Add `unicode.simple_fold`; Finish `(strings|bytes).equal_fold`

1 files changed, 79 insertions, 0 deletions

diff --git a/core/unicode/fold.odin b/core/unicode/fold.odin
new file mode 100644
index 000000000..edd8a3156
--- /dev/null
+++ b/core/unicode/fold.odin
@@ -0,0 +1,79 @@
+package unicode
+
+// simple_fold iterates over the Unicode code points equivalent under the Unicode defined simple case folding.
+// simple_fold returns the smallest rune > r if one exists, or the smallest rune >= 0.
+// If no valid Unicode code point exists, r is returned.
+//
+// Example:
+// 	simple_fold('A')      == 'a'
+// 	simple_fold('a')      == 'A'
+// 	simple_fold('Z')      == 'z'
+// 	simple_fold('z')      == 'Z'
+// 	simple_fold('7')      == '7'
+// 	simple_fold('k')      == '\u212a' (Kelvin symbol, K)
+// 	simple_fold('\u212a') == 'k'
+// 	simple_fold(-3)       == -3
+@(require_results)
+simple_fold :: proc(r: rune) -> rune {
+	Fold_Pair :: struct {
+		from: u16,
+		to:   u16,
+	}
+
+	@(static, rodata)
+	ASCII_FOLD := [MAX_ASCII + 1]u16{
+		0x0000, 0x0001, 0x0002, 0x0003, 0x0004, 0x0005, 0x0006, 0x0007, 0x0008, 0x0009, 0x000a, 0x000b, 0x000c, 0x000d, 0x000e, 0x000f,
+		0x0010, 0x0011, 0x0012, 0x0013, 0x0014, 0x0015, 0x0016, 0x0017, 0x0018, 0x0019, 0x001a, 0x001b, 0x001c, 0x001d, 0x001e, 0x001f,
+		0x0020, 0x0021, 0x0022, 0x0023, 0x0024, 0x0025, 0x0026, 0x0027, 0x0028, 0x0029, 0x002a, 0x002b, 0x002c, 0x002d, 0x002e, 0x002f,
+		0x0030, 0x0031, 0x0032, 0x0033, 0x0034, 0x0035, 0x0036, 0x0037, 0x0038, 0x0039, 0x003a, 0x003b, 0x003c, 0x003d, 0x003e, 0x003f,
+		0x0040, 0x0061, 0x0062, 0x0063, 0x0064, 0x0065, 0x0066, 0x0067, 0x0068, 0x0069, 0x006a, 0x006b, 0x006c, 0x006d, 0x006e, 0x006f,
+		0x0070, 0x0071, 0x0072, 0x0073, 0x0074, 0x0075, 0x0076, 0x0077, 0x0078, 0x0079, 0x007a, 0x005b, 0x005c, 0x005d, 0x005e, 0x005f,
+		0x0060, 0x0041, 0x0042, 0x0043, 0x0044, 0x0045, 0x0046, 0x0047, 0x0048, 0x0049, 0x004a, 0x212a, 0x004c, 0x004d, 0x004e, 0x004f,
+		0x0050, 0x0051, 0x0052, 0x017f, 0x0054, 0x0055, 0x0056, 0x0057, 0x0058, 0x0059, 0x005a, 0x007b, 0x007c, 0x007d, 0x007e, 0x007f,
+	}
+
+	@(static, rodata)
+	CASE_ORBIT := [?]Fold_Pair{
+		{0x004B, 0x006B}, {0x0053, 0x0073}, {0x006B, 0x212A}, {0x0073, 0x017F}, {0x00B5, 0x039C}, {0x00C5, 0x00E5}, {0x00DF, 0x1E9E},
+		{0x00E5, 0x212B}, {0x0130, 0x0130}, {0x0131, 0x0131}, {0x017F, 0x0053}, {0x01C4, 0x01C5}, {0x01C5, 0x01C6}, {0x01C6, 0x01C4},
+		{0x01C7, 0x01C8}, {0x01C8, 0x01C9}, {0x01C9, 0x01C7}, {0x01CA, 0x01CB}, {0x01CB, 0x01CC}, {0x01CC, 0x01CA}, {0x01F1, 0x01F2},
+		{0x01F2, 0x01F3}, {0x01F3, 0x01F1}, {0x0345, 0x0399}, {0x0392, 0x03B2}, {0x0395, 0x03B5}, {0x0398, 0x03B8}, {0x0399, 0x03B9},
+		{0x039A, 0x03BA}, {0x039C, 0x03BC}, {0x03A0, 0x03C0}, {0x03A1, 0x03C1}, {0x03A3, 0x03C2}, {0x03A6, 0x03C6}, {0x03A9, 0x03C9},
+		{0x03B2, 0x03D0}, {0x03B5, 0x03F5}, {0x03B8, 0x03D1}, {0x03B9, 0x1FBE}, {0x03BA, 0x03F0}, {0x03BC, 0x00B5}, {0x03C0, 0x03D6},
+		{0x03C1, 0x03F1}, {0x03C2, 0x03C3}, {0x03C3, 0x03A3}, {0x03C6, 0x03D5}, {0x03C9, 0x2126}, {0x03D0, 0x0392}, {0x03D1, 0x03F4},
+		{0x03D5, 0x03A6}, {0x03D6, 0x03A0}, {0x03F0, 0x039A}, {0x03F1, 0x03A1}, {0x03F4, 0x0398}, {0x03F5, 0x0395}, {0x0412, 0x0432},
+		{0x0414, 0x0434}, {0x041E, 0x043E}, {0x0421, 0x0441}, {0x0422, 0x0442}, {0x042A, 0x044A}, {0x0432, 0x1C80}, {0x0434, 0x1C81},
+		{0x043E, 0x1C82}, {0x0441, 0x1C83}, {0x0442, 0x1C84}, {0x044A, 0x1C86}, {0x0462, 0x0463}, {0x0463, 0x1C87}, {0x1C80, 0x0412},
+		{0x1C81, 0x0414}, {0x1C82, 0x041E}, {0x1C83, 0x0421}, {0x1C84, 0x1C85}, {0x1C85, 0x0422}, {0x1C86, 0x042A}, {0x1C87, 0x0462},
+		{0x1C88, 0xA64A}, {0x1E60, 0x1E61}, {0x1E61, 0x1E9B}, {0x1E9B, 0x1E60}, {0x1E9E, 0x00DF}, {0x1FBE, 0x0345}, {0x2126, 0x03A9},
+		{0x212A, 0x004B}, {0x212B, 0x00C5}, {0xA64A, 0xA64B}, {0xA64B, 0x1C88},
+	}
+
+	if r < 0 || r > MAX_RUNE {
+		return r
+	}
+	if int(r) < len(ASCII_FOLD) {
+		return rune(ASCII_FOLD[r])
+	}
+
+	lo, hi := 0, len(CASE_ORBIT)
+	for lo < hi {
+		m := int(uint(lo+hi) >> 1)
+		if rune(CASE_ORBIT[m].from) < r {
+			lo = m + 1
+		} else {
+			hi = m
+		}
+	}
+
+	if lo < len(CASE_ORBIT) && rune(CASE_ORBIT[lo].from) == r {
+		return rune(CASE_ORBIT[lo].to)
+	}
+
+
+	l := to_lower(r)
+	if l != r {
+		return l
+	}
+	return to_upper(r)
+}