# define PERL_VARIANTS_WORD_MASK 0x80808080
# define PERL_WORD_BOUNDARY_MASK 0x3
# elif PERL_WORDSIZE == 8
-# define PERL_VARIANTS_WORD_MASK 0x8080808080808080
+# define PERL_VARIANTS_WORD_MASK UINT64_C(0x8080808080808080)
# define PERL_WORD_BOUNDARY_MASK 0x7
# else
# error Unexpected word size
#define is_utf8_string(s, len) is_utf8_string_loclen(s, len, NULL, NULL)
+#if defined(PERL_CORE) || defined (PERL_EXT)
+
+/*
+=for apidoc is_utf8_non_invariant_string
+
+Returns TRUE if L<perlapi/is_utf8_invariant_string> returns FALSE for the first
+C<len> bytes of the string C<s>, but they are, nonetheless, legal Perl-extended
+UTF-8; otherwise returns FALSE.
+
+A TRUE return means that at least one code point represented by the sequence
+either is a wide character not representable as a single byte, or the
+representation differs depending on whether the sequence is encoded in UTF-8 or
+not.
+
+See also
+C<L<perlapi/is_utf8_invariant_string>>,
+C<L<perlapi/is_utf8_string>>
+
+=cut
+
+This is commonly used to determine if a SV's UTF-8 flag should be turned on.
+It needn't be if its string is entirely UTF-8 invariant, and it shouldn't be if
+it otherwise contains invalid UTF-8.
+
+It is an internal function because khw thinks that XS code shouldn't be working
+at this low a level. A valid use case could change that.
+
+*/
+
+PERL_STATIC_INLINE bool
+S_is_utf8_non_invariant_string(const U8* const s, STRLEN len)
+{
+ const U8 * first_variant;
+
+ PERL_ARGS_ASSERT_IS_UTF8_NON_INVARIANT_STRING;
+
+ if (is_utf8_invariant_string_loc(s, len, &first_variant)) {
+ return FALSE;
+ }
+
+ return is_utf8_string(first_variant, len - (first_variant - s));
+}
+
+#endif
+
/*
=for apidoc is_strict_utf8_string
PERL_STATIC_INLINE bool
S_is_utf8_string_flags(const U8 *s, STRLEN len, const U32 flags)
{
- const U8* send;
- const U8* x = s;
+ const U8 * first_variant;
PERL_ARGS_ASSERT_IS_UTF8_STRING_FLAGS;
assert(0 == (flags & ~(UTF8_DISALLOW_ILLEGAL_INTERCHANGE
return is_c9strict_utf8_string(s, len);
}
- send = s + len;
- while (x < send) {
- STRLEN cur_len = isUTF8_CHAR_flags(x, send, flags);
- if (UNLIKELY(! cur_len)) {
- return FALSE;
+ if (! is_utf8_invariant_string_loc(s, len, &first_variant)) {
+ const U8* const send = s + len;
+ const U8* x = first_variant;
+
+ while (x < send) {
+ STRLEN cur_len = isUTF8_CHAR_flags(x, send, flags);
+ if (UNLIKELY(! cur_len)) {
+ return FALSE;
+ }
+ x += cur_len;
}
- x += cur_len;
}
return TRUE;
*/
PERL_STATIC_INLINE bool
-Perl_is_utf8_string_loclen(const U8 *s, const STRLEN len, const U8 **ep, STRLEN *el)
+Perl_is_utf8_string_loclen(const U8 *s, STRLEN len, const U8 **ep, STRLEN *el)
{
- const U8* const send = s + (len ? len : strlen((const char *)s));
- const U8* x = s;
- STRLEN outlen = 0;
+ const U8 * first_variant;
PERL_ARGS_ASSERT_IS_UTF8_STRING_LOCLEN;
- while (x < send) {
- const STRLEN cur_len = isUTF8_CHAR(x, send);
- if (UNLIKELY(! cur_len)) {
- break;
- }
- x += cur_len;
- outlen++;
+ if (len == 0) {
+ len = strlen((const char *) s);
}
- if (el)
- *el = outlen;
+ if (is_utf8_invariant_string_loc(s, len, &first_variant)) {
+ if (el)
+ *el = len;
+
+ if (ep) {
+ *ep = s + len;
+ }
- if (ep) {
- *ep = x;
+ return TRUE;
}
- return (x == send);
+ {
+ const U8* const send = s + len;
+ const U8* x = first_variant;
+ STRLEN outlen = first_variant - s;
+
+ while (x < send) {
+ const STRLEN cur_len = isUTF8_CHAR(x, send);
+ if (UNLIKELY(! cur_len)) {
+ break;
+ }
+ x += cur_len;
+ outlen++;
+ }
+
+ if (el)
+ *el = outlen;
+
+ if (ep) {
+ *ep = x;
+ }
+
+ return (x == send);
+ }
}
/*
*/
PERL_STATIC_INLINE bool
-S_is_strict_utf8_string_loclen(const U8 *s, const STRLEN len, const U8 **ep, STRLEN *el)
+S_is_strict_utf8_string_loclen(const U8 *s, STRLEN len, const U8 **ep, STRLEN *el)
{
- const U8* const send = s + (len ? len : strlen((const char *)s));
- const U8* x = s;
- STRLEN outlen = 0;
+ const U8 * first_variant;
PERL_ARGS_ASSERT_IS_STRICT_UTF8_STRING_LOCLEN;
- while (x < send) {
- const STRLEN cur_len = isSTRICT_UTF8_CHAR(x, send);
- if (UNLIKELY(! cur_len)) {
- break;
- }
- x += cur_len;
- outlen++;
+ if (len == 0) {
+ len = strlen((const char *) s);
}
- if (el)
- *el = outlen;
+ if (is_utf8_invariant_string_loc(s, len, &first_variant)) {
+ if (el)
+ *el = len;
- if (ep) {
- *ep = x;
+ if (ep) {
+ *ep = s + len;
+ }
+
+ return TRUE;
}
- return (x == send);
+ {
+ const U8* const send = s + len;
+ const U8* x = first_variant;
+ STRLEN outlen = first_variant - s;
+
+ while (x < send) {
+ const STRLEN cur_len = isSTRICT_UTF8_CHAR(x, send);
+ if (UNLIKELY(! cur_len)) {
+ break;
+ }
+ x += cur_len;
+ outlen++;
+ }
+
+ if (el)
+ *el = outlen;
+
+ if (ep) {
+ *ep = x;
+ }
+
+ return (x == send);
+ }
}
/*
*/
PERL_STATIC_INLINE bool
-S_is_c9strict_utf8_string_loclen(const U8 *s, const STRLEN len, const U8 **ep, STRLEN *el)
+S_is_c9strict_utf8_string_loclen(const U8 *s, STRLEN len, const U8 **ep, STRLEN *el)
{
- const U8* const send = s + (len ? len : strlen((const char *)s));
- const U8* x = s;
- STRLEN outlen = 0;
+ const U8 * first_variant;
PERL_ARGS_ASSERT_IS_C9STRICT_UTF8_STRING_LOCLEN;
- while (x < send) {
- const STRLEN cur_len = isC9_STRICT_UTF8_CHAR(x, send);
- if (UNLIKELY(! cur_len)) {
- break;
- }
- x += cur_len;
- outlen++;
+ if (len == 0) {
+ len = strlen((const char *) s);
}
- if (el)
- *el = outlen;
+ if (is_utf8_invariant_string_loc(s, len, &first_variant)) {
+ if (el)
+ *el = len;
+
+ if (ep) {
+ *ep = s + len;
+ }
- if (ep) {
- *ep = x;
+ return TRUE;
}
- return (x == send);
+ {
+ const U8* const send = s + len;
+ const U8* x = first_variant;
+ STRLEN outlen = first_variant - s;
+
+ while (x < send) {
+ const STRLEN cur_len = isC9_STRICT_UTF8_CHAR(x, send);
+ if (UNLIKELY(! cur_len)) {
+ break;
+ }
+ x += cur_len;
+ outlen++;
+ }
+
+ if (el)
+ *el = outlen;
+
+ if (ep) {
+ *ep = x;
+ }
+
+ return (x == send);
+ }
}
/*
PERL_STATIC_INLINE bool
S_is_utf8_string_loclen_flags(const U8 *s, STRLEN len, const U8 **ep, STRLEN *el, const U32 flags)
{
- const U8* send;
- const U8* x = s;
- STRLEN outlen = 0;
+ const U8 * first_variant;
PERL_ARGS_ASSERT_IS_UTF8_STRING_LOCLEN_FLAGS;
assert(0 == (flags & ~(UTF8_DISALLOW_ILLEGAL_INTERCHANGE
|UTF8_DISALLOW_PERL_EXTENDED)));
if (len == 0) {
- len = strlen((const char *)s);
+ len = strlen((const char *) s);
}
if (flags == 0) {
return is_c9strict_utf8_string_loclen(s, len, ep, el);
}
- send = s + len;
- while (x < send) {
- const STRLEN cur_len = isUTF8_CHAR_flags(x, send, flags);
- if (UNLIKELY(! cur_len)) {
- break;
+ if (is_utf8_invariant_string_loc(s, len, &first_variant)) {
+ if (el)
+ *el = len;
+
+ if (ep) {
+ *ep = s + len;
}
- x += cur_len;
- outlen++;
+
+ return TRUE;
}
- if (el)
- *el = outlen;
+ {
+ const U8* send = s + len;
+ const U8* x = first_variant;
+ STRLEN outlen = first_variant - s;
+
+ while (x < send) {
+ const STRLEN cur_len = isUTF8_CHAR_flags(x, send, flags);
+ if (UNLIKELY(! cur_len)) {
+ break;
+ }
+ x += cur_len;
+ outlen++;
+ }
- if (ep) {
- *ep = x;
- }
+ if (el)
+ *el = outlen;
- return (x == send);
+ if (ep) {
+ *ep = x;
+ }
+
+ return (x == send);
+ }
}
/*
PERL_STATIC_INLINE bool
S_is_utf8_fixed_width_buf_loclen_flags(const U8 * const s,
- const STRLEN len,
+ STRLEN len,
const U8 **ep,
STRLEN *el,
const U32 flags)
https://perl5.git.perl.org / perl5.git / blobdiff