Would these patches have any (positive) efftect on xft version 2.1.14?
I tried to patch it out if curiousity but hunk 1/10 failed on xftglyphs.c.
Code:
patching file src/xftfreetype.c
Hunk #1 succeeded at 584 (offset 18 lines).
patching file src/xftglyphs.c
Hunk #1 FAILED at 23.
Hunk #2 succeeded at 42 (offset 2 lines).
Hunk #3 succeeded at 134 (offset 2 lines).
Hunk #4 succeeded at 176 (offset 2 lines).
Hunk #5 succeeded at 197 (offset 2 lines).
Hunk #6 succeeded at 276 (offset 2 lines).
Hunk #7 succeeded at 370 (offset 2 lines).
Hunk #8 succeeded at 382 (offset 2 lines).
Hunk #9 succeeded at 400 (offset 2 lines).
Hunk #10 succeeded at 519 (offset 2 lines).
1 out of 10 hunks FAILED -- saving rejects to file src/xftglyphs.c.rej
Here is the reject file:
Code:
--- src/xftglyphs.c 2006-06-03 18:30:56.000000000 +0800
+++ src/xftglyphs.c 2006-06-19 01:15:28.000000000 +0800
@@ -23,8 +23,23 @@
*/
#include "xftint.h"
-#include <freetype/ftoutln.h>
+#include FT_OUTLINE_H
+#if HAVE_FT_GLYPHSLOT_EMBOLDEN
+#include FT_SYNTHESIS_H
+#endif
+
+
+/* define the following macro if you want to use a small FIR filter to
+ * reduce color fringes for LCD rendering. If undefined, the original
+ * weird pixel-local color balancing algorithm will be used
+ */
+#define FIR_FILTER
+#ifdef FIR_FILTER
+/* note: keep the filter symetric, or bad things will happen */
+static const int fir_filter[5] = { 0x10, 0x40, 0x70, 0x40, 0x10 };
+
+#else /* !FIR_FILTER */
static const int filters[3][3] = {
/* red */
#if 0
Bookmarks