Re: [Gimp-developer] combo boxes in osx
Thanks Sven and Ek I wasn't sure; I've been trying to use macports libraries as much as possible. So I've uninstalled all of the older gtks and x11s installed XQuartz 2.6.0 (xorg-server 1.9.3) then gtk2 /opt/local/bin/pkg-config --modversion gtk+-2.0 says 2.22.1 The problem persists, but I've noticed its not all combo boxes only those within dockable dialogues.The file type dropdown, pixel size and all of the combo boxes in the preferences panes work fine. I then set dyld to print libraries and found that then combo boxes within gtk-demo and pygtk-demo work fine (all started from the X11 xterm) and use dyld: loaded: /opt/local/lib/libgtk-x11-2.0.0.dylib dyld: loaded: /opt/local/lib/libgdk-x11-2.0.0.dylib as does my compiled gimp I'll try to delve into the differences between the combos that work and those that don't All sugestions welcome! Thanks Sean On 10/01/2011 8:52PM, Sven Neumann wrote: On Mon, 2011-01-10 at 17:14 +, s...@startide.org wrote: Hi I've just got gimp to compile from git on OSX, I've just noticed that the combo boxes popup is displaying behind the dialogue panel so they are hidden. Thats every combo box. Are you running GTK+ on X11 or did you compile GTK+ with the Quartz backend? Sven ___ Gimp-developer mailing list Gimp-developer@lists.XCF.Berkeley.EDU https://lists.XCF.Berkeley.EDU/mailman/listinfo/gimp-developer
Re: [Gimp-developer] combo boxes in osx
The tooltips are also displaying in the background,behind the dialogues. Sean ___ Gimp-developer mailing list Gimp-developer@lists.XCF.Berkeley.EDU https://lists.XCF.Berkeley.EDU/mailman/listinfo/gimp-developer
[Gimp-developer] combo boxes in osx
Hi I've just got gimp to compile from git on OSX, I've just noticed that the combo boxes popup is displaying behind the dialogue panel so they are hidden. Thats every combo box. It can be seen if docked in the toolbox then squeezed to the edge. When you hit the combo box the options popup is visible at the side behind the toolbox dialogue. Does anyone else have this problem on OSX or is it just my ruff compile setup? Thanks Sean ___ Gimp-developer mailing list Gimp-developer@lists.XCF.Berkeley.EDU https://lists.XCF.Berkeley.EDU/mailman/listinfo/gimp-developer
[Gimp-developer] luminosity layer mode for 3.0?
I've become a great fan of Dan Margulis and his books on color correction. See Photoshop Professional and Photoshop Lab Color. (BTW, while the books are aimed at PS users, I've found them very good for a basic understanding.) Almost all of his ideas can be implemented in the Gimp, except for his use of the luminosity blending mode in layers. Value blending mode is related, but it really doesn't work as well the luminosity mode does in PS. I can see Gimp calculates luminosity, but I don't know if this is the same luminosity PS calculates, or, for that matter, how it's actually used as a blending mode. In any event, I'd like to put it on the wish list for 3.0 (dare I hope for 2.8?) sean ___ Gimp-developer mailing list Gimp-developer@lists.XCF.Berkeley.EDU https://lists.XCF.Berkeley.EDU/mailman/listinfo/gimp-developer
[Gimp-developer] Re: color balance (preserve luminosity) bug
I'm not working on directly with the Gimp source, but I did fix the
bug with Preserve Luminosity. The offending code in color-balance.c
lines 182-187:
if (cb-preserve_luminosity)
{
gimp_rgb_to_hsl_int (r_n, g_n, b_n);
b_n = gimp_rgb_to_l_int (r, g, b);
gimp_hsl_to_rgb_int (r_n, g_n, b_n);
}
I'm working in Objective-C, so my code isn't copy paste ready.
Basically, if cb- preserve_luminosity, re-scale the slider values so
that the lightness of R1+G1+B1 = the lightness of the base pixel
R+G+B, now use these modified slider values when changing the pixel
values:
-(NSBitmapImageRep *)colorAdjustedImage:(NSBitmapImageRep *)baseImageRep
preserveLumenance:(BOOL)preserveLumenance
colorTone:(LSColorTone)colorTone
{
int w,h,x,y = 0;
unsigned char *srcData, *p1;
int n, red, green, blue;
double shadowScale, midScale, highlightScale;
shadowScale = midScale = highlightScale = 1.0;
if (preserveLumenance == YES) {
shadowScale =1.0 / ( ([_shadowToneRed doubleValue] / 100.0)
+
([_shadowToneGreen doubleValue] / 100.0) + ([_shadowToneBlue
doubleValue] / 100.0) );
midScale = 1.0 / ( ( [_midToneRed doubleValue] / 100.0)
+ (
[_midToneGreen doubleValue] / 100.0) + ( [_midToneBlue doubleValue]
/ 100.0) );
highlightScale = 1.0 / ( ( [_highToneRed doubleValue] / 100.0)
+ (
[_highToneGreen doubleValue] / 100.0) + ( [_highToneBlue doubleValue]
/ 100.0) );
shadowScale = shadowScale -1/3 ? -1/3 : shadowScale;
shadowScale = shadowScale 1/3 ? 1/3 : shadowScale;
shadowScale = shadowScale == 0 ? 1.0 : shadowScale;
midScale = midScale -1/3 ? -1/3 : midScale;
midScale = midScale 1/3 ? 1/3 : midScale;
midScale = midScale == 0 ? 1.0 : midScale;
highlightScale = highlightScale -1/3 ? -1/3 : highlightScale;
highlightScale = highlightScale 1/3 ? 1/3 : highlightScale;
highlightScale = highlightScale == 0 ? 1.0 : highlightScale;
}
// update the tonal range values from any changes in the UI
[self applyToneToColorAdjustment:colorToneMatrix];
// now set the slider values for the different tonal ranges
cyan_red[LSColorToneShadow] = shadowScale * [_shadowToneRed
doubleValue];
magenta_green[LSColorToneShadow] = shadowScale * [_shadowToneGreen
doubleValue];
yellow_blue[LSColorToneShadow] = shadowScale * [_shadowToneBlue
doubleValue];
cyan_red[LSColorToneMidtone] = midScale * [_midToneRed doubleValue];
magenta_green[LSColorToneMidtone] = midScale * [_midToneGreen doubleValue];
yellow_blue[LSColorToneMidtone] = midScale * [_midToneBlue doubleValue];
cyan_red[LSColorToneHighlights] = highlightScale *
[_highToneRed doubleValue];
magenta_green[LSColorToneHighlights] = highlightScale *
[_highToneGreen doubleValue];
yellow_blue[LSColorToneHighlights] = highlightScale *
[_highToneBlue doubleValue];
// create the color lookup tables for the current slider values
and tonal ranges
[self color_balance_create_lookup_tables];
// change the pixels
w = [baseImageRep pixelsWide];
h = [baseImageRep pixelsHigh];
srcData = [baseImageRep bitmapData];
n = [baseImageRep bitsPerPixel] / 8;
for ( y = 0 ; y h ; y++) {
for ( x = 0 ; x w ; x++) {
int r, g, b = 0;
p1 = srcData + n * (y * w + x);
// get the pixels color components
r = p1[0];
g = p1[1];
b = p1[2];
// lookup the new color values in the LUTS contructed from
slider values and tonal selections
red = r_lookup[r];
green = g_lookup[g];
blue = b_lookup[b];
// preserve lumenance if requested to do so
// original code from adapted from the Gimp, which does
not work correctly.
//if (preserveLumenance == YES) {
//gimp_rgb_to_hsl_int (red, green, blue); // red
becomes hue, green becomes saturation, blue becomes lightness
//blue = gimp_rgb_to_l_int (r, g, b); // get the
lightness, set it
//gimp_hsl_to_rgb_int (red, green, blue);
//}
// stomp the pixel component values
p1[0] = red;
p1[1] = green;
p1[2] = blue;
}
}
return baseImageRep;
}
___
Gimp-developer mailing list
Gimp-developer@lists.XCF.Berkeley.EDU
https://lists.XCF.Berkeley.EDU/mailman/listinfo/gimp-developer
[Gimp-developer] color balance (preserve luminosity) bug
I found a bug in Gimp's preserve luminosity feature located in the color balance tool (code is in color-balance.c).The easiest way to see the bug is to turn on preserve luminosity and adjust R G B to any amount that is equal. The result should be no change to the image at all because they should cancel each other out and the brightness should stay the same due to PL.
Any ideas on why the current PL feature isn't working correctly. At a first glance the code method for doing this looks ok... I'm not sure about the function gimp_rgb_to_l_int however. Specifically, I'm not sure if this is the correct formula to find the lightness. Why would it not be:
L = (R + G + B) / 3I tried making this change, but it looks even worse. This code chunk might not be the issue, but it's the first thing that jumped out at me. Any suggestions?- Sean==
snip from color-balance.c==/*** gimp_rgb_to_l_int:* @red: Red channel* @green: Green channel* @blue: Blue channel** Calculates the lightness value of an RGB triplet with the formula
* L = (max(R, G, B) + min (R, G, B)) / 2* * Return value: Luminance vaue corresponding to the input RGB value**/intgimp_rgb_to_l_int (int red, int green, int blue)
{ int min, max; if (red green) { max = MAX (red, blue); min = MIN (green, blue); } else { max = MAX (green, blue); min = MIN (red, blue);
} return ROUND ((max + min) / 2.0);}
___
Gimp-developer mailing list
Gimp-developer@lists.XCF.Berkeley.EDU
https://lists.XCF.Berkeley.EDU/mailman/listinfo/gimp-developer
[Gimp-developer] siod
hello; just thought that some folks might find this of use or interest -- SIOD packages, for versions up to 3.0 or so, in the CMU AI repository: http://www-2.cs.cmu.edu/afs/cs/project/ai-repository/ai/lang/scheme/impl/siod/ SIOD docs, too: http://www.cs.indiana.edu/scheme-repository/imp/siod.html -- sean ___ Gimp-developer mailing list [EMAIL PROTECTED] http://lists.xcf.berkeley.edu/mailman/listinfo/gimp-developer
[Gimp-developer] looking for a script-fu command-line shell
hello; i'm looking for, or to make, something that can be used for sending commands, from BASH or what, directly to a gimp script-fu listener. like: echo '(foo bar etc)' | fu-client -h img.host just wanted to see whether anyone might know of something like this, already written. (else, i'll have to learn to use C streams, heh; need to anyway) incidentally, this came up when looking at gimp-shell.el [1], then trying to make an ilisp 'dialect' definition for SIOD, then one for script-fu good day, all -- sean Footnotes: [1] http://www.geekware.de/software/emacs/ release the hounds! -- someone ___ Gimp-developer mailing list [EMAIL PROTECTED] http://lists.xcf.berkeley.edu/mailman/listinfo/gimp-developer
[Gimp-developer] someone's hairy to-do list
hello; i scratched-together a to-do list for my own GIMP wants wishes, and put it online, thought it might be worth sharing: http://gimbal.paunix.org/cs/img/gimp/todo.html this isn't a wish-list; more like a here's what i plan on doing; of course, it's all open for discussion, completely, with the disclaimer/warning that i can be fairly stubborn on things. I'm not wanting to just barge in on the development efforts, to any effect of wasted work. So, hello. Simply, I know what I want to do with it, and there's a list of the same. Thanks! -- sean ___ Gimp-developer mailing list [EMAIL PROTECTED] http://lists.xcf.berkeley.edu/mailman/listinfo/gimp-developer
