Change layout of case mapping table

Previously the case mapping table was segregated by the type of
the character (upper, lower, title) and always stored the other
two variants (key, other1, other2). Now the table is segregated
by the target type (key, other). As only very few characters have
more than one target this only slightly increases the size of the
table.

The advantage of this layout is that we only need to perform a
single table lookup in the case table. Previously, depending on
the case that was hit, either one lookup in the property table,
or two lookups in the property table and one lookup in the case
table were required.

This changes the layout from libunicode in the OpenLDAP project
-- however, the last commit there was over 10 years ago, so I
don't see value in keeping this in sync.

1 files changed, 27 insertions, 87 deletions

diff --git a/ext/mbstring/php_unicode.c b/ext/mbstring/php_unicode.c
index 2baa4690c9..25d9947713 100644
--- a/ext/mbstring/php_unicode.c
+++ b/ext/mbstring/php_unicode.c
@@ -113,7 +113,9 @@ MBSTRING_API int php_unicode_is_prop(unsigned long code, ...)
 	return result;
 }
 
-static unsigned long case_lookup(unsigned long code, long l, long r, int field)
+#define CODE_NOT_FOUND ((unsigned long) -1)
+
+static unsigned long case_lookup(unsigned long code, long l, long r)
 {
 	long m;
 	const unsigned int *tmp;
@@ -127,23 +129,20 @@ static unsigned long case_lookup(unsigned long code, long l, long r, int field)
 		 * the beginning of a case mapping triple.
 		 */
 		m = (l + r) >> 1;
-		tmp = &_uccase_map[m*3];
+		tmp = &_uccase_map[m*2];
 		if (code > *tmp)
 			l = m + 1;
 		else if (code < *tmp)
 			r = m - 1;
 		else if (code == *tmp)
-			return tmp[field];
+			return tmp[1];
 	}
 
-	return code;
+	return CODE_NOT_FOUND;
 }
 
 MBSTRING_API unsigned long php_unicode_toupper(unsigned long code, enum mbfl_no_encoding enc)
 {
-	int field;
-	long l, r;
-
 	if (code < 0x80) {
 		/* Fast path for ASCII */
 		if (code >= 0x61 && code <= 0x7A) {
@@ -153,34 +152,19 @@ MBSTRING_API unsigned long php_unicode_toupper(unsigned long code, enum mbfl_no_
 			return code - 0x20;
 		}
 		return code;
-	}
-
-	if (php_unicode_is_upper(code))
-		return code;
-
-	if (php_unicode_is_lower(code)) {
-		/*
-		 * The character is lower case.
-		 */
-		field = 1;
-		l = _uccase_len[0];
-		r = (l + _uccase_len[1]) - 1;
 	} else {
-		/*
-		 * The character is title case.
-		 */
-		field = 1;
-		l = _uccase_len[0] + _uccase_len[1];
-		r = _uccase_size - 1;
+		long l = 0;
+		long r = _uccase_len[0] - 1;
+		unsigned long new_code = case_lookup(code, l, r);
+		if (new_code != CODE_NOT_FOUND) {
+			return new_code;
+		}
+		return code;
 	}
-	return case_lookup(code, l, r, field);
 }
 
 MBSTRING_API unsigned long php_unicode_tolower(unsigned long code, enum mbfl_no_encoding enc)
 {
-	int field;
-	long l, r;
-
 	if (code < 0x80) {
 		/* Fast path for ASCII */
 		if (code >= 0x41 && code <= 0x5A) {
@@ -190,72 +174,28 @@ MBSTRING_API unsigned long php_unicode_tolower(unsigned long code, enum mbfl_no_
 			return code + 0x20;
 		}
 		return code;
-	}
-
-	if (php_unicode_is_lower(code))
-		return code;
-
-	if (php_unicode_is_upper(code)) {
-		/*
-		 * The character is upper case.
-		 */
-		field = 1;
-		l = 0;
-		r = _uccase_len[0] - 1;
 	} else {
-		/*
-		 * The character is title case.
-		 */
-		field = 2;
-		l = _uccase_len[0] + _uccase_len[1];
-		r = _uccase_size - 1;
+		long l = _uccase_len[0];
+		long r = _uccase_len[0] + _uccase_len[1] - 1;
+		unsigned long new_code = case_lookup(code, l, r);
+		if (new_code != CODE_NOT_FOUND) {
+			return new_code;
+		}
+		return code;
 	}
-	return case_lookup(code, l, r, field);
 }
 
 MBSTRING_API unsigned long php_unicode_totitle(unsigned long code, enum mbfl_no_encoding enc)
 {
-	int field;
-	long l, r;
-
-	if (php_unicode_is_title(code))
-		return code;
-
-	/*
-	 * The offset will always be the same for converting to title case.
-	 */
-	field = 2;
-
-	if (php_unicode_is_upper(code)) {
-		/*
-		 * The character is upper case.
-		 */
-		l = 0;
-		r = _uccase_len[0] - 1;
-	} else {
-		/*
-		 * The character is lower case.
-		 */
-		l = _uccase_len[0];
-		r = (l + _uccase_len[1]) - 1;
+	long l = _uccase_len[0] + _uccase_len[1];
+	long r = _uccase_size - 1;
+	unsigned long new_code = case_lookup(code, l, r);
+	if (new_code != CODE_NOT_FOUND) {
+		return new_code;
 	}
-	return case_lookup(code, l, r, field);
-
-}
-
-
-#define BE_ARY_TO_UINT32(ptr) (\
-	((unsigned char*)(ptr))[0]<<24 |\
-	((unsigned char*)(ptr))[1]<<16 |\
-	((unsigned char*)(ptr))[2]<< 8 |\
-	((unsigned char*)(ptr))[3] )
 
-#define UINT32_TO_BE_ARY(ptr,val) { \
-	unsigned int v = val; \
-	((unsigned char*)(ptr))[0] = (v>>24) & 0xff,\
-	((unsigned char*)(ptr))[1] = (v>>16) & 0xff,\
-	((unsigned char*)(ptr))[2] = (v>> 8) & 0xff,\
-	((unsigned char*)(ptr))[3] = (v    ) & 0xff;\
+	/* No dedicated title-case variant, use to-upper instead */
+	return php_unicode_toupper(code, enc);
 }
 
 struct convert_case_data {