{
PERL_ARGS_ASSERT_TO_UNI_LOWER;
- uvchr_to_utf8(p, c);
- return to_utf8_lower(p, p, lenp);
+ if (c > 255) {
+ uvchr_to_utf8(p, c);
+ return to_utf8_lower(p, p, lenp);
+ }
+
+ /* We have the latin1-range values compiled into the core, so just use
+ * those, converting the result to utf8 */
+ c = toLOWER_LATIN1(c);
+ if (UNI_IS_INVARIANT(c)) {
+ *p = c;
+ *lenp = 1;
+ }
+ else {
+ *p = UTF8_TWO_BYTE_HI(c);
+ *(p+1) = UTF8_TWO_BYTE_LO(c);
+ *lenp = 2;
+ }
+ return c;
}
UV
* handle locale rules */
}
else {
- if ((flags & FOLDEQ_UTF8_LOCALE)
- && (! u1 || UTF8_IS_INVARIANT(*p1) || UTF8_IS_DOWNGRADEABLE_START(*p1)))
- {
- /* There is no mixing of code points above and below 255. */
- if (u2 && (! UTF8_IS_INVARIANT(*p2)
- && ! UTF8_IS_DOWNGRADEABLE_START(*p2)))
+ if ((flags & FOLDEQ_UTF8_LOCALE)
+ && (! u1 || UTF8_IS_INVARIANT(*p1)
+ || UTF8_IS_DOWNGRADEABLE_START(*p1)))
{
- return 0;
- }
+ /* There is no mixing of code points above and below 255. */
+ if (u2 && (! UTF8_IS_INVARIANT(*p2)
+ && ! UTF8_IS_DOWNGRADEABLE_START(*p2)))
+ {
+ return 0;
+ }
- /* We handle locale rules by converting, if necessary, the code
- * point to a single byte. */
- if (! u1 || UTF8_IS_INVARIANT(*p1)) {
- *foldbuf1 = *p1;
+ /* We handle locale rules by converting, if necessary, the
+ * code point to a single byte. */
+ if (! u1 || UTF8_IS_INVARIANT(*p1)) {
+ *foldbuf1 = *p1;
+ }
+ else {
+ *foldbuf1 = TWO_BYTE_UTF8_TO_UNI(*p1, *(p1 + 1));
+ }
+ n1 = 1;
}
- else {
- *foldbuf1 = TWO_BYTE_UTF8_TO_UNI(*p1, *(p1 + 1));
+ else if (isASCII(*p1)) { /* Note, that here won't be
+ both ASCII and using locale
+ rules */
+
+ /* If trying to mix non- with ASCII, and not supposed to,
+ * fail */
+ if ((flags & FOLDEQ_UTF8_NOMIX_ASCII) && ! isASCII(*p2)) {
+ return 0;
+ }
+ n1 = 1;
+ *foldbuf1 = toLOWER(*p1); /* Folds in the ASCII range are
+ just lowercased */
}
- n1 = 1;
- }
- else if (isASCII(*p1)) { /* Note, that here won't be both ASCII
- and using locale rules */
-
- /* If trying to mix non- with ASCII, and not supposed to, fail */
- if ((flags & FOLDEQ_UTF8_NOMIX_ASCII) && ! isASCII(*p2)) {
- return 0;
+ else if (u1) {
+ to_utf8_fold(p1, foldbuf1, &n1);
}
- n1 = 1;
- *foldbuf1 = toLOWER(*p1); /* Folds in the ASCII range are
- just lowercased */
- }
- else if (u1) {
- to_utf8_fold(p1, foldbuf1, &n1);
- }
- else { /* Not utf8, convert to it first and then get fold */
- uvuni_to_utf8(natbuf, (UV) NATIVE_TO_UNI(((UV)*p1)));
- to_utf8_fold(natbuf, foldbuf1, &n1);
- }
- f1 = foldbuf1;
+ else { /* Not utf8, convert to it first and then get fold */
+ uvuni_to_utf8(natbuf, (UV) NATIVE_TO_UNI(((UV)*p1)));
+ to_utf8_fold(natbuf, foldbuf1, &n1);
+ }
+ f1 = foldbuf1;
}
}
n2 = UTF8SKIP(f2);
}
else {
- if ((flags & FOLDEQ_UTF8_LOCALE)
- && (! u2 || UTF8_IS_INVARIANT(*p2) || UTF8_IS_DOWNGRADEABLE_START(*p2)))
- {
- /* Here, the next char in s2 is < 256. We've already worked on
- * s1, and if it isn't also < 256, can't match */
- if (u1 && (! UTF8_IS_INVARIANT(*p1)
- && ! UTF8_IS_DOWNGRADEABLE_START(*p1)))
+ if ((flags & FOLDEQ_UTF8_LOCALE)
+ && (! u2 || UTF8_IS_INVARIANT(*p2) || UTF8_IS_DOWNGRADEABLE_START(*p2)))
{
- return 0;
- }
- if (! u2 || UTF8_IS_INVARIANT(*p2)) {
- *foldbuf2 = *p2;
+ /* Here, the next char in s2 is < 256. We've already
+ * worked on s1, and if it isn't also < 256, can't match */
+ if (u1 && (! UTF8_IS_INVARIANT(*p1)
+ && ! UTF8_IS_DOWNGRADEABLE_START(*p1)))
+ {
+ return 0;
+ }
+ if (! u2 || UTF8_IS_INVARIANT(*p2)) {
+ *foldbuf2 = *p2;
+ }
+ else {
+ *foldbuf2 = TWO_BYTE_UTF8_TO_UNI(*p2, *(p2 + 1));
+ }
+
+ /* Use another function to handle locale rules. We've made
+ * sure that both characters to compare are single bytes */
+ if (! foldEQ_locale((char *) f1, (char *) foldbuf2, 1)) {
+ return 0;
+ }
+ n1 = n2 = 0;
}
- else {
- *foldbuf2 = TWO_BYTE_UTF8_TO_UNI(*p2, *(p2 + 1));
+ else if (isASCII(*p2)) {
+ if (flags && ! isASCII(*p1)) {
+ return 0;
+ }
+ n2 = 1;
+ *foldbuf2 = toLOWER(*p2);
}
-
- /* Use another function to handle locale rules. We've made
- * sure that both characters to compare are single bytes */
- if (! foldEQ_locale((char *) f1, (char *) foldbuf2, 1)) {
- return 0;
+ else if (u2) {
+ to_utf8_fold(p2, foldbuf2, &n2);
}
- n1 = n2 = 0;
- }
- else if (isASCII(*p2)) {
- if (flags && ! isASCII(*p1)) {
- return 0;
+ else {
+ uvuni_to_utf8(natbuf, (UV) NATIVE_TO_UNI(((UV)*p2)));
+ to_utf8_fold(natbuf, foldbuf2, &n2);
}
- n2 = 1;
- *foldbuf2 = toLOWER(*p2);
- }
- else if (u2) {
- to_utf8_fold(p2, foldbuf2, &n2);
- }
- else {
- uvuni_to_utf8(natbuf, (UV) NATIVE_TO_UNI(((UV)*p2)));
- to_utf8_fold(natbuf, foldbuf2, &n2);
- }
- f2 = foldbuf2;
+ f2 = foldbuf2;
}
}
/* Here f1 and f2 point to the beginning of the strings to compare.
- * These strings are the folds of the input characters, stored in utf8.
- */
+ * These strings are the folds of the next character from each input
+ * string, stored in utf8. */
/* While there is more to look for in both folds, see if they
* continue to match */
https://perl5.git.perl.org / perl5.git / blobdiff