clarifications.
* app/makefile.{cygwin,msc}
* libgimp/makefile.{cygwin,msc}
* plug-ins/makefile.{cygwin,msc}: Changes corresponding to the GTk+
source reorg. Add new files.
* app/text_tool.c: Remove now unnecessary workaround for Win32
POINTS identifier clash.
2000-01-01 Garry R. Osgood <gosgood@idt.net>
Happy New Year!
* app/disp_callbacks.c
app/disp_callbacks.c: Checked RGB/GRAY/INDEXED product box RE
testing #4708. added gdisplay_update_title() to refresh
window title bar so that a layer-filled display won't
continue its insistence of being empty. Nice fix, Sven.
* Makefile.am: cosmetic fix
* libgimp/color_display.h
* modules/cdisplay_gamma.c: add clone method
* app/color_area.[ch]
* app/gdisplay.[ch]
* app/gdisplay_color.c
* app/gdisplay_color_ui.c: preliminary support for color_area filter
(really just a sync to home machine)
* app/nav_window.c: minor cleanup
* tools/pdbgen/app.pl: initialize gbooleans to FALSE, not 0
* tools/pdbgen/lib.pl: some more arg work
-Yosh
Thu Dec 30 15:21:29 GMT 1999 Adam D. Moss <adam@gimp.org>
* plug-ins/common/animoptimize.c: PDB interface patch
from Andreas Jaekel <jaekel@cablecats.de>
* Makefile.am: added spec file to EXTRA_DIST
* docs/Makefile.am: added undo.txt to EXTRA_DIST
* app/interface.c: removed the DODGY warning, doesn't look dodgy to me
* libgimp/*.h: missed an s/Library/Lesser/
* tools/pdbgen/*: lots of work on the libgimp portion
-Yosh
1999-12-25 Garry R. Osgood <gosgood@idt.net>
Season's Greetings!
* app/clone.c
* app/paint_core.c
* app/paint_core.h
* MAINTAINERS
MAINTAINERS: Updated my entry (it wasn't there ;)
app/paint_core.[ch] supplied new PaintTool states to clone_paint_func() so that
writes of temporary markings made directly to the window are not
clobbered by buffered writes stemming from gdisplay_flush_xxx()
routines. clone_tool_paint_func() has been modified to take advantage
of these new states, retiring bug #2184 in a way that does not change
user interface semantics. There are small additions to the PaintCore
interface that do not affect clientele unaware of added semantics.
These changes are detailed at http://idt.net/~gosgood/gimp-patch/patch03.html.
I'm not sure if I understand how this is supposed to work. At least the
interface is quite confusing and the fact that it does nothing doesn't
really help...
--Sven
* app/menus.c
* app/tools.c
* app/tools.h
* app/toolsF.h: moved the xinput_airbrush before the measure tool,
so it groups nicely with the paint tools. Removed the toolbox_position
from the ToolInfo structure, since it was never used and added some
separators into the Tools menu.
--Sven
when using "Once Forward/Backward" modes. Isn't that what was wanted in the
first place? It feels much more intuitive and useful to me. Please complain
if you don't like this.
--Sven
Wed Dec 22 12:06:22 EST 1999 Gregory McLean <gregm@comstar.net>
* gimp.spec: Final clean up this results in a nice binary package
that should work for all. It will only work on intel archs though
as perl is being stubborn.
* app/gimpdrawable.c: added an assert for name != NULL
* tools/pdbgen/pdb/channel.pdb
* tools/pdbgen/pdb/layer.pdb: sanity check the name parameters
-Yosh
1999-12-19 Michael Natterer <mitch@gimp.org>
* gimprc.in
* app/gimprc.[ch]
* app/preferences_dialog.c: new gimprc variable "help-browser"
which can be either "gimp", or "netscape".
* app/gimphelp.[ch]: depending on the value of help_browser, show
the help page with the webbrowser plugin.
1999-12-19 Michael Natterer <mitch@gimp.org>
* plug-ins/helpbrowser/helpbrowser.c: immediately return from
the wheel mouse callback if no vertical scrollbar is visible.
1999-12-18 Michael Natterer <mitch@gimp.org>
* app/preferences_dialog.c: Added a special help function which
shows the help for the active notebook page.
1999-12-18 Tor Lillqvist <tml@iki.fi>
* plug-ins/gfig/gfig.c: Use g_get_home_dir(). Use
G_DIR_SEPARATOR. Fix a bug I introduced (for Win32) a week ago.
1999-12-17 Michael Natterer <mitch@gimp.org>
* app/app_procs.c
* app/gdisplay_ops.c: Set "allow_grow" to FALSE for "Really Quit?" and
"Changes were made to ... Close Anyway?".
* app/gimpui.c: fixed a compiler warning.
* app/tips_dialog.c: Same as above and did some code cleanup.
* app/colormap_dialog.i.c
* app/colormap_dialog.p.h: Allow dragging a color with mouse2
without changing the active color. Replaced the "index" entry with
a spinbutton. Code & UI cleanup.
Fri Dec 17 20:29:12 GMT 1999 Adam D. Moss <adam@gimp.org>
* app/edit_selection.c: Fixed a couple of bugs with translating
the selection mask (move tool, alt-drag):
- Selection mask was being clipped whilst moved around, not just
at its final resting place.
- Selection mask translation was being performed 'live' like the
opaque moves even though there's simply nothing exciting to see.
Now the process is much faster.
Will remove the edit_selection.c dead-code later if this change
does not cause new trouble.
Sven Neumann