678b61a3ac
Bug 742623: Equalize operation fills buffer with nan values in floating point precision
...
Use histogram bins count to process equalization instead of "256" hard-coded value.
2015-02-10 12:08:04 +01:00
e629f4c14b
app: add GimpOperationScalarMultiply
...
which multiplies each float in a gegl_format_n(float, n) input
by a factor.
2014-11-07 00:42:58 +01:00
dbfbc221a5
Bug 736232 - Strange result using transparent layer, blending modes...
...
...and layer opacity
Use a temporary variable to store intermediate result instead of
writting 2 times to the output buffer, because input and output
buffer might be the same.
2014-10-03 00:17:49 +02:00
fe3c0e4020
Bug 732787 - Posterize doesn't work as expected with alpha channel
...
GimpOperationPosterize: also posterize the alpha channel. Seems like a
no-brainer to me, and is actually useful (see bug). The old behavior
is still available by simply disabling the alpha channel in the
channels dialog.
2014-09-25 23:12:35 +02:00
1db5c06f54
Bug 735902 - Drawing a gradient should never produce negative RGB channel values
...
Use MAX(value, 0.0) on the result of dithering.
2014-09-08 22:49:51 +02:00
14978bfe5d
Simplify calls to g_output_stream_write_all()
...
It returns TRUE only if all bytes were written, so there is no need to
get and check the bytes_written return value.
2014-09-07 20:30:14 +02:00
9a245989c0
fix bogus checks against unsigned variables.
...
Spotted by Andrey Karpov using static code analysis:
http://www.viva64.com/en/b/0273/
2014-08-16 00:37:13 +02:00
dbd1f0500e
app: Switch to the GeglAccessMode enum in GimpOperationBlend
...
See b5530e9e3d
2014-07-22 14:09:20 -04:00
8fef1918a9
app: Support gradients with fg/bg segments in the blend tool preview.
2014-07-22 13:13:47 -04:00
5839b5e4af
app: Add a gegl op to render gradients
2014-07-22 13:13:47 -04:00
3f52828896
app: change GimpImageFile and GimpDocumentList to speak GFile
2014-07-07 23:58:11 +02:00
29427fa4f1
app: port GimpImageMapTool settings import/export to streams
...
Simplifies and centralizes error checking and reporting.
2014-07-04 18:31:00 +02:00
b4cdecb4c3
app: port the curves and levels cruft format loaders to GIO
2014-07-04 13:18:01 +02:00
249041d927
Bug 732698 - Select > Shrink/Grow/Border not working in GIMP master
...
Always pass level = 0 to gegl_buffer_set(). This actually makes a
difference since GEGL's level processing got some fixes.
2014-07-03 20:47:02 +02:00
a572638656
app: fix copy/paste GError bug in some calls to g_file_replace()
2014-07-03 13:01:01 +02:00
4e6a6079f4
app: port levels cruft format saving to GIO
2014-07-02 23:45:16 +02:00
87ecc83885
app: port curves cruft format saving to GIO
2014-07-02 23:40:26 +02:00
b5530e9e3d
Use the new GeglAccessMode enum instead of the old values
2014-07-02 02:00:35 +02:00
784e74e1d8
app: remove bogus (but harmless) assignment from GimpOperationColorize
2014-06-30 17:28:04 +02:00
847606d177
app: add back gimp_curves_config_new_spline,explicit()
...
but as versions using gdouble instead of guint8. Use the new functions
from the just renamed _cruft() variants.
2014-06-17 22:23:41 +02:00
f5e4f01c52
app: make gimp_curves_config_new_spline,explicit() handle variable #points
...
Also, append _cruft to the function name because they take guint8*
arrays, will add proper ones next.
2014-06-17 21:10:33 +02:00
0206effba0
Bug 731697 - Colorize operation does not restitute alpha channel correctly
...
Copy the src alpha value in GimpOperationColorize.
2014-06-15 23:39:09 +02:00
a7c82d2964
app: rework gimp:shapeburst towards more floats, try to improve readability.
2014-06-14 02:56:33 +02:00
8aa0019c2e
app: clean up inclusion of gegl-plugin.h, mostly removals
2014-06-10 03:13:09 +02:00
8be52f6aa7
app, libgimp*: pass translated blurbs to GIMP_CONFIG_INSTALL_PROP_FOO()
...
instead of passing N_()-strings; and remove gettext() calls on these
strings when using them. Reduces complexitx, and fixes double- and
untranslated strings. Also enables to treat properties of GIMP and
GEGL objects the same way, which was totally broken before.
2014-05-14 23:34:01 +02:00
23037b5230
app: convert all stock IDs kept around by the core by icon names
...
Particularly GimpViewable's stock_id. Make sure old config files
containing stock IDs are still properly parsed.
2014-05-07 01:01:56 +02:00
d9cf36d07d
app: remove gimp:cast-format and use gegl:cast-format
2014-05-05 19:30:21 +02:00
8f648471e6
app: add new operation gimp:cast-format
...
which casts between two arbitrary formats of the same bpp.
2014-05-04 20:42:58 +02:00
16819917f5
app: add a boolean "normalize" property to gimp:shapeburst
...
and do the normalization in the op, instead of exporting the
max_iterations via a property and doing the normalization outside.
2014-05-01 20:40:55 +02:00
830dbf6d5b
app: remove some disabled code from gimp:shapeburst
2014-05-01 20:33:57 +02:00
540d3bb6ae
app: port gimp:shapeburst's input from u8 to float
2014-05-01 20:32:26 +02:00
3915ac01b4
Bug 728607 - Patch to make divide blend mode work at 32-bit floating point
...
Properly port divide mode to 32-bit float.
2014-04-30 00:16:11 +02:00
22c222291d
libgimpbase,*: clean up enum values in gimpbaseenums.h
...
GIMP_ADD_foo_MASK -> GIMP_ADD_MASK_foo
GIMP_foo_MODE -> GIMP_BLEND_foo
GIMP_foo_CLONE -> GIMP_CLONE_foo
GIMP_foo -> GIMP_DODGE_BURN_TYPE_foo
GIMP_foo -> GIMP_TRANSFER_foo
Add compat values for the old names and compat code to script-fu
and pygimp.
2014-04-29 20:58:30 +02:00
fcecafa479
Bug 723213 - colors/levels/auto is broken for high bitdepth images
...
Port gimp_levels_config_stretch_channel() to histograms with
variable number of bins. Still behaves differently for different
bit depths tho.
2014-01-30 00:19:37 +01:00
84c132addc
Bug 721553 - License text contains obsolete FSF postal address
...
These files were forgotten when changing license from GPL2 to GPL3.
2014-01-05 14:25:46 +01:00
697572ccc0
app,libgimp*: fix includes to follow our include policy
...
- don't include <gdk-pixbuf/gdk-pixbuf.h> in headers in app/
- instead, include it in many .c files instead of <glib-object.h>,
finally acknowledging the fact that app/ depends on gdk-pixbuf almost
globally
- fix up includes as if libgimpbase depended in GIO, which it soon will
2013-10-15 01:58:39 +02:00
ed1b7f4697
app: minor typo spotted on gimp:grow operation. s/diection/direction/
2013-09-16 00:43:44 +12:00
05646749f5
app: port the mask border operation from u8 to float
2013-08-24 19:48:01 +02:00
7ac385caec
app: port the mask shrink and grow operations from u8 to float
2013-08-24 19:08:48 +02:00
fe02ef78ca
Revert "app: enable caching on layer mode ops"
...
This reverts commit 35b5b624e9
2013-08-10 17:44:39 +02:00
35b5b624e9
app: enable caching on layer mode ops
2013-08-10 16:25:50 +02:00
b58e18944a
app: don't set a name or any keys on GimpOperationPointLayerMode
...
it's an abstract base class and cannot be used on its own anyway.
2013-07-31 01:11:28 +02:00
fda356bd10
app: Fix phantom dependency of libappoperation.a
2013-06-29 09:54:49 -07:00
009b5a967a
app: whitespace cleanup in operations/Makefile.am
2013-06-27 09:13:18 +02:00
cccc3550af
SSE2 & SSE4.1 versions of GimpOperationNormalMode
...
* Add configure checks for SSE intrinsic code.
* Use SSE helper libraries in app/operations to compile
code with different CFLAGS.
2013-06-25 20:42:15 -07:00
b10b1ba931
app: Change the opacity param of _pixels to gfloat
...
All the internal math is float, and avoiding the conversion
from double is much faster when using SSE math.
2013-06-19 16:19:34 -07:00
4725529dcb
app: remove forgotten members from struct GimpOperationReplaceMode
2013-06-19 22:12:12 +02:00
77e6ab690d
app: move deref after NULL check
2013-06-16 01:31:14 +02:00
1947d8def8
app: avoid reading uninitialized memory
...
gimp_rgb_to_hsl and hsl_to_rgb reads the a member, so set it before
calling them. Add gimp_hsl_set_alpha() for this purpose.
Found by coverity
2013-06-15 22:40:49 +02:00
3dafeeed15
app: add a generic fast path to GimpOperationPointLayerMode
...
If there is no "aux" buffer, or opacity is 0.0, pass through the input
buffer as-is.
2013-06-12 21:08:04 +02:00
8d1cbd95d2
app: rename gimp_histogram_get_channel() to get_component()
...
Because that's what it is. It also takes a "gint component"
(0..4) and not a "GimpHistogramChannel channel".
2013-06-11 21:32:08 +02:00
a7f42de4c0
app: turn GimpHistorgram into a GimpObject, no other changes
2013-06-11 21:23:32 +02:00
264d09b417
Rename colour and greyscale to color and grayscale respectively
2013-06-06 23:26:16 +02:00
ec786816bb
*/Makefile.am: merge INCLUDES into AM_CPPFLAGS
...
automake-1.13 finally warns about this anachronism.
2013-06-05 20:48:37 +02:00
963275e846
app: use gimp_prop_color_button_new() in GimpColorizeTool
...
which removes the remaining old callbacks. Remove the set/get_color()
API from GimpColorizeConfig and instead add an unserialized "color"
property that stores its value in the other properties.
2013-05-29 00:45:24 +02:00
9072bf331a
app: rename GimpImageMapConfig to GimpSettings
...
It was misnamed from the beginning and has no relation to GimpImageMap
except that it happens to be used by GimpImageMapTools. Now it feels
less weird to potentially use it for other settings too.
2013-05-25 15:26:18 +02:00
a182819420
app: brightness-contrast move a tan call out of the loop
2013-05-21 19:26:43 +02:00
1498b0e972
gimpoperationvaluemode: Unbreak value mode
...
Copy paste failure in my last commit
2013-05-21 07:07:28 -07:00
413a516f8e
Extract layer mode blend functions
...
Extract layer mode blend functions form their operations so
they can be used directly by paintcore.
2013-05-21 04:02:26 -07:00
319738532a
app: fix some oprtation descriptions and make them translatable
...
They are going to be visible in the UI.
2013-05-20 16:42:26 +02:00
6781439fdc
Bug 700653 - Duplicate "(time ...)" values in gimp-curves-tool.settings
...
and gimp-levels-tool.settings
Serialize the channel properties manually (not using
gimp_config_serialize_properties()), so the parent class' "time"
property doesn't end up in the config file once per channel.
2013-05-19 21:44:00 +02:00
77d1b49edc
app: small optimization in gimp_operation_normal_mode_process()
...
Move some calculations out of the R,G,B loop, including a floating
point division, now there is only one division left per pixel.
2013-04-28 23:35:23 +02:00
9de6e41abb
app: remove prepare() implementations from all layer mode operations
...
so the one in GimpOperationPointLayerMode is used which can switch
between RGBA and R'G'B'A.
2013-04-26 23:48:05 +02:00
5322d6a544
app: sense the input format in gimp:mask-components
...
because it can work on RGBA and R'G'B'A without conversions.
2013-04-26 23:29:03 +02:00
c865d8f141
Bug 622054 - Levels Tool gray point picker causes lockup
...
Bail out in gimp_levels_config_adjust_by_colors() if pure
back or white was picked as gray (gamma).
2013-03-27 22:28:29 +01:00
3a0daf540a
app: enable the fast path again in GimpOperationNormalMode
...
but fix it to not take the fast path if opacity is != 1.0 or there is
a mask.
2013-03-11 15:38:28 +01:00
2dc36a9dee
app: add a fast path in GimpOperationMaskComponent
...
Pass through the input buffers nop-style if the mask is either 0 or ALL.
2013-03-11 00:21:23 +01:00
6b0d1038cc
Bug 692641 - Various spelling fixes
2013-01-27 18:59:02 +01:00
9b6703371e
app: remove premultiplied blending code from the few ops that had it
...
Change GimpOperationPointLayerMode's "premultiplied" to "linear" and
set format to "RGBA float" if it's TRUE. Everything defaults to FALSE
so nothing changes.
2012-12-13 22:58:38 +01:00
cf0ac5a57b
Bug 685830 - opacity and mask of bottom layer ignored on Normal Mode
...
Let's prefer correct over broken but fast: disabled broken
optimization for the bottom layer.
2012-10-09 21:52:12 +02:00
d169244955
Bug 143315 - Turning off alpha channel visibility should make alpha=255
...
When the alpha channel is "invisible", make it 1.0 not 0.0.
2012-10-07 22:36:42 +02:00
4cc1084c11
app: allow a NULL "aux" in gimp:mask-components
...
and use black transparency instead, so it can be used for implementing
the image's component visibility.
2012-10-06 23:45:59 +02:00
d0d1962d10
app: include cleanup and s/M_PI/G_PI/ in test-operations
2012-08-03 23:41:29 +02:00
a9179601c9
app: remove g_thread_init() from test-operations.c
2012-07-31 16:47:57 +02:00
a6e1eace45
app: temporarily disable operations testing
2012-06-28 19:07:29 +03:00
1e3db809c2
app: make test-operations build again
2012-06-26 13:36:41 +02:00
ea497fb1ca
app: shortened gegl operations testing makefile
2012-06-14 21:30:11 +03:00
1edb18c30a
app: modified and moved gegl operation testing to operations/tests
2012-06-14 21:30:00 +03:00
e35427071b
app: added test for gegl operations
2012-06-14 21:29:52 +03:00
83680825d0
app: removed unnecessary argument from gimp_operations_init
2012-06-14 21:29:43 +03:00
e13b892c6e
app: fix the (unused) "mask and value" case in GimpOperationSetAlpha
...
and operate in linear RGBA, this op doesn't touch the colors anyway.
2012-06-11 12:41:32 +02:00
e2510b2f98
app: cleaned code of blending modes
2012-05-31 22:22:25 +03:00
8079285c91
app: added gegl version of color erase mode
2012-05-31 22:22:00 +03:00
3456fcb2fc
app: remove inclusion of "paint-funcs/paint-funcs.h"
2012-05-19 11:46:21 +02:00
4699725821
app: port floating selection blending to the new mode ops
...
Also, remove the fallback layer mode code from GimpOperationPointLayerMode,
and pass premultiplied = FALSE to gimp_gegl_node_set_layer_mode().
2012-05-18 23:37:56 +02:00
e377eb01cd
app: modified gegl blending modes to take mask and opacity inputs
2012-05-18 22:52:48 +02:00
cf442d3386
app: added gegl value and behind blending modes
2012-05-11 20:37:47 +03:00
c713e80e63
app: added initial gegl color blending mode
2012-05-11 18:53:49 +03:00
cc0410a226
app: added initial gegl version of saturation blending mode
2012-05-11 18:28:49 +03:00
06382a0037
app: add operations-types.h to Makefile.am
2012-05-11 15:57:31 +02:00
d9ba8b81da
Bug 675825 - (cage) simply using cage transform tool makes gimp to crash.
...
Just check if there is enough handles before closing
2012-05-11 12:02:58 +09:00
8f40eba702
change the pow() workaround to be more HDR safe
2012-05-10 21:41:56 +02:00
7101ee191a
app: move all GEGL operations to new directory app/operations/
2012-05-10 21:22:44 +02:00
Thomas Manni