first working version of uuterm!

- at this point it is purely experimental. only ascii characters are
  visible (using builtin font) although all characters are processed.
- there are known bugs, including crashes.
- there are major missing features.
- but it works. ^_^

1 files changed, 196 insertions, 0 deletions

diff --git a/comb.c b/comb.c
new file mode 100644
index 0000000..c1d6ea3
--- /dev/null
+++ b/comb.c
@@ -0,0 +1,196 @@
+/* uuterm, Copyright (C) 2006 Rich Felker; licensed under GNU GPL v2 only */
+
+#define R(a,b) { (a), (b)-(a) }
+
+static const unsigned short common[][2] = {
+	R( 0x300, 0x341 ),
+	R( 0x346, 0x362 ),
+	R( 0x200B, 0x200F ),
+	R( 0x202A, 0x202E ),
+	R( 0x2060, 0x206F ),
+	R( 0x20D0, 0x20EA ),
+	{ 0, 0 }
+};
+
+static const unsigned short latin[][2] = {
+	R( 0x363, 0x36F ),
+	{ 0, 0 }
+};
+
+static const unsigned short greek[][2] = {
+	R( 0x342, 0x345 ),
+	{ 0, 0 }
+};
+
+static const unsigned short cyrillic[][2] = {
+	R( 0x483, 0x489 ),
+	{ 0, 0 }
+};
+
+static const unsigned short hebrew[][2] = {
+	R( 0x591, 0x5C4 ),
+	{ 0, 0 }
+};
+
+static const unsigned short arabic[][2] = {
+	R( 0x600, 0x603 ),
+	R( 0x610, 0x615 ),
+	R( 0x64B, 0x658 ),
+	R( 0x670, 0x670 ),
+	R( 0x6D6, 0x6ED ),
+	{ 0, 0 }
+};
+
+static const unsigned short syriac[][2] = {
+	R( 0x70F, 0x711 ),
+	R( 0x730, 0x74A ),
+	{ 0, 0 }
+};
+
+static const unsigned short thaana[][2] = {
+	R( 0x7A6, 0x7B0 ),
+	{ 0, 0 }
+};
+
+static const unsigned short devanagari[][2] = {
+	R( 0x901, 0x902 ),
+	R( 0x93C, 0x963 ),
+	{ 0, 0 }
+};
+
+static const unsigned short bengali[][2] = {
+	R( 0x981, 0x981 ),
+	R( 0x9BC, 0x9E3 ),
+	{ 0, 0 }
+};
+
+static const unsigned short gurmukhi[][2] = {
+	R( 0xA01, 0xA02 ),
+	R( 0xA3C, 0xA4D ),
+	R( 0xA70, 0xA71 ),
+	{ 0, 0 }
+};
+
+static const unsigned short gujarati[][2] = {
+	R( 0xA81, 0xA82 ),
+	R( 0xABC, 0xAE3 ),
+	{ 0, 0 }
+};
+
+static const unsigned short oriya[][2] = {
+	R( 0xB01, 0xB01 ),
+	R( 0xB3C, 0xB4D ),
+	R( 0xB56, 0xB56 ),
+	{ 0, 0 }
+};
+
+static const unsigned short tamil[][2] = {
+	R( 0xB82, 0xB82 ),
+	R( 0xBC0, 0xBCD ),
+	{ 0, 0 }
+};
+
+static const unsigned short telugu[][2] = {
+	R( 0xC3E, 0xC56 ),
+	{ 0, 0 }
+};
+
+static const unsigned short kannada[][2] = {
+	R( 0xCBC, 0xCCD ),
+	{ 0, 0 }
+};
+
+static const unsigned short malayalam[][2] = {
+	R( 0xD41, 0xD4D ),
+	{ 0, 0 }
+};
+
+static const unsigned short sinhala[][2] = {
+	R( 0xDCA, 0xDD6 ),
+	{ 0, 0 }
+};
+
+static const unsigned short thai[][2] = {
+	R( 0xE31, 0xE3A ),
+	R( 0xE47, 0xE4E ),
+	{ 0, 0 }
+};
+
+static const unsigned short lao[][2] = {
+	R( 0xEB1, 0xECD ),
+	{ 0, 0 }
+};
+
+static const unsigned short tibetan[][2] = {
+	R( 0xF18, 0xF19 ),
+	R( 0xF35, 0xF35 ),
+	R( 0xF39, 0xF39 ),
+	R( 0xF71, 0xF84 ),
+	R( 0xF90, 0xFBC ),
+	R( 0xFC6, 0xFC6 ),
+	{ 0, 0 }
+};
+
+static const unsigned short burmese[][2] = {
+	R( 0x102D, 0x1039 ),
+	R( 0x1058, 0x1059 ),
+	{ 0, 0 }
+};
+
+static const unsigned short misc_scripts[][2] = {
+	R( 0x1732, 0x1734 ), /* hanunoo */
+	R( 0x1752, 0x1753 ), /* buhid */
+	R( 0x17B4, 0x17BD ), /* khmer */
+	R( 0x17C6, 0x17D3 ),
+	R( 0x17DD, 0x17DD ),
+	R( 0x18A9, 0x18A9 ), /* mongolian */
+	R( 0x1920, 0x193B ), /* limbu (can be broken down more) */
+	{ 0, 0 }
+};
+
+#undef R
+#define R(a,b,s) { (a), (b)-(a), (s) }
+
+static const struct {
+	unsigned a, l;
+	const unsigned short (*r)[2];
+} scripts[] = {
+	R( 0x400, 0x52F, cyrillic ),
+	R( 0x590, 0x5FF, hebrew ),
+	R( 0x600, 0x6FF, arabic ),
+	R( 0x700, 0x74F, syriac ),
+	R( 0x780, 0x7B1, thaana ),
+	R( 0x900, 0x97F, devanagari ),
+	R( 0x980, 0x9FF, bengali ),
+	R( 0xA00, 0xA7F, gurmukhi ),
+	R( 0xA80, 0xAFF, gujarati ),
+	R( 0xB00, 0xB7F, oriya ),
+	R( 0xB80, 0xBFF, tamil ),
+	R( 0xC00, 0xC7F, telugu ),
+	R( 0xC80, 0xCFF, kannada ),
+	R( 0xD00, 0xD7F, malayalam ),
+	R( 0xD80, 0xDFF, sinhala ),
+	R( 0xF00, 0xFFF, tibetan ),
+	R( 0x1000, 0x108F, burmese ),
+	R( 0x1720, 0x19FF, misc_scripts ),
+	R( 0x1D2B, 0x1D2B, cyrillic ),
+	R( 0x0000, 0x10FFFF, common ),
+	{ }
+};
+
+#undef R
+
+int uu_combine_involution(unsigned b, unsigned c)
+{
+	int i;
+	unsigned code = 1;
+	const unsigned short (*r)[2];
+	for (i=0; scripts[i].l; i++)
+		if (b - scripts[i].a <= scripts[i].l)
+			for (r = scripts[i].r; r[0][0]; code += r++[0][1])
+				if (c - r[0][0] <= r[0][1])
+					return c - r[0][0] + code;
+				else if (c - code <= r[0][1])
+					return c + r[0][0] - code;
+	return 0;
+}