Re: [Gimp-developer] HSL patch
Hi, On Thu, 2007-01-11 at 21:10 -0500, Robert L Krawitz wrote: Uh, maybe that I didn't look closely enough :-? Here's an updated patch. Thanks. I think it's about time that you open a bug report for this new feature and attach your patch to it. That's still the preferred way of submitting patches. BTW, in the process of preparing this diff, I notice that the tarball contains devel-docs/app/Makefile, which almost surely it shouldn't. The build rules for devel-docs/app are rather difficult as it shouldn't be built by default but still the directory is supposed to be in the tarball. The current solution is far from perfect as it allows glitches to happen like the one you observed. Sven ___ Gimp-developer mailing list Gimp-developer@lists.XCF.Berkeley.EDU https://lists.XCF.Berkeley.EDU/mailman/listinfo/gimp-developer
Re: [Gimp-developer] HSL patch
From: Sven Neumann [EMAIL PROTECTED] Date: Fri, 12 Jan 2007 21:33:01 +0100 On Thu, 2007-01-11 at 21:10 -0500, Robert L Krawitz wrote: Uh, maybe that I didn't look closely enough :-? Here's an updated patch. Thanks. I think it's about time that you open a bug report for this new feature and attach your patch to it. That's still the preferred way of submitting patches. Done (bug 395928). -- Robert Krawitz [EMAIL PROTECTED] Tall Clubs International -- http://www.tall.org/ or 1-888-IM-TALL-2 Member of the League for Programming Freedom -- mail [EMAIL PROTECTED] Project lead for Gutenprint --http://gimp-print.sourceforge.net Linux doesn't dictate how I work, I dictate how Linux works. --Eric Crampton ___ Gimp-developer mailing list Gimp-developer@lists.XCF.Berkeley.EDU https://lists.XCF.Berkeley.EDU/mailman/listinfo/gimp-developer
Re: [Gimp-developer] HSL patch
From: Sven Neumann [EMAIL PROTECTED] Date: Thu, 11 Jan 2007 08:22:20 +0100 On Wed, 2007-01-10 at 06:53 -0500, Robert L Krawitz wrote: From: Sven Neumann [EMAIL PROTECTED] Date: Wed, 10 Jan 2007 08:53:02 +0100 can you explain to me why your patch introduces new functions in libgimpcolor? Couldn't you just use the existing gimp_rgb_to_hsl() and gimp_hsl_to_rgb()? I did it the way I did to use the same implementation as the existing compose_hsv() function. In addition, it isn't clear to me why there should be gimp_rgb_to_hsv4 but not gimp_rgb_to_hsl4. gimp_rgb_to_hsv4() is there for historical reasons. If we could, we would probably remove it from the API. I don't think we want to introduce a HSL variant of it now. I'll try to get to it tonight, although I think it would be useful to have the floating point version also -- if someone's writing a plugin that's doing multiple transformations, it keeps more precision to do the math in floating point vs. with chars. In this case, I suggest that the other floating point variants should be marked deprecated. ___ Gimp-developer mailing list Gimp-developer@lists.XCF.Berkeley.EDU https://lists.XCF.Berkeley.EDU/mailman/listinfo/gimp-developer
Re: [Gimp-developer] HSL patch
Hi, On Thu, 2007-01-11 at 07:23 -0500, Robert L Krawitz wrote: I'll try to get to it tonight, although I think it would be useful to have the floating point version also -- if someone's writing a plugin that's doing multiple transformations, it keeps more precision to do the math in floating point vs. with chars. In this case, I suggest that the other floating point variants should be marked deprecated. But gimp_rgb_to_hsl() and gimp_hsl_to_rgb() are floating point versions. What's your point? Sven ___ Gimp-developer mailing list Gimp-developer@lists.XCF.Berkeley.EDU https://lists.XCF.Berkeley.EDU/mailman/listinfo/gimp-developer
Re: [Gimp-developer] HSL patch
From: Sven Neumann [EMAIL PROTECTED] Date: Wed, 10 Jan 2007 08:53:02 +0100 can you explain to me why your patch introduces new functions in libgimpcolor? Couldn't you just use the existing gimp_rgb_to_hsl() and gimp_hsl_to_rgb()? I did it the way I did to use the same implementation as the existing compose_hsv() function. In addition, it isn't clear to me why there should be gimp_rgb_to_hsv4 but not gimp_rgb_to_hsl4. -- Robert Krawitz [EMAIL PROTECTED] Tall Clubs International -- http://www.tall.org/ or 1-888-IM-TALL-2 Member of the League for Programming Freedom -- mail [EMAIL PROTECTED] Project lead for Gutenprint --http://gimp-print.sourceforge.net Linux doesn't dictate how I work, I dictate how Linux works. --Eric Crampton ___ Gimp-developer mailing list Gimp-developer@lists.XCF.Berkeley.EDU https://lists.XCF.Berkeley.EDU/mailman/listinfo/gimp-developer
Re: [Gimp-developer] HSL patch
Hi, On Wed, 2007-01-10 at 06:53 -0500, Robert L Krawitz wrote: From: Sven Neumann [EMAIL PROTECTED] Date: Wed, 10 Jan 2007 08:53:02 +0100 can you explain to me why your patch introduces new functions in libgimpcolor? Couldn't you just use the existing gimp_rgb_to_hsl() and gimp_hsl_to_rgb()? I did it the way I did to use the same implementation as the existing compose_hsv() function. In addition, it isn't clear to me why there should be gimp_rgb_to_hsv4 but not gimp_rgb_to_hsl4. gimp_rgb_to_hsv4() is there for historical reasons. If we could, we would probably remove it from the API. I don't think we want to introduce a HSL variant of it now. Sven ___ Gimp-developer mailing list Gimp-developer@lists.XCF.Berkeley.EDU https://lists.XCF.Berkeley.EDU/mailman/listinfo/gimp-developer
[Gimp-developer] HSL patch
Unfortunately, there's a little bug in the HSL compose patch, in
gimp_hsl_to_rgb4 in the case of zero saturation. This fixes that.
I'd like to have an option to use luminosity rather than value in
curves and layers; I'm not sure how to do it without introducing
additional complexity.
--- ./libgimpcolor/gimpcolorspace.c~2006-06-27 06:33:48.0 -0400
+++ ./libgimpcolor/gimpcolorspace.c 2007-01-06 14:57:28.0 -0500
@@ -1003,6 +1003,90 @@
}
/**
+ * gimp_rgb_to_hsl4:
+ * @rgb:RGB triplet, rgb[0] is red channel, rgb[1] is green,
+ * rgb[2] is blue (0..255)
+ * @hue:Pointer to hue channel (0..1)
+ * @saturation: Pointer to saturation channel (0..1)
+ * @luma: Pointer to luma channel (0..1)
+ *
+ * Convert an RGB color value to an HSL (Hue, Saturation, Lightness)
+ * color value.
+ *
+ * Since: GIMP 2.4
+ **/
+void
+gimp_rgb_to_hsl4 (const guchar *rgb,
+ gdouble *hue,
+ gdouble *saturation,
+ gdouble *luma)
+{
+ gdouble red, green, blue;
+ gdouble h, s, l;
+ gdouble min, max;
+ gdouble delta;
+
+ red = rgb[0] / 255.0;
+ green = rgb[1] / 255.0;
+ blue = rgb[2] / 255.0;
+
+ h = 0.0; /* Shut up -Wall */
+
+ if (red green)
+{
+ max = MAX (red, blue);
+ min = MIN (green, blue);
+}
+ else
+{
+ max = MAX (green, blue);
+ min = MIN (red, blue);
+}
+
+ l = (max + min) / 2.0;
+
+ if (max == min)
+{
+ s = 0.0;
+ h = GIMP_HSL_UNDEFINED;
+}
+ else
+{
+ if (l = 0.5)
+s = (max - min) / (max + min);
+ else
+s = (max - min) / (2.0 - max - min);
+
+ delta = max - min;
+
+ if (delta == 0.0)
+delta = 1.0;
+
+ if (red == max)
+{
+ h = (green - blue) / delta;
+}
+ else if (green == max)
+{
+ h = 2.0 + (blue - red) / delta;
+}
+ else if (blue == max)
+{
+ h = 4.0 + (red - green) / delta;
+}
+
+ h /= 6.0;
+
+ if (h 0.0)
+h += 1.0;
+}
+
+ *hue= h;
+ *saturation = s;
+ *luma = l;
+}
+
+/**
* gimp_hsv_to_rgb4:
* @rgb:RGB triplet, rgb[0] is red channel, rgb[1] is green,
* rgb[2] is blue (0..255)
@@ -1083,3 +1167,45 @@
rgb[1] = ROUND (saturation * 255.0);
rgb[2] = ROUND (value * 255.0);
}
+
+/**
+ * gimp_hsl_to_rgb4:
+ * @rgb:RGB triplet, rgb[0] is red channel, rgb[1] is green,
+ * rgb[2] is blue (0..255)
+ * @hue:Hue channel (0..1)
+ * @saturation: Saturation channel (0..1)
+ * @luma: Luma channel (0..1)
+ *
+ * Convert an HSL color value (Hue, Saturation, Lightness) to an RGB
+ * color value.
+ *
+ * Since: GIMP 2.4
+ **/
+void
+gimp_hsl_to_rgb4 (guchar *rgb,
+ gdouble hue,
+ gdouble saturation,
+ gdouble luma)
+{
+ if (saturation == 0.0)
+{
+ rgb[0] = ROUND(luma * 255.0);
+ rgb[1] = ROUND(luma * 255.0);
+ rgb[2] = ROUND(luma * 255.0);
+}
+ else
+{
+ gdouble m1, m2;
+
+ if (luma = 0.5)
+m2 = luma * (1.0 + saturation);
+ else
+m2 = luma + saturation - luma * saturation;
+
+ m1 = 2.0 * luma - m2;
+
+ rgb[0] = ROUND (gimp_hsl_value (m1, m2, hue * 6.0 + 2.0) * 255.0);
+ rgb[1] = ROUND (gimp_hsl_value (m1, m2, hue * 6.0) * 255.0);
+ rgb[2] = ROUND (gimp_hsl_value (m1, m2, hue * 6.0 - 2.0) * 255.0);
+}
+}
--- ./libgimpcolor/gimpcolorspace.h~2006-06-27 06:33:48.0 -0400
+++ ./libgimpcolor/gimpcolorspace.h 2007-01-06 13:07:47.0 -0500
@@ -94,6 +94,14 @@
gdouble hue,
gdouble saturation,
gdouble value);
+voidgimp_rgb_to_hsl4(const guchar *rgb,
+ gdouble *hue,
+ gdouble *saturation,
+ gdouble *luma);
+voidgimp_hsl_to_rgb4(guchar *rgb,
+ gdouble hue,
+ gdouble saturation,
+ gdouble luma);
G_END_DECLS
--- ./libgimpcolor/gimpcolor.def~ 2006-09-01 07:14:32.0 -0400
+++ ./libgimpcolor/gimpcolor.def2007-01-06 14:42:27.0 -0500
@@ -18,6 +18,7 @@
gimp_cmyka_set_uchar
gimp_hsl_get_type
gimp_hsl_to_rgb
+ gimp_hsl_to_rgb4
gimp_hsl_to_rgb_int
gimp_hsv_clamp
gimp_hsv_get_type
@@ -55,6 +56,7 @@
gimp_rgb_to_cmyk
gimp_rgb_to_cmyk_int
gimp_rgb_to_hsl
+ gimp_rgb_to_hsl4
gimp_rgb_to_hsl_int
gimp_rgb_to_hsv
gimp_rgb_to_hsv4
--- ./plug-ins/common/decompose.c~ 2006-06-27 06:33:49.0 -0400
+++ ./plug-ins/common/decompose.c 2007-01-06
Re: [Gimp-developer] HSL patch
Hi, can you explain to me why your patch introduces new functions in libgimpcolor? Couldn't you just use the existing gimp_rgb_to_hsl() and gimp_hsl_to_rgb()? Sven ___ Gimp-developer mailing list Gimp-developer@lists.XCF.Berkeley.EDU https://lists.XCF.Berkeley.EDU/mailman/listinfo/gimp-developer
