/* utf8.c
*
- * Copyright (c) 1998-2002, Larry Wall
+ * Copyright (C) 2000, 2001, 2002, 2003, 2004, 2005, 2006, 2007
+ * by Larry Wall and others
*
* You may distribute under the terms of either the GNU General Public
* License or the Artistic License, as specified in the README file.
#define PERL_IN_UTF8_C
#include "perl.h"
+#ifndef EBCDIC
+/* Separate prototypes needed because in ASCII systems these
+ * usually macros but they still are compiled as code, too. */
+PERL_CALLCONV UV Perl_utf8n_to_uvchr(pTHX_ const U8 *s, STRLEN curlen, STRLEN *retlen, U32 flags);
+PERL_CALLCONV U8* Perl_uvchr_to_utf8(pTHX_ U8 *d, UV uv);
+#endif
+
+static const char unees[] =
+ "Malformed UTF-8 character (unexpected end of string)";
+
/*
=head1 Unicode Support
+This file contains various utility functions for manipulating UTF8-encoded
+strings. For the uninitiated, this is a method of representing arbitrary
+Unicode characters as a variable number of bytes, in such a way that
+characters in the ASCII range are unmodified, and a zero byte never appears
+within non-zero characters.
+
=for apidoc A|U8 *|uvuni_to_utf8_flags|U8 *d|UV uv|UV flags
-Adds the UTF8 representation of the Unicode codepoint C<uv> to the end
-of the string C<d>; C<d> should be have at least C<UTF8_MAXLEN+1> free
+Adds the UTF-8 representation of the Unicode codepoint C<uv> to the end
+of the string C<d>; C<d> should be have at least C<UTF8_MAXBYTES+1> free
bytes available. The return value is the pointer to the byte after the
end of the new character. In other words,
if (ckWARN(WARN_UTF8)) {
if (UNICODE_IS_SURROGATE(uv) &&
!(flags & UNICODE_ALLOW_SURROGATE))
- Perl_warner(aTHX_ WARN_UTF8, "UTF-16 surrogate 0x%04"UVxf, uv);
+ Perl_warner(aTHX_ packWARN(WARN_UTF8), "UTF-16 surrogate 0x%04"UVxf, uv);
else if (
((uv >= 0xFDD0 && uv <= 0xFDEF &&
!(flags & UNICODE_ALLOW_FDD0))
||
- ((uv & 0xFFFF) == 0xFFFE &&
- !(flags & UNICODE_ALLOW_FFFE))
- ||
- ((uv & 0xFFFF) == 0xFFFF &&
+ ((uv & 0xFFFE) == 0xFFFE && /* Either FFFE or FFFF. */
!(flags & UNICODE_ALLOW_FFFF))) &&
/* UNICODE_ALLOW_SUPER includes
* FFFEs and FFFFs beyond 0x10FFFF. */
((uv <= PERL_UNICODE_MAX) ||
!(flags & UNICODE_ALLOW_SUPER))
)
- Perl_warner(aTHX_ WARN_UTF8,
+ Perl_warner(aTHX_ packWARN(WARN_UTF8),
"Unicode character 0x%04"UVxf" is illegal", uv);
}
if (UNI_IS_INVARIANT(uv)) {
- *d++ = UTF_TO_NATIVE(uv);
+ *d++ = (U8)UTF_TO_NATIVE(uv);
return d;
}
#if defined(EBCDIC)
STRLEN len = UNISKIP(uv);
U8 *p = d+len-1;
while (p > d) {
- *p-- = UTF_TO_NATIVE((uv & UTF_CONTINUATION_MASK) | UTF_CONTINUATION_MARK);
+ *p-- = (U8)UTF_TO_NATIVE((uv & UTF_CONTINUATION_MASK) | UTF_CONTINUATION_MARK);
uv >>= UTF_ACCUMULATION_SHIFT;
}
- *p = UTF_TO_NATIVE((uv & UTF_START_MASK(len)) | UTF_START_MARK(len));
+ *p = (U8)UTF_TO_NATIVE((uv & UTF_START_MASK(len)) | UTF_START_MARK(len));
return d+len;
}
#else /* Non loop style */
if (uv < 0x800) {
- *d++ = (( uv >> 6) | 0xc0);
- *d++ = (( uv & 0x3f) | 0x80);
+ *d++ = (U8)(( uv >> 6) | 0xc0);
+ *d++ = (U8)(( uv & 0x3f) | 0x80);
return d;
}
if (uv < 0x10000) {
- *d++ = (( uv >> 12) | 0xe0);
- *d++ = (((uv >> 6) & 0x3f) | 0x80);
- *d++ = (( uv & 0x3f) | 0x80);
+ *d++ = (U8)(( uv >> 12) | 0xe0);
+ *d++ = (U8)(((uv >> 6) & 0x3f) | 0x80);
+ *d++ = (U8)(( uv & 0x3f) | 0x80);
return d;
}
if (uv < 0x200000) {
- *d++ = (( uv >> 18) | 0xf0);
- *d++ = (((uv >> 12) & 0x3f) | 0x80);
- *d++ = (((uv >> 6) & 0x3f) | 0x80);
- *d++ = (( uv & 0x3f) | 0x80);
+ *d++ = (U8)(( uv >> 18) | 0xf0);
+ *d++ = (U8)(((uv >> 12) & 0x3f) | 0x80);
+ *d++ = (U8)(((uv >> 6) & 0x3f) | 0x80);
+ *d++ = (U8)(( uv & 0x3f) | 0x80);
return d;
}
if (uv < 0x4000000) {
- *d++ = (( uv >> 24) | 0xf8);
- *d++ = (((uv >> 18) & 0x3f) | 0x80);
- *d++ = (((uv >> 12) & 0x3f) | 0x80);
- *d++ = (((uv >> 6) & 0x3f) | 0x80);
- *d++ = (( uv & 0x3f) | 0x80);
+ *d++ = (U8)(( uv >> 24) | 0xf8);
+ *d++ = (U8)(((uv >> 18) & 0x3f) | 0x80);
+ *d++ = (U8)(((uv >> 12) & 0x3f) | 0x80);
+ *d++ = (U8)(((uv >> 6) & 0x3f) | 0x80);
+ *d++ = (U8)(( uv & 0x3f) | 0x80);
return d;
}
if (uv < 0x80000000) {
- *d++ = (( uv >> 30) | 0xfc);
- *d++ = (((uv >> 24) & 0x3f) | 0x80);
- *d++ = (((uv >> 18) & 0x3f) | 0x80);
- *d++ = (((uv >> 12) & 0x3f) | 0x80);
- *d++ = (((uv >> 6) & 0x3f) | 0x80);
- *d++ = (( uv & 0x3f) | 0x80);
+ *d++ = (U8)(( uv >> 30) | 0xfc);
+ *d++ = (U8)(((uv >> 24) & 0x3f) | 0x80);
+ *d++ = (U8)(((uv >> 18) & 0x3f) | 0x80);
+ *d++ = (U8)(((uv >> 12) & 0x3f) | 0x80);
+ *d++ = (U8)(((uv >> 6) & 0x3f) | 0x80);
+ *d++ = (U8)(( uv & 0x3f) | 0x80);
return d;
}
#ifdef HAS_QUAD
if (uv < UTF8_QUAD_MAX)
#endif
{
- *d++ = 0xfe; /* Can't match U+FEFF! */
- *d++ = (((uv >> 30) & 0x3f) | 0x80);
- *d++ = (((uv >> 24) & 0x3f) | 0x80);
- *d++ = (((uv >> 18) & 0x3f) | 0x80);
- *d++ = (((uv >> 12) & 0x3f) | 0x80);
- *d++ = (((uv >> 6) & 0x3f) | 0x80);
- *d++ = (( uv & 0x3f) | 0x80);
+ *d++ = 0xfe; /* Can't match U+FEFF! */
+ *d++ = (U8)(((uv >> 30) & 0x3f) | 0x80);
+ *d++ = (U8)(((uv >> 24) & 0x3f) | 0x80);
+ *d++ = (U8)(((uv >> 18) & 0x3f) | 0x80);
+ *d++ = (U8)(((uv >> 12) & 0x3f) | 0x80);
+ *d++ = (U8)(((uv >> 6) & 0x3f) | 0x80);
+ *d++ = (U8)(( uv & 0x3f) | 0x80);
return d;
}
#ifdef HAS_QUAD
{
- *d++ = 0xff; /* Can't match U+FFFE! */
- *d++ = 0x80; /* 6 Reserved bits */
- *d++ = (((uv >> 60) & 0x0f) | 0x80); /* 2 Reserved bits */
- *d++ = (((uv >> 54) & 0x3f) | 0x80);
- *d++ = (((uv >> 48) & 0x3f) | 0x80);
- *d++ = (((uv >> 42) & 0x3f) | 0x80);
- *d++ = (((uv >> 36) & 0x3f) | 0x80);
- *d++ = (((uv >> 30) & 0x3f) | 0x80);
- *d++ = (((uv >> 24) & 0x3f) | 0x80);
- *d++ = (((uv >> 18) & 0x3f) | 0x80);
- *d++ = (((uv >> 12) & 0x3f) | 0x80);
- *d++ = (((uv >> 6) & 0x3f) | 0x80);
- *d++ = (( uv & 0x3f) | 0x80);
+ *d++ = 0xff; /* Can't match U+FFFE! */
+ *d++ = 0x80; /* 6 Reserved bits */
+ *d++ = (U8)(((uv >> 60) & 0x0f) | 0x80); /* 2 Reserved bits */
+ *d++ = (U8)(((uv >> 54) & 0x3f) | 0x80);
+ *d++ = (U8)(((uv >> 48) & 0x3f) | 0x80);
+ *d++ = (U8)(((uv >> 42) & 0x3f) | 0x80);
+ *d++ = (U8)(((uv >> 36) & 0x3f) | 0x80);
+ *d++ = (U8)(((uv >> 30) & 0x3f) | 0x80);
+ *d++ = (U8)(((uv >> 24) & 0x3f) | 0x80);
+ *d++ = (U8)(((uv >> 18) & 0x3f) | 0x80);
+ *d++ = (U8)(((uv >> 12) & 0x3f) | 0x80);
+ *d++ = (U8)(((uv >> 6) & 0x3f) | 0x80);
+ *d++ = (U8)(( uv & 0x3f) | 0x80);
return d;
}
#endif
#endif /* Loop style */
}
-
-U8 *
-Perl_uvuni_to_utf8(pTHX_ U8 *d, UV uv)
-{
- return Perl_uvuni_to_utf8_flags(aTHX_ d, uv, 0);
-}
-
/*
-=for apidoc A|STRLEN|is_utf8_char|U8 *s
Tests if some arbitrary number of bytes begins in a valid UTF-8
-character. Note that an INVARIANT (i.e. ASCII) character is a valid UTF-8 character.
-The actual number of bytes in the UTF-8 character will be returned if
-it is valid, otherwise 0.
+character. Note that an INVARIANT (i.e. ASCII) character is a valid
+UTF-8 character. The actual number of bytes in the UTF-8 character
+will be returned if it is valid, otherwise 0.
+
+This is the "slow" version as opposed to the "fast" version which is
+the "unrolled" IS_UTF8_CHAR(). E.g. for t/uni/class.t the speed
+difference is a factor of 2 to 3. For lengths (UTF8SKIP(s)) of four
+or less you should use the IS_UTF8_CHAR(), for lengths of five or more
+you should use the _slow(). In practice this means that the _slow()
+will be used very rarely, since the maximum Unicode code point (as of
+Unicode 4.1) is U+10FFFF, which encodes in UTF-8 to four bytes. Only
+the "Perl extended UTF-8" (the infamous 'v-strings') will encode into
+five bytes or more.
-=cut
-*/
-STRLEN
-Perl_is_utf8_char(pTHX_ U8 *s)
+=cut */
+STATIC STRLEN
+S_is_utf8_char_slow(const U8 *s, const STRLEN len)
{
U8 u = *s;
- STRLEN slen, len;
+ STRLEN slen;
UV uv, ouv;
if (UTF8_IS_INVARIANT(u))
if (!UTF8_IS_START(u))
return 0;
- len = UTF8SKIP(s);
-
if (len < 2 || !UTF8_IS_CONTINUATION(s[1]))
return 0;
slen = len - 1;
s++;
+#ifdef EBCDIC
+ u = NATIVE_TO_UTF(u);
+#endif
u &= UTF_START_MASK(len);
uv = u;
ouv = uv;
s++;
}
- if (UNISKIP(uv) < len)
+ if ((STRLEN)UNISKIP(uv) < len)
return 0;
return len;
}
/*
-=for apidoc A|bool|is_utf8_string|U8 *s|STRLEN len
+=for apidoc A|STRLEN|is_utf8_char|const U8 *s
+
+Tests if some arbitrary number of bytes begins in a valid UTF-8
+character. Note that an INVARIANT (i.e. ASCII) character is a valid
+UTF-8 character. The actual number of bytes in the UTF-8 character
+will be returned if it is valid, otherwise 0.
-Returns true if first C<len> bytes of the given string form a valid UTF8
-string, false otherwise. Note that 'a valid UTF8 string' does not mean
-'a string that contains UTF8' because a valid ASCII string is a valid
-UTF8 string.
+=cut */
+STRLEN
+Perl_is_utf8_char(pTHX_ const U8 *s)
+{
+ const STRLEN len = UTF8SKIP(s);
+ PERL_UNUSED_CONTEXT;
+#ifdef IS_UTF8_CHAR
+ if (IS_UTF8_CHAR_FAST(len))
+ return IS_UTF8_CHAR(s, len) ? len : 0;
+#endif /* #ifdef IS_UTF8_CHAR */
+ return is_utf8_char_slow(s, len);
+}
+
+/*
+=for apidoc A|bool|is_utf8_string|const U8 *s|STRLEN len
+
+Returns true if first C<len> bytes of the given string form a valid
+UTF-8 string, false otherwise. Note that 'a valid UTF-8 string' does
+not mean 'a string that contains code points above 0x7F encoded in UTF-8'
+because a valid ASCII string is a valid UTF-8 string.
+
+See also is_utf8_string_loclen() and is_utf8_string_loc().
=cut
*/
bool
-Perl_is_utf8_string(pTHX_ U8 *s, STRLEN len)
+Perl_is_utf8_string(pTHX_ const U8 *s, STRLEN len)
{
- U8* x = s;
- U8* send;
- STRLEN c;
+ const U8* const send = s + (len ? len : strlen((const char *)s));
+ const U8* x = s;
- if (!len)
- len = strlen((char *)s);
- send = s + len;
+ PERL_UNUSED_CONTEXT;
while (x < send) {
- c = is_utf8_char(x);
- if (!c)
- return FALSE;
+ STRLEN c;
+ /* Inline the easy bits of is_utf8_char() here for speed... */
+ if (UTF8_IS_INVARIANT(*x))
+ c = 1;
+ else if (!UTF8_IS_START(*x))
+ goto out;
+ else {
+ /* ... and call is_utf8_char() only if really needed. */
+#ifdef IS_UTF8_CHAR
+ c = UTF8SKIP(x);
+ if (IS_UTF8_CHAR_FAST(c)) {
+ if (!IS_UTF8_CHAR(x, c))
+ c = 0;
+ }
+ else
+ c = is_utf8_char_slow(x, c);
+#else
+ c = is_utf8_char(x);
+#endif /* #ifdef IS_UTF8_CHAR */
+ if (!c)
+ goto out;
+ }
x += c;
}
+
+ out:
if (x != send)
return FALSE;
}
/*
-=for apidoc A|UV|utf8n_to_uvuni|U8 *s|STRLEN curlen|STRLEN *retlen|U32 flags
+Implemented as a macro in utf8.h
+
+=for apidoc A|bool|is_utf8_string_loc|const U8 *s|STRLEN len|const U8 **ep
+
+Like is_utf8_string() but stores the location of the failure (in the
+case of "utf8ness failure") or the location s+len (in the case of
+"utf8ness success") in the C<ep>.
+
+See also is_utf8_string_loclen() and is_utf8_string().
+
+=for apidoc A|bool|is_utf8_string_loclen|const U8 *s|STRLEN len|const U8 **ep|const STRLEN *el
+
+Like is_utf8_string() but stores the location of the failure (in the
+case of "utf8ness failure") or the location s+len (in the case of
+"utf8ness success") in the C<ep>, and the number of UTF-8
+encoded characters in the C<el>.
+
+See also is_utf8_string_loc() and is_utf8_string().
+
+=cut
+*/
+
+bool
+Perl_is_utf8_string_loclen(pTHX_ 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 c;
+ STRLEN outlen = 0;
+ PERL_UNUSED_CONTEXT;
+
+ while (x < send) {
+ /* Inline the easy bits of is_utf8_char() here for speed... */
+ if (UTF8_IS_INVARIANT(*x))
+ c = 1;
+ else if (!UTF8_IS_START(*x))
+ goto out;
+ else {
+ /* ... and call is_utf8_char() only if really needed. */
+#ifdef IS_UTF8_CHAR
+ c = UTF8SKIP(x);
+ if (IS_UTF8_CHAR_FAST(c)) {
+ if (!IS_UTF8_CHAR(x, c))
+ c = 0;
+ } else
+ c = is_utf8_char_slow(x, c);
+#else
+ c = is_utf8_char(x);
+#endif /* #ifdef IS_UTF8_CHAR */
+ if (!c)
+ goto out;
+ }
+ x += c;
+ outlen++;
+ }
+
+ out:
+ if (el)
+ *el = outlen;
+
+ if (ep)
+ *ep = x;
+ return (x == send);
+}
+
+/*
+
+=for apidoc A|UV|utf8n_to_uvuni|const U8 *s|STRLEN curlen|STRLEN *retlen|U32 flags
Bottom level UTF-8 decode routine.
Returns the unicode code point value of the first character in the string C<s>
-which is assumed to be in UTF8 encoding and no longer than C<curlen>;
+which is assumed to be in UTF-8 encoding and no longer than C<curlen>;
C<retlen> will be set to the length, in bytes, of that character.
-If C<s> does not point to a well-formed UTF8 character, the behaviour
+If C<s> does not point to a well-formed UTF-8 character, the behaviour
is dependent on the value of C<flags>: if it contains UTF8_CHECK_ONLY,
it is assumed that the caller will raise a warning, and this function
will silently just set C<retlen> to C<-1> and return zero. If the
*/
UV
-Perl_utf8n_to_uvuni(pTHX_ U8 *s, STRLEN curlen, STRLEN *retlen, U32 flags)
+Perl_utf8n_to_uvuni(pTHX_ const U8 *s, STRLEN curlen, STRLEN *retlen, U32 flags)
{
- U8 *s0 = s;
+ dVAR;
+ const U8 * const s0 = s;
UV uv = *s, ouv = 0;
STRLEN len = 1;
- bool dowarn = ckWARN_d(WARN_UTF8);
- UV startbyte = *s;
+ const bool dowarn = ckWARN_d(WARN_UTF8);
+ const UV startbyte = *s;
STRLEN expectlen = 0;
U32 warning = 0;
#define UTF8_WARN_SHORT 5
#define UTF8_WARN_OVERFLOW 6
#define UTF8_WARN_SURROGATE 7
-#define UTF8_WARN_BOM 8
-#define UTF8_WARN_LONG 9
-#define UTF8_WARN_FFFF 10
+#define UTF8_WARN_LONG 8
+#define UTF8_WARN_FFFF 9 /* Also FFFE. */
if (curlen == 0 &&
!(flags & UTF8_ALLOW_EMPTY)) {
if (!(uv > ouv)) {
/* These cannot be allowed. */
if (uv == ouv) {
- if (!(flags & UTF8_ALLOW_LONG)) {
+ if (expectlen != 13 && !(flags & UTF8_ALLOW_LONG)) {
warning = UTF8_WARN_LONG;
goto malformed;
}
!(flags & UTF8_ALLOW_SURROGATE)) {
warning = UTF8_WARN_SURROGATE;
goto malformed;
- } else if (UNICODE_IS_BYTE_ORDER_MARK(uv) &&
- !(flags & UTF8_ALLOW_BOM)) {
- warning = UTF8_WARN_BOM;
- goto malformed;
- } else if ((expectlen > UNISKIP(uv)) &&
+ } else if ((expectlen > (STRLEN)UNISKIP(uv)) &&
!(flags & UTF8_ALLOW_LONG)) {
warning = UTF8_WARN_LONG;
goto malformed;
if (flags & UTF8_CHECK_ONLY) {
if (retlen)
- *retlen = -1;
+ *retlen = ((STRLEN) -1);
return 0;
}
if (dowarn) {
- SV* sv = sv_2mortal(newSVpv("Malformed UTF-8 character ", 0));
+ SV* const sv = sv_2mortal(newSVpvs("Malformed UTF-8 character "));
switch (warning) {
case 0: /* Intentionally empty. */ break;
case UTF8_WARN_EMPTY:
- Perl_sv_catpvf(aTHX_ sv, "(empty string)");
+ sv_catpvs(sv, "(empty string)");
break;
case UTF8_WARN_CONTINUATION:
Perl_sv_catpvf(aTHX_ sv, "(unexpected continuation byte 0x%02"UVxf", with no preceding start byte)", uv);
if (s == s0)
Perl_sv_catpvf(aTHX_ sv, "(unexpected non-continuation byte 0x%02"UVxf", immediately after start byte 0x%02"UVxf")",
(UV)s[1], startbyte);
- else
+ else {
+ const int len = (int)(s-s0);
Perl_sv_catpvf(aTHX_ sv, "(unexpected non-continuation byte 0x%02"UVxf", %d byte%s after start byte 0x%02"UVxf", expected %d bytes)",
- (UV)s[1], s - s0, s - s0 > 1 ? "s" : "", startbyte, expectlen);
-
+ (UV)s[1], len, len > 1 ? "s" : "", startbyte, (int)expectlen);
+ }
+
break;
case UTF8_WARN_FE_FF:
Perl_sv_catpvf(aTHX_ sv, "(byte 0x%02"UVxf")", uv);
break;
case UTF8_WARN_SHORT:
Perl_sv_catpvf(aTHX_ sv, "(%d byte%s, need %d, after start byte 0x%02"UVxf")",
- curlen, curlen == 1 ? "" : "s", expectlen, startbyte);
+ (int)curlen, curlen == 1 ? "" : "s", (int)expectlen, startbyte);
expectlen = curlen; /* distance for caller to skip */
break;
case UTF8_WARN_OVERFLOW:
case UTF8_WARN_SURROGATE:
Perl_sv_catpvf(aTHX_ sv, "(UTF-16 surrogate 0x%04"UVxf")", uv);
break;
- case UTF8_WARN_BOM:
- Perl_sv_catpvf(aTHX_ sv, "(byte order mark 0x%04"UVxf")", uv);
- break;
case UTF8_WARN_LONG:
Perl_sv_catpvf(aTHX_ sv, "(%d byte%s, need %d, after start byte 0x%02"UVxf")",
- expectlen, expectlen == 1 ? "": "s", UNISKIP(uv), startbyte);
+ (int)expectlen, expectlen == 1 ? "": "s", UNISKIP(uv), startbyte);
break;
case UTF8_WARN_FFFF:
Perl_sv_catpvf(aTHX_ sv, "(character 0x%04"UVxf")", uv);
break;
default:
- Perl_sv_catpvf(aTHX_ sv, "(unknown reason)");
+ sv_catpvs(sv, "(unknown reason)");
break;
}
if (warning) {
- char *s = SvPVX(sv);
+ const char * const s = SvPVX_const(sv);
if (PL_op)
- Perl_warner(aTHX_ WARN_UTF8,
+ Perl_warner(aTHX_ packWARN(WARN_UTF8),
"%s in %s", s, OP_DESC(PL_op));
else
- Perl_warner(aTHX_ WARN_UTF8, "%s", s);
+ Perl_warner(aTHX_ packWARN(WARN_UTF8), "%s", s);
}
}
}
/*
-=for apidoc A|UV|utf8_to_uvchr|U8 *s|STRLEN *retlen
+=for apidoc A|UV|utf8_to_uvchr|const U8 *s|STRLEN *retlen
Returns the native character value of the first character in the string C<s>
-which is assumed to be in UTF8 encoding; C<retlen> will be set to the
+which is assumed to be in UTF-8 encoding; C<retlen> will be set to the
length, in bytes, of that character.
-If C<s> does not point to a well-formed UTF8 character, zero is
+If C<s> does not point to a well-formed UTF-8 character, zero is
returned and retlen is set, if possible, to -1.
=cut
*/
UV
-Perl_utf8_to_uvchr(pTHX_ U8 *s, STRLEN *retlen)
+Perl_utf8_to_uvchr(pTHX_ const U8 *s, STRLEN *retlen)
{
- return Perl_utf8n_to_uvchr(aTHX_ s, UTF8_MAXLEN, retlen, 0);
+ return utf8n_to_uvchr(s, UTF8_MAXBYTES, retlen,
+ ckWARN(WARN_UTF8) ? 0 : UTF8_ALLOW_ANY);
}
/*
-=for apidoc A|UV|utf8_to_uvuni|U8 *s|STRLEN *retlen
+=for apidoc A|UV|utf8_to_uvuni|const U8 *s|STRLEN *retlen
Returns the Unicode code point of the first character in the string C<s>
-which is assumed to be in UTF8 encoding; C<retlen> will be set to the
+which is assumed to be in UTF-8 encoding; C<retlen> will be set to the
length, in bytes, of that character.
This function should only be used when returned UV is considered
an index into the Unicode semantic tables (e.g. swashes).
-If C<s> does not point to a well-formed UTF8 character, zero is
+If C<s> does not point to a well-formed UTF-8 character, zero is
returned and retlen is set, if possible, to -1.
=cut
*/
UV
-Perl_utf8_to_uvuni(pTHX_ U8 *s, STRLEN *retlen)
+Perl_utf8_to_uvuni(pTHX_ const U8 *s, STRLEN *retlen)
{
/* Call the low level routine asking for checks */
- return Perl_utf8n_to_uvuni(aTHX_ s, UTF8_MAXLEN, retlen, 0);
+ return Perl_utf8n_to_uvuni(aTHX_ s, UTF8_MAXBYTES, retlen,
+ ckWARN(WARN_UTF8) ? 0 : UTF8_ALLOW_ANY);
}
/*
-=for apidoc A|STRLEN|utf8_length|U8 *s|U8 *e
+=for apidoc A|STRLEN|utf8_length|const U8 *s|const U8 *e
Return the length of the UTF-8 char encoded string C<s> in characters.
Stops at C<e> (inclusive). If C<e E<lt> s> or if the scan would end
*/
STRLEN
-Perl_utf8_length(pTHX_ U8 *s, U8 *e)
+Perl_utf8_length(pTHX_ const U8 *s, const U8 *e)
{
+ dVAR;
STRLEN len = 0;
+ U8 t = 0;
/* Note: cannot use UTF8_IS_...() too eagerly here since e.g.
* the bitops (especially ~) can create illegal UTF-8.
* In other words: in Perl UTF-8 is not just for Unicode. */
if (e < s)
- Perl_croak(aTHX_ "panic: utf8_length: unexpected end");
+ goto warn_and_return;
while (s < e) {
- U8 t = UTF8SKIP(s);
-
- if (e - s < t)
- Perl_croak(aTHX_ "panic: utf8_length: unaligned end");
+ t = UTF8SKIP(s);
+ if (e - s < t) {
+ warn_and_return:
+ if (ckWARN_d(WARN_UTF8)) {
+ if (PL_op)
+ Perl_warner(aTHX_ packWARN(WARN_UTF8),
+ "%s in %s", unees, OP_DESC(PL_op));
+ else
+ Perl_warner(aTHX_ packWARN(WARN_UTF8), unees);
+ }
+ return len;
+ }
s += t;
len++;
}
}
/*
-=for apidoc A|IV|utf8_distance|U8 *a|U8 *b
+=for apidoc A|IV|utf8_distance|const U8 *a|const U8 *b
-Returns the number of UTF8 characters between the UTF-8 pointers C<a>
+Returns the number of UTF-8 characters between the UTF-8 pointers C<a>
and C<b>.
WARNING: use only if you *know* that the pointers point inside the
*/
IV
-Perl_utf8_distance(pTHX_ U8 *a, U8 *b)
+Perl_utf8_distance(pTHX_ const U8 *a, const U8 *b)
{
- IV off = 0;
-
- /* Note: cannot use UTF8_IS_...() too eagerly here since e.g.
- * the bitops (especially ~) can create illegal UTF-8.
- * In other words: in Perl UTF-8 is not just for Unicode. */
-
- if (a < b) {
- while (a < b) {
- U8 c = UTF8SKIP(a);
-
- if (b - a < c)
- Perl_croak(aTHX_ "panic: utf8_distance: unaligned end");
- a += c;
- off--;
- }
- }
- else {
- while (b < a) {
- U8 c = UTF8SKIP(b);
-
- if (a - b < c)
- Perl_croak(aTHX_ "panic: utf8_distance: unaligned end");
- b += c;
- off++;
- }
- }
-
- return off;
+ return (a < b) ? -1 * (IV) utf8_length(a, b) : (IV) utf8_length(b, a);
}
/*
*/
U8 *
-Perl_utf8_hop(pTHX_ U8 *s, I32 off)
+Perl_utf8_hop(pTHX_ const U8 *s, I32 off)
{
+ PERL_UNUSED_CONTEXT;
/* Note: cannot use UTF8_IS_...() too eagerly here since e.g
* the bitops (especially ~) can create illegal UTF-8.
* In other words: in Perl UTF-8 is not just for Unicode. */
s--;
}
}
- return s;
+ return (U8 *)s;
}
/*
=for apidoc A|U8 *|utf8_to_bytes|U8 *s|STRLEN *len
-Converts a string C<s> of length C<len> from UTF8 into byte encoding.
+Converts a string C<s> of length C<len> from UTF-8 into byte encoding.
Unlike C<bytes_to_utf8>, this over-writes the original string, and
updates len to contain the new length.
Returns zero on failure, setting C<len> to -1.
+If you need a copy of the string, see C<bytes_from_utf8>.
+
=cut
*/
U8 *
Perl_utf8_to_bytes(pTHX_ U8 *s, STRLEN *len)
{
- U8 *send;
+ U8 * const save = s;
+ U8 * const send = s + *len;
U8 *d;
- U8 *save = s;
- /* ensure valid UTF8 and chars < 256 before updating string */
- for (send = s + *len; s < send; ) {
+ /* ensure valid UTF-8 and chars < 256 before updating string */
+ while (s < send) {
U8 c = *s++;
if (!UTF8_IS_INVARIANT(c) &&
(!UTF8_IS_DOWNGRADEABLE_START(c) || (s >= send)
|| !(c = *s++) || !UTF8_IS_CONTINUATION(c))) {
- *len = -1;
+ *len = ((STRLEN) -1);
return 0;
}
}
}
/*
-=for apidoc A|U8 *|bytes_from_utf8|U8 *s|STRLEN *len|bool *is_utf8
+=for apidoc A|U8 *|bytes_from_utf8|const U8 *s|STRLEN *len|bool *is_utf8
-Converts a string C<s> of length C<len> from UTF8 into byte encoding.
-Unlike <utf8_to_bytes> but like C<bytes_to_utf8>, returns a pointer to
+Converts a string C<s> of length C<len> from UTF-8 into byte encoding.
+Unlike C<utf8_to_bytes> but like C<bytes_to_utf8>, returns a pointer to
the newly-created string, and updates C<len> to contain the new
length. Returns the original string if no conversion occurs, C<len>
is unchanged. Do nothing if C<is_utf8> points to 0. Sets C<is_utf8> to
*/
U8 *
-Perl_bytes_from_utf8(pTHX_ U8 *s, STRLEN *len, bool *is_utf8)
+Perl_bytes_from_utf8(pTHX_ const U8 *s, STRLEN *len, bool *is_utf8)
{
U8 *d;
- U8 *start = s;
- U8 *send;
+ const U8 *start = s;
+ const U8 *send;
I32 count = 0;
+ PERL_UNUSED_CONTEXT;
if (!*is_utf8)
- return start;
+ return (U8 *)start;
- /* ensure valid UTF8 and chars < 256 before converting string */
+ /* ensure valid UTF-8 and chars < 256 before converting string */
for (send = s + *len; s < send;) {
- U8 c = *s++;
+ U8 c = *s++;
if (!UTF8_IS_INVARIANT(c)) {
if (UTF8_IS_DOWNGRADEABLE_START(c) && s < send &&
(c = *s++) && UTF8_IS_CONTINUATION(c))
count++;
else
- return start;
+ return (U8 *)start;
}
}
- *is_utf8 = 0;
+ *is_utf8 = FALSE;
- Newz(801, d, (*len) - count + 1, U8);
+ Newx(d, (*len) - count + 1, U8);
s = start; start = d;
while (s < send) {
U8 c = *s++;
}
*d = '\0';
*len = d - start;
- return start;
+ return (U8 *)start;
}
/*
-=for apidoc A|U8 *|bytes_to_utf8|U8 *s|STRLEN *len
+=for apidoc A|U8 *|bytes_to_utf8|const U8 *s|STRLEN *len
-Converts a string C<s> of length C<len> from ASCII into UTF8 encoding.
+Converts a string C<s> of length C<len> from ASCII into UTF-8 encoding.
Returns a pointer to the newly-created string, and sets C<len> to
reflect the new length.
+If you want to convert to UTF-8 from other encodings than ASCII,
+see sv_recode_to_utf8().
+
=cut
*/
U8*
-Perl_bytes_to_utf8(pTHX_ U8 *s, STRLEN *len)
+Perl_bytes_to_utf8(pTHX_ const U8 *s, STRLEN *len)
{
- U8 *send;
+ const U8 * const send = s + (*len);
U8 *d;
U8 *dst;
- send = s + (*len);
+ PERL_UNUSED_CONTEXT;
- Newz(801, d, (*len) * 2 + 1, U8);
+ Newx(d, (*len) * 2 + 1, U8);
dst = d;
while (s < send) {
- UV uv = NATIVE_TO_ASCII(*s++);
+ const UV uv = NATIVE_TO_ASCII(*s++);
if (UNI_IS_INVARIANT(uv))
- *d++ = UTF_TO_NATIVE(uv);
+ *d++ = (U8)UTF_TO_NATIVE(uv);
else {
- *d++ = UTF8_EIGHT_BIT_HI(uv);
- *d++ = UTF8_EIGHT_BIT_LO(uv);
+ *d++ = (U8)UTF8_EIGHT_BIT_HI(uv);
+ *d++ = (U8)UTF8_EIGHT_BIT_LO(uv);
}
}
*d = '\0';
U8* pend;
U8* dstart = d;
+ if (bytelen == 1 && p[0] == 0) { /* Be understanding. */
+ d[0] = 0;
+ *newlen = 1;
+ return d;
+ }
+
if (bytelen & 1)
- Perl_croak(aTHX_ "panic: utf16_to_utf8: odd bytelen");
+ Perl_croak(aTHX_ "panic: utf16_to_utf8: odd bytelen %"UVuf, (UV)bytelen);
pend = p + bytelen;
UV uv = (p[0] << 8) + p[1]; /* UTF-16BE */
p += 2;
if (uv < 0x80) {
- *d++ = uv;
+#ifdef EBCDIC
+ *d++ = UNI_TO_NATIVE(uv);
+#else
+ *d++ = (U8)uv;
+#endif
continue;
}
if (uv < 0x800) {
- *d++ = (( uv >> 6) | 0xc0);
- *d++ = (( uv & 0x3f) | 0x80);
+ *d++ = (U8)(( uv >> 6) | 0xc0);
+ *d++ = (U8)(( uv & 0x3f) | 0x80);
continue;
}
if (uv >= 0xd800 && uv < 0xdbff) { /* surrogates */
- UV low = *p++;
+ UV low = (p[0] << 8) + p[1];
+ p += 2;
if (low < 0xdc00 || low >= 0xdfff)
Perl_croak(aTHX_ "Malformed UTF-16 surrogate");
uv = ((uv - 0xd800) << 10) + (low - 0xdc00) + 0x10000;
}
if (uv < 0x10000) {
- *d++ = (( uv >> 12) | 0xe0);
- *d++ = (((uv >> 6) & 0x3f) | 0x80);
- *d++ = (( uv & 0x3f) | 0x80);
+ *d++ = (U8)(( uv >> 12) | 0xe0);
+ *d++ = (U8)(((uv >> 6) & 0x3f) | 0x80);
+ *d++ = (U8)(( uv & 0x3f) | 0x80);
continue;
}
else {
- *d++ = (( uv >> 18) | 0xf0);
- *d++ = (((uv >> 12) & 0x3f) | 0x80);
- *d++ = (((uv >> 6) & 0x3f) | 0x80);
- *d++ = (( uv & 0x3f) | 0x80);
+ *d++ = (U8)(( uv >> 18) | 0xf0);
+ *d++ = (U8)(((uv >> 12) & 0x3f) | 0x80);
+ *d++ = (U8)(((uv >> 6) & 0x3f) | 0x80);
+ *d++ = (U8)(( uv & 0x3f) | 0x80);
continue;
}
}
Perl_utf16_to_utf8_reversed(pTHX_ U8* p, U8* d, I32 bytelen, I32 *newlen)
{
U8* s = (U8*)p;
- U8* send = s + bytelen;
+ U8* const send = s + bytelen;
while (s < send) {
- U8 tmp = s[0];
+ const U8 tmp = s[0];
s[0] = s[1];
s[1] = tmp;
s += 2;
bool
Perl_is_uni_alnum(pTHX_ UV c)
{
- U8 tmpbuf[UTF8_MAXLEN+1];
+ U8 tmpbuf[UTF8_MAXBYTES+1];
uvchr_to_utf8(tmpbuf, c);
return is_utf8_alnum(tmpbuf);
}
bool
Perl_is_uni_alnumc(pTHX_ UV c)
{
- U8 tmpbuf[UTF8_MAXLEN+1];
+ U8 tmpbuf[UTF8_MAXBYTES+1];
uvchr_to_utf8(tmpbuf, c);
return is_utf8_alnumc(tmpbuf);
}
bool
Perl_is_uni_idfirst(pTHX_ UV c)
{
- U8 tmpbuf[UTF8_MAXLEN+1];
+ U8 tmpbuf[UTF8_MAXBYTES+1];
uvchr_to_utf8(tmpbuf, c);
return is_utf8_idfirst(tmpbuf);
}
bool
Perl_is_uni_alpha(pTHX_ UV c)
{
- U8 tmpbuf[UTF8_MAXLEN+1];
+ U8 tmpbuf[UTF8_MAXBYTES+1];
uvchr_to_utf8(tmpbuf, c);
return is_utf8_alpha(tmpbuf);
}
bool
Perl_is_uni_ascii(pTHX_ UV c)
{
- U8 tmpbuf[UTF8_MAXLEN+1];
+ U8 tmpbuf[UTF8_MAXBYTES+1];
uvchr_to_utf8(tmpbuf, c);
return is_utf8_ascii(tmpbuf);
}
bool
Perl_is_uni_space(pTHX_ UV c)
{
- U8 tmpbuf[UTF8_MAXLEN+1];
+ U8 tmpbuf[UTF8_MAXBYTES+1];
uvchr_to_utf8(tmpbuf, c);
return is_utf8_space(tmpbuf);
}
bool
Perl_is_uni_digit(pTHX_ UV c)
{
- U8 tmpbuf[UTF8_MAXLEN+1];
+ U8 tmpbuf[UTF8_MAXBYTES+1];
uvchr_to_utf8(tmpbuf, c);
return is_utf8_digit(tmpbuf);
}
bool
Perl_is_uni_upper(pTHX_ UV c)
{
- U8 tmpbuf[UTF8_MAXLEN+1];
+ U8 tmpbuf[UTF8_MAXBYTES+1];
uvchr_to_utf8(tmpbuf, c);
return is_utf8_upper(tmpbuf);
}
bool
Perl_is_uni_lower(pTHX_ UV c)
{
- U8 tmpbuf[UTF8_MAXLEN+1];
+ U8 tmpbuf[UTF8_MAXBYTES+1];
uvchr_to_utf8(tmpbuf, c);
return is_utf8_lower(tmpbuf);
}
bool
Perl_is_uni_cntrl(pTHX_ UV c)
{
- U8 tmpbuf[UTF8_MAXLEN+1];
+ U8 tmpbuf[UTF8_MAXBYTES+1];
uvchr_to_utf8(tmpbuf, c);
return is_utf8_cntrl(tmpbuf);
}
bool
Perl_is_uni_graph(pTHX_ UV c)
{
- U8 tmpbuf[UTF8_MAXLEN+1];
+ U8 tmpbuf[UTF8_MAXBYTES+1];
uvchr_to_utf8(tmpbuf, c);
return is_utf8_graph(tmpbuf);
}
bool
Perl_is_uni_print(pTHX_ UV c)
{
- U8 tmpbuf[UTF8_MAXLEN+1];
+ U8 tmpbuf[UTF8_MAXBYTES+1];
uvchr_to_utf8(tmpbuf, c);
return is_utf8_print(tmpbuf);
}
bool
Perl_is_uni_punct(pTHX_ UV c)
{
- U8 tmpbuf[UTF8_MAXLEN+1];
+ U8 tmpbuf[UTF8_MAXBYTES+1];
uvchr_to_utf8(tmpbuf, c);
return is_utf8_punct(tmpbuf);
}
bool
Perl_is_uni_xdigit(pTHX_ UV c)
{
- U8 tmpbuf[UTF8_MAXLEN_UCLC+1];
+ U8 tmpbuf[UTF8_MAXBYTES_CASE+1];
uvchr_to_utf8(tmpbuf, c);
return is_utf8_xdigit(tmpbuf);
}
UV
Perl_to_uni_upper(pTHX_ UV c, U8* p, STRLEN *lenp)
{
- U8 tmpbuf[UTF8_MAXLEN_UCLC+1];
- uvchr_to_utf8(tmpbuf, c);
- return to_utf8_upper(tmpbuf, p, lenp);
+ uvchr_to_utf8(p, c);
+ return to_utf8_upper(p, p, lenp);
}
UV
Perl_to_uni_title(pTHX_ UV c, U8* p, STRLEN *lenp)
{
- U8 tmpbuf[UTF8_MAXLEN_UCLC+1];
- uvchr_to_utf8(tmpbuf, c);
- return to_utf8_title(tmpbuf, p, lenp);
+ uvchr_to_utf8(p, c);
+ return to_utf8_title(p, p, lenp);
}
UV
Perl_to_uni_lower(pTHX_ UV c, U8* p, STRLEN *lenp)
{
- U8 tmpbuf[UTF8_MAXLEN_UCLC+1];
- uvchr_to_utf8(tmpbuf, c);
- return to_utf8_lower(tmpbuf, p, lenp);
+ uvchr_to_utf8(p, c);
+ return to_utf8_lower(p, p, lenp);
}
UV
Perl_to_uni_fold(pTHX_ UV c, U8* p, STRLEN *lenp)
{
- U8 tmpbuf[UTF8_MAXLEN_FOLD+1];
- uvchr_to_utf8(tmpbuf, c);
- return to_utf8_fold(tmpbuf, p, lenp);
+ uvchr_to_utf8(p, c);
+ return to_utf8_fold(p, p, lenp);
}
/* for now these all assume no locale info available for Unicode > 255 */
/* XXX returns only the first character -- do not use XXX */
/* XXX no locale support yet */
STRLEN len;
- U8 tmpbuf[UTF8_MAXLEN_UCLC+1];
+ U8 tmpbuf[UTF8_MAXBYTES_CASE+1];
return (U32)to_uni_upper(c, tmpbuf, &len);
}
/* XXX returns only the first character XXX -- do not use XXX */
/* XXX no locale support yet */
STRLEN len;
- U8 tmpbuf[UTF8_MAXLEN_UCLC+1];
+ U8 tmpbuf[UTF8_MAXBYTES_CASE+1];
return (U32)to_uni_title(c, tmpbuf, &len);
}
/* XXX returns only the first character -- do not use XXX */
/* XXX no locale support yet */
STRLEN len;
- U8 tmpbuf[UTF8_MAXLEN_UCLC+1];
+ U8 tmpbuf[UTF8_MAXBYTES_CASE+1];
return (U32)to_uni_lower(c, tmpbuf, &len);
}
-bool
-Perl_is_utf8_alnum(pTHX_ U8 *p)
+static bool
+S_is_utf8_common(pTHX_ const U8 *const p, SV **swash,
+ const char *const swashname)
{
+ dVAR;
if (!is_utf8_char(p))
return FALSE;
- if (!PL_utf8_alnum)
- /* NOTE: "IsWord", not "IsAlnum", since Alnum is a true
- * descendant of isalnum(3), in other words, it doesn't
- * contain the '_'. --jhi */
- PL_utf8_alnum = swash_init("utf8", "IsWord", &PL_sv_undef, 0, 0);
- return swash_fetch(PL_utf8_alnum, p, TRUE);
-/* return *p == '_' || is_utf8_alpha(p) || is_utf8_digit(p); */
-#ifdef SURPRISINGLY_SLOWER /* probably because alpha is usually true */
- if (!PL_utf8_alnum)
- PL_utf8_alnum = swash_init("utf8", "",
- sv_2mortal(newSVpv("+utf8::IsAlpha\n+utf8::IsDigit\n005F\n",0)), 0, 0);
- return swash_fetch(PL_utf8_alnum, p, TRUE);
-#endif
+ if (!*swash)
+ *swash = swash_init("utf8", swashname, &PL_sv_undef, 1, 0);
+ return swash_fetch(*swash, p, TRUE) != 0;
}
bool
-Perl_is_utf8_alnumc(pTHX_ U8 *p)
+Perl_is_utf8_alnum(pTHX_ const U8 *p)
{
- if (!is_utf8_char(p))
- return FALSE;
- if (!PL_utf8_alnum)
- PL_utf8_alnum = swash_init("utf8", "IsAlnumC", &PL_sv_undef, 0, 0);
- return swash_fetch(PL_utf8_alnum, p, TRUE);
-/* return is_utf8_alpha(p) || is_utf8_digit(p); */
-#ifdef SURPRISINGLY_SLOWER /* probably because alpha is usually true */
- if (!PL_utf8_alnum)
- PL_utf8_alnum = swash_init("utf8", "",
- sv_2mortal(newSVpv("+utf8::IsAlpha\n+utf8::IsDigit\n005F\n",0)), 0, 0);
- return swash_fetch(PL_utf8_alnum, p, TRUE);
-#endif
+ dVAR;
+ /* NOTE: "IsWord", not "IsAlnum", since Alnum is a true
+ * descendant of isalnum(3), in other words, it doesn't
+ * contain the '_'. --jhi */
+ return is_utf8_common(p, &PL_utf8_alnum, "IsWord");
}
bool
-Perl_is_utf8_idfirst(pTHX_ U8 *p)
+Perl_is_utf8_alnumc(pTHX_ const U8 *p)
{
- return *p == '_' || is_utf8_alpha(p);
+ dVAR;
+ return is_utf8_common(p, &PL_utf8_alnumc, "IsAlnumC");
}
bool
-Perl_is_utf8_alpha(pTHX_ U8 *p)
+Perl_is_utf8_idfirst(pTHX_ const U8 *p) /* The naming is historical. */
{
- if (!is_utf8_char(p))
- return FALSE;
- if (!PL_utf8_alpha)
- PL_utf8_alpha = swash_init("utf8", "IsAlpha", &PL_sv_undef, 0, 0);
- return swash_fetch(PL_utf8_alpha, p, TRUE);
+ dVAR;
+ if (*p == '_')
+ return TRUE;
+ /* is_utf8_idstart would be more logical. */
+ return is_utf8_common(p, &PL_utf8_idstart, "IdStart");
}
bool
-Perl_is_utf8_ascii(pTHX_ U8 *p)
+Perl_is_utf8_idcont(pTHX_ const U8 *p)
{
- if (!is_utf8_char(p))
- return FALSE;
- if (!PL_utf8_ascii)
- PL_utf8_ascii = swash_init("utf8", "IsAscii", &PL_sv_undef, 0, 0);
- return swash_fetch(PL_utf8_ascii, p, TRUE);
+ dVAR;
+ if (*p == '_')
+ return TRUE;
+ return is_utf8_common(p, &PL_utf8_idcont, "IdContinue");
}
bool
-Perl_is_utf8_space(pTHX_ U8 *p)
+Perl_is_utf8_alpha(pTHX_ const U8 *p)
{
- if (!is_utf8_char(p))
- return FALSE;
- if (!PL_utf8_space)
- PL_utf8_space = swash_init("utf8", "IsSpacePerl", &PL_sv_undef, 0, 0);
- return swash_fetch(PL_utf8_space, p, TRUE);
+ dVAR;
+ return is_utf8_common(p, &PL_utf8_alpha, "IsAlpha");
}
bool
-Perl_is_utf8_digit(pTHX_ U8 *p)
+Perl_is_utf8_ascii(pTHX_ const U8 *p)
{
- if (!is_utf8_char(p))
- return FALSE;
- if (!PL_utf8_digit)
- PL_utf8_digit = swash_init("utf8", "IsDigit", &PL_sv_undef, 0, 0);
- return swash_fetch(PL_utf8_digit, p, TRUE);
+ dVAR;
+ return is_utf8_common(p, &PL_utf8_ascii, "IsAscii");
}
bool
-Perl_is_utf8_upper(pTHX_ U8 *p)
+Perl_is_utf8_space(pTHX_ const U8 *p)
{
- if (!is_utf8_char(p))
- return FALSE;
- if (!PL_utf8_upper)
- PL_utf8_upper = swash_init("utf8", "IsUpper", &PL_sv_undef, 0, 0);
- return swash_fetch(PL_utf8_upper, p, TRUE);
+ dVAR;
+ return is_utf8_common(p, &PL_utf8_space, "IsSpacePerl");
}
bool
-Perl_is_utf8_lower(pTHX_ U8 *p)
+Perl_is_utf8_digit(pTHX_ const U8 *p)
{
- if (!is_utf8_char(p))
- return FALSE;
- if (!PL_utf8_lower)
- PL_utf8_lower = swash_init("utf8", "IsLower", &PL_sv_undef, 0, 0);
- return swash_fetch(PL_utf8_lower, p, TRUE);
+ dVAR;
+ return is_utf8_common(p, &PL_utf8_digit, "IsDigit");
}
bool
-Perl_is_utf8_cntrl(pTHX_ U8 *p)
+Perl_is_utf8_upper(pTHX_ const U8 *p)
{
- if (!is_utf8_char(p))
- return FALSE;
- if (!PL_utf8_cntrl)
- PL_utf8_cntrl = swash_init("utf8", "IsCntrl", &PL_sv_undef, 0, 0);
- return swash_fetch(PL_utf8_cntrl, p, TRUE);
+ dVAR;
+ return is_utf8_common(p, &PL_utf8_upper, "IsUppercase");
}
bool
-Perl_is_utf8_graph(pTHX_ U8 *p)
+Perl_is_utf8_lower(pTHX_ const U8 *p)
{
- if (!is_utf8_char(p))
- return FALSE;
- if (!PL_utf8_graph)
- PL_utf8_graph = swash_init("utf8", "IsGraph", &PL_sv_undef, 0, 0);
- return swash_fetch(PL_utf8_graph, p, TRUE);
+ dVAR;
+ return is_utf8_common(p, &PL_utf8_lower, "IsLowercase");
}
bool
-Perl_is_utf8_print(pTHX_ U8 *p)
+Perl_is_utf8_cntrl(pTHX_ const U8 *p)
{
- if (!is_utf8_char(p))
- return FALSE;
- if (!PL_utf8_print)
- PL_utf8_print = swash_init("utf8", "IsPrint", &PL_sv_undef, 0, 0);
- return swash_fetch(PL_utf8_print, p, TRUE);
+ dVAR;
+ return is_utf8_common(p, &PL_utf8_cntrl, "IsCntrl");
}
bool
-Perl_is_utf8_punct(pTHX_ U8 *p)
+Perl_is_utf8_graph(pTHX_ const U8 *p)
{
- if (!is_utf8_char(p))
- return FALSE;
- if (!PL_utf8_punct)
- PL_utf8_punct = swash_init("utf8", "IsPunct", &PL_sv_undef, 0, 0);
- return swash_fetch(PL_utf8_punct, p, TRUE);
+ dVAR;
+ return is_utf8_common(p, &PL_utf8_graph, "IsGraph");
}
bool
-Perl_is_utf8_xdigit(pTHX_ U8 *p)
+Perl_is_utf8_print(pTHX_ const U8 *p)
{
- if (!is_utf8_char(p))
- return FALSE;
- if (!PL_utf8_xdigit)
- PL_utf8_xdigit = swash_init("utf8", "IsXDigit", &PL_sv_undef, 0, 0);
- return swash_fetch(PL_utf8_xdigit, p, TRUE);
+ dVAR;
+ return is_utf8_common(p, &PL_utf8_print, "IsPrint");
}
bool
-Perl_is_utf8_mark(pTHX_ U8 *p)
+Perl_is_utf8_punct(pTHX_ const U8 *p)
{
- if (!is_utf8_char(p))
- return FALSE;
- if (!PL_utf8_mark)
- PL_utf8_mark = swash_init("utf8", "IsM", &PL_sv_undef, 0, 0);
- return swash_fetch(PL_utf8_mark, p, TRUE);
+ dVAR;
+ return is_utf8_common(p, &PL_utf8_punct, "IsPunct");
+}
+
+bool
+Perl_is_utf8_xdigit(pTHX_ const U8 *p)
+{
+ dVAR;
+ return is_utf8_common(p, &PL_utf8_xdigit, "Isxdigit");
+}
+
+bool
+Perl_is_utf8_mark(pTHX_ const U8 *p)
+{
+ dVAR;
+ return is_utf8_common(p, &PL_utf8_mark, "IsM");
}
/*
conversion result to. The "lenp" is a pointer to the length
of the result.
-The "swash" is a pointer to the swash to use.
+The "swashp" is a pointer to the swash to use.
-The "normal" is a string like "ToLower" which means the swash
-$utf8::ToLower, which is stored in lib/unicore/To/Lower.pl,
-and loaded by SWASHGET, using lib/utf8_heavy.pl.
+Both the special and normal mappings are stored lib/unicore/To/Foo.pl,
+and loaded by SWASHNEW, using lib/utf8_heavy.pl. The special (usually,
+but not always, a multicharacter mapping), is tried first.
-The "special" is a string like "utf8::ToSpecLower", which means
-the hash %utf8::ToSpecLower, which is stored in the same file,
-lib/unicore/To/Lower.pl, and also loaded by SWASHGET. The access
-to the hash is by Perl_to_utf8_case().
+The "special" is a string like "utf8::ToSpecLower", which means the
+hash %utf8::ToSpecLower. The access to the hash is through
+Perl_to_utf8_case().
-=cut
- */
+The "normal" is a string like "ToLower" which means the swash
+%utf8::ToLower.
+
+=cut */
UV
-Perl_to_utf8_case(pTHX_ U8 *p, U8* ustrp, STRLEN *lenp, SV **swashp, char *normal, char *special)
+Perl_to_utf8_case(pTHX_ const U8 *p, U8* ustrp, STRLEN *lenp,
+ SV **swashp, const char *normal, const char *special)
{
- UV uv0, uv1, uv2;
- U8 tmpbuf[UTF8_MAXLEN_FOLD+1];
- STRLEN len;
+ dVAR;
+ U8 tmpbuf[UTF8_MAXBYTES_CASE+1];
+ STRLEN len = 0;
- if (!*swashp)
- *swashp = swash_init("utf8", normal, &PL_sv_undef, 4, 0);
- uv0 = utf8_to_uvchr(p, 0);
+ const UV uv0 = utf8_to_uvchr(p, NULL);
/* The NATIVE_TO_UNI() and UNI_TO_NATIVE() mappings
* are necessary in EBCDIC, they are redundant no-ops
* in ASCII-ish platforms, and hopefully optimized away. */
- uv1 = NATIVE_TO_UNI(uv0);
+ const UV uv1 = NATIVE_TO_UNI(uv0);
uvuni_to_utf8(tmpbuf, uv1);
- uv2 = swash_fetch(*swashp, tmpbuf, TRUE);
- if (uv2) {
- /* It was "normal" (single character mapping). */
- UV uv3 = UNI_TO_NATIVE(uv2);
- len = uvchr_to_utf8(ustrp, uv3) - ustrp;
- if (lenp)
- *lenp = len;
+ if (!*swashp) /* load on-demand */
+ *swashp = swash_init("utf8", normal, &PL_sv_undef, 4, 0);
- return uv3;
- }
- else {
- HV *hv;
- SV *keysv;
- HE *he;
+ /* The 0xDF is the only special casing Unicode code point below 0x100. */
+ if (special && (uv1 == 0xDF || uv1 > 0xFF)) {
+ /* It might be "special" (sometimes, but not always,
+ * a multicharacter mapping) */
+ HV * const hv = get_hv(special, FALSE);
+ SV **svp;
- if ((hv = get_hv(special, FALSE)) &&
- (keysv = sv_2mortal(Perl_newSVpvf(aTHX_ "%04"UVXf, uv1))) &&
- (he = hv_fetch_ent(hv, keysv, FALSE, 0))) {
- SV *val = HeVAL(he);
- char *s = SvPV(val, len);
+ if (hv &&
+ (svp = hv_fetch(hv, (const char*)tmpbuf, UNISKIP(uv1), FALSE)) &&
+ (*svp)) {
+ const char *s;
- if (len > 1) {
- Copy(s, ustrp, len, U8);
+ s = SvPV_const(*svp, len);
+ if (len == 1)
+ len = uvuni_to_utf8(ustrp, NATIVE_TO_UNI(*(U8*)s)) - ustrp;
+ else {
#ifdef EBCDIC
- {
- /* If we have EBCDIC we need to remap the
- * characters coming in from the "special"
- * (usually, but not always multicharacter)
- * mapping, since any characters in the low 256
- * are in Unicode code points, not EBCDIC.
- * --jhi */
- U8 *d = tmpbuf;
- U8 *t, *tend;
+ /* If we have EBCDIC we need to remap the characters
+ * since any characters in the low 256 are Unicode
+ * code points, not EBCDIC. */
+ U8 *t = (U8*)s, *tend = t + len, *d;
+
+ d = tmpbuf;
+ if (SvUTF8(*svp)) {
+ STRLEN tlen = 0;
- if (SvUTF8(val)) {
- STRLEN tlen = 0;
-
- for (t = ustrp, tend = t + len;
- t < tend; t += tlen) {
- UV c = utf8_to_uvchr(t, &tlen);
-
- if (tlen > 0)
- d = uvchr_to_utf8(d, UNI_TO_NATIVE(c));
- else
- break;
+ while (t < tend) {
+ const UV c = utf8_to_uvchr(t, &tlen);
+ if (tlen > 0) {
+ d = uvchr_to_utf8(d, UNI_TO_NATIVE(c));
+ t += tlen;
}
- } else {
- for (t = ustrp, tend = t + len;
- t < tend; t++)
- d = uvchr_to_utf8(d, UNI_TO_NATIVE(*t));
+ else
+ break;
}
- len = d - tmpbuf;
- Copy(tmpbuf, ustrp, len, U8);
}
+ else {
+ while (t < tend) {
+ d = uvchr_to_utf8(d, UNI_TO_NATIVE(*t));
+ t++;
+ }
+ }
+ len = d - tmpbuf;
+ Copy(tmpbuf, ustrp, len, U8);
+#else
+ Copy(s, ustrp, len, U8);
#endif
}
- else {
- UV c = UNI_TO_NATIVE(*(U8*)s);
- U8 *d = uvchr_to_utf8(ustrp, c);
+ }
+ }
- len = d - ustrp;
- }
- if (lenp)
- *lenp = len;
+ if (!len && *swashp) {
+ const UV uv2 = swash_fetch(*swashp, tmpbuf, TRUE);
- return utf8_to_uvchr(ustrp, 0);
+ if (uv2) {
+ /* It was "normal" (a single character mapping). */
+ const UV uv3 = UNI_TO_NATIVE(uv2);
+ len = uvchr_to_utf8(ustrp, uv3) - ustrp;
}
+ }
- /* So it was not "special": just copy it. */
+ if (!len) /* Neither: just copy. */
len = uvchr_to_utf8(ustrp, uv0) - ustrp;
- if (lenp)
- *lenp = len;
- return uv0;
- }
+ if (lenp)
+ *lenp = len;
+
+ return len ? utf8_to_uvchr(ustrp, 0) : 0;
}
/*
-=for apidoc A|UV|to_utf8_upper|U8 *p|U8 *ustrp|STRLEN *lenp
+=for apidoc A|UV|to_utf8_upper|const U8 *p|U8 *ustrp|STRLEN *lenp
Convert the UTF-8 encoded character at p to its uppercase version and
store that in UTF-8 in ustrp and its length in bytes in lenp. Note
-that the ustrp needs to be at least UTF8_MAXLEN_UCLC+1 bytes since the
-uppercase version may be longer than the original character (up to two
-characters).
+that the ustrp needs to be at least UTF8_MAXBYTES_CASE+1 bytes since
+the uppercase version may be longer than the original character.
The first character of the uppercased version is returned
(but note, as explained above, that there may be more.)
=cut */
UV
-Perl_to_utf8_upper(pTHX_ U8 *p, U8* ustrp, STRLEN *lenp)
+Perl_to_utf8_upper(pTHX_ const U8 *p, U8* ustrp, STRLEN *lenp)
{
+ dVAR;
return Perl_to_utf8_case(aTHX_ p, ustrp, lenp,
&PL_utf8_toupper, "ToUpper", "utf8::ToSpecUpper");
}
/*
-=for apidoc A|UV|to_utf8_title|U8 *p|U8 *ustrp|STRLEN *lenp
+=for apidoc A|UV|to_utf8_title|const U8 *p|U8 *ustrp|STRLEN *lenp
Convert the UTF-8 encoded character at p to its titlecase version and
store that in UTF-8 in ustrp and its length in bytes in lenp. Note
-that the ustrp needs to be at least UTF8_MAXLEN_UCLC+1 bytes since the
-titlecase version may be longer than the original character (up to two
-characters).
+that the ustrp needs to be at least UTF8_MAXBYTES_CASE+1 bytes since the
+titlecase version may be longer than the original character.
The first character of the titlecased version is returned
(but note, as explained above, that there may be more.)
=cut */
UV
-Perl_to_utf8_title(pTHX_ U8 *p, U8* ustrp, STRLEN *lenp)
+Perl_to_utf8_title(pTHX_ const U8 *p, U8* ustrp, STRLEN *lenp)
{
+ dVAR;
return Perl_to_utf8_case(aTHX_ p, ustrp, lenp,
&PL_utf8_totitle, "ToTitle", "utf8::ToSpecTitle");
}
/*
-=for apidoc A|UV|to_utf8_lower|U8 *p|U8 *ustrp|STRLEN *lenp
+=for apidoc A|UV|to_utf8_lower|const U8 *p|U8 *ustrp|STRLEN *lenp
Convert the UTF-8 encoded character at p to its lowercase version and
store that in UTF-8 in ustrp and its length in bytes in lenp. Note
-that the ustrp needs to be at least UTF8_MAXLEN_UCLC+1 bytes since the
-lowercase version may be longer than the original character (up to two
-characters).
+that the ustrp needs to be at least UTF8_MAXBYTES_CASE+1 bytes since the
+lowercase version may be longer than the original character.
The first character of the lowercased version is returned
(but note, as explained above, that there may be more.)
=cut */
UV
-Perl_to_utf8_lower(pTHX_ U8 *p, U8* ustrp, STRLEN *lenp)
+Perl_to_utf8_lower(pTHX_ const U8 *p, U8* ustrp, STRLEN *lenp)
{
+ dVAR;
return Perl_to_utf8_case(aTHX_ p, ustrp, lenp,
&PL_utf8_tolower, "ToLower", "utf8::ToSpecLower");
}
/*
-=for apidoc A|UV|to_utf8_fold|U8 *p|U8 *ustrp|STRLEN *lenp
+=for apidoc A|UV|to_utf8_fold|const U8 *p|U8 *ustrp|STRLEN *lenp
Convert the UTF-8 encoded character at p to its foldcase version and
store that in UTF-8 in ustrp and its length in bytes in lenp. Note
-that the ustrp needs to be at least UTF8_MAXLEN_FOLD+1 bytes since the
+that the ustrp needs to be at least UTF8_MAXBYTES_CASE+1 bytes since the
foldcase version may be longer than the original character (up to
three characters).
=cut */
UV
-Perl_to_utf8_fold(pTHX_ U8 *p, U8* ustrp, STRLEN *lenp)
+Perl_to_utf8_fold(pTHX_ const U8 *p, U8* ustrp, STRLEN *lenp)
{
+ dVAR;
return Perl_to_utf8_case(aTHX_ p, ustrp, lenp,
&PL_utf8_tofold, "ToFold", "utf8::ToSpecFold");
}
-/* a "swash" is a swatch hash */
-
+/* Note:
+ * A "swash" is a swatch hash.
+ * A "swatch" is a bit vector generated by utf8.c:S_swash_get().
+ * C<pkg> is a pointer to a package name for SWASHNEW, should be "utf8".
+ * For other parameters, see utf8::SWASHNEW in lib/utf8_heavy.pl.
+ */
SV*
-Perl_swash_init(pTHX_ char* pkg, char* name, SV *listsv, I32 minbits, I32 none)
+Perl_swash_init(pTHX_ const char* pkg, const char* name, SV *listsv, I32 minbits, I32 none)
{
+ dVAR;
SV* retval;
- SV* tokenbufsv = sv_2mortal(NEWSV(0,0));
+ SV* const tokenbufsv = sv_newmortal();
dSP;
- HV *stash = gv_stashpvn(pkg, strlen(pkg), FALSE);
+ const size_t pkg_len = strlen(pkg);
+ const size_t name_len = strlen(name);
+ HV * const stash = gv_stashpvn(pkg, pkg_len, 0);
SV* errsv_save;
+ PUSHSTACKi(PERLSI_MAGIC);
+ ENTER;
+ SAVEI32(PL_hints);
+ PL_hints = 0;
+ save_re_context();
if (!gv_fetchmeth(stash, "SWASHNEW", 8, -1)) { /* demand load utf8 */
ENTER;
errsv_save = newSVsv(ERRSV);
- Perl_load_module(aTHX_ PERL_LOADMOD_NOIMPORT, newSVpv(pkg,0), Nullsv);
+ /* It is assumed that callers of this routine are not passing in any
+ user derived data. */
+ /* Need to do this after save_re_context() as it will set PL_tainted to
+ 1 while saving $1 etc (see the code after getrx: in Perl_magic_get).
+ Even line to create errsv_save can turn on PL_tainted. */
+ SAVEBOOL(PL_tainted);
+ PL_tainted = 0;
+ Perl_load_module(aTHX_ PERL_LOADMOD_NOIMPORT, newSVpvn(pkg,pkg_len),
+ NULL);
if (!SvTRUE(ERRSV))
sv_setsv(ERRSV, errsv_save);
SvREFCNT_dec(errsv_save);
LEAVE;
}
SPAGAIN;
- PUSHSTACKi(PERLSI_MAGIC);
PUSHMARK(SP);
EXTEND(SP,5);
- PUSHs(sv_2mortal(newSVpvn(pkg, strlen(pkg))));
- PUSHs(sv_2mortal(newSVpvn(name, strlen(name))));
+ PUSHs(sv_2mortal(newSVpvn(pkg, pkg_len)));
+ PUSHs(sv_2mortal(newSVpvn(name, name_len)));
PUSHs(listsv);
PUSHs(sv_2mortal(newSViv(minbits)));
PUSHs(sv_2mortal(newSViv(none)));
PUTBACK;
- ENTER;
- SAVEI32(PL_hints);
- PL_hints = 0;
- save_re_context();
- if (PL_curcop == &PL_compiling)
+ if (IN_PERL_COMPILETIME) {
/* XXX ought to be handled by lex_start */
+ SAVEI32(PL_in_my);
+ PL_in_my = 0;
sv_setpv(tokenbufsv, PL_tokenbuf);
+ }
errsv_save = newSVsv(ERRSV);
if (call_method("SWASHNEW", G_SCALAR))
retval = newSVsv(*PL_stack_sp--);
SvREFCNT_dec(errsv_save);
LEAVE;
POPSTACK;
- if (PL_curcop == &PL_compiling) {
+ if (IN_PERL_COMPILETIME) {
STRLEN len;
- char* pv = SvPV(tokenbufsv, len);
+ const char* const pv = SvPV_const(tokenbufsv, len);
Copy(pv, PL_tokenbuf, len+1, char);
- PL_curcop->op_private = PL_hints;
+ CopHINTS_set(PL_curcop, PL_hints);
}
- if (!SvROK(retval) || SvTYPE(SvRV(retval)) != SVt_PVHV)
+ if (!SvROK(retval) || SvTYPE(SvRV(retval)) != SVt_PVHV) {
+ if (SvPOK(retval))
+ Perl_croak(aTHX_ "Can't find Unicode property definition \"%"SVf"\"",
+ SVfARG(retval));
Perl_croak(aTHX_ "SWASHNEW didn't return an HV ref");
+ }
return retval;
}
* (see lib/unicore/SpecCase.txt) The SWASHGET in lib/utf8_heavy.pl is
* the lower-level routine, and it is similarly broken for returning
* multiple values. --jhi */
+/* Now SWASHGET is recasted into S_swash_get in this file. */
+
+/* Note:
+ * Returns the value of property/mapping C<swash> for the first character
+ * of the string C<ptr>. If C<do_utf8> is true, the string C<ptr> is
+ * assumed to be in utf8. If C<do_utf8> is false, the string C<ptr> is
+ * assumed to be in native 8-bit encoding. Caches the swatch in C<swash>.
+ */
UV
-Perl_swash_fetch(pTHX_ SV *sv, U8 *ptr, bool do_utf8)
+Perl_swash_fetch(pTHX_ SV *swash, const U8 *ptr, bool do_utf8)
{
- HV* hv = (HV*)SvRV(sv);
+ dVAR;
+ HV* const hv = (HV*)SvRV(swash);
U32 klen;
U32 off;
STRLEN slen;
STRLEN needents;
- U8 *tmps = NULL;
+ const U8 *tmps = NULL;
U32 bit;
- SV *retval;
+ SV *swatch;
U8 tmputf8[2];
- UV c = NATIVE_TO_ASCII(*ptr);
+ const UV c = NATIVE_TO_ASCII(*ptr);
if (!do_utf8 && !UNI_IS_INVARIANT(c)) {
- tmputf8[0] = UTF8_EIGHT_BIT_HI(c);
- tmputf8[1] = UTF8_EIGHT_BIT_LO(c);
- ptr = tmputf8;
+ tmputf8[0] = (U8)UTF8_EIGHT_BIT_HI(c);
+ tmputf8[1] = (U8)UTF8_EIGHT_BIT_LO(c);
+ ptr = tmputf8;
}
/* Given a UTF-X encoded char 0xAA..0xYY,0xZZ
* then the "swatch" is a vec() for al the chars which start
klen = UTF8SKIP(ptr) - 1;
off = ptr[klen];
- if (klen == 0)
- {
+ if (klen == 0) {
/* If char in invariant then swatch is for all the invariant chars
- * In both UTF-8 and UTF8-MOD that happens to be UTF_CONTINUATION_MARK
+ * In both UTF-8 and UTF-8-MOD that happens to be UTF_CONTINUATION_MARK
*/
- needents = UTF_CONTINUATION_MARK;
- off = NATIVE_TO_UTF(ptr[klen]);
- }
- else
- {
+ needents = UTF_CONTINUATION_MARK;
+ off = NATIVE_TO_UTF(ptr[klen]);
+ }
+ else {
/* If char is encoded then swatch is for the prefix */
- needents = (1 << UTF_ACCUMULATION_SHIFT);
- off = NATIVE_TO_UTF(ptr[klen]) & UTF_CONTINUATION_MASK;
- }
+ needents = (1 << UTF_ACCUMULATION_SHIFT);
+ off = NATIVE_TO_UTF(ptr[klen]) & UTF_CONTINUATION_MASK;
+ }
/*
* This single-entry cache saves about 1/3 of the utf8 overhead in test
}
else {
/* Try our second-level swatch cache, kept in a hash. */
- SV** svp = hv_fetch(hv, (char*)ptr, klen, FALSE);
+ SV** svp = hv_fetch(hv, (const char*)ptr, klen, FALSE);
- /* If not cached, generate it via utf8::SWASHGET */
- if (!svp || !SvPOK(*svp) || !(tmps = (U8*)SvPV(*svp, slen))) {
- dSP;
+ /* If not cached, generate it via swash_get */
+ if (!svp || !SvPOK(*svp)
+ || !(tmps = (const U8*)SvPV_const(*svp, slen))) {
/* We use utf8n_to_uvuni() as we want an index into
Unicode tables, not a native character number.
*/
- UV code_point = utf8n_to_uvuni(ptr, UTF8_MAXLEN, NULL, 0);
- SV *errsv_save;
- ENTER;
- SAVETMPS;
- save_re_context();
- PUSHSTACKi(PERLSI_MAGIC);
- PUSHMARK(SP);
- EXTEND(SP,3);
- PUSHs((SV*)sv);
- /* On EBCDIC & ~(0xA0-1) isn't a useful thing to do */
- PUSHs(sv_2mortal(newSViv((klen) ?
- (code_point & ~(needents - 1)) : 0)));
- PUSHs(sv_2mortal(newSViv(needents)));
- PUTBACK;
- errsv_save = newSVsv(ERRSV);
- if (call_method("SWASHGET", G_SCALAR))
- retval = newSVsv(*PL_stack_sp--);
- else
- retval = &PL_sv_undef;
- if (!SvTRUE(ERRSV))
- sv_setsv(ERRSV, errsv_save);
- SvREFCNT_dec(errsv_save);
- POPSTACK;
- FREETMPS;
- LEAVE;
- if (PL_curcop == &PL_compiling)
- PL_curcop->op_private = PL_hints;
-
- svp = hv_store(hv, (char*)ptr, klen, retval, 0);
-
- if (!svp || !(tmps = (U8*)SvPV(*svp, slen)) || (slen << 3) < needents)
- Perl_croak(aTHX_ "SWASHGET didn't return result of proper length");
+ const UV code_point = utf8n_to_uvuni(ptr, UTF8_MAXBYTES, 0,
+ ckWARN(WARN_UTF8) ?
+ 0 : UTF8_ALLOW_ANY);
+ swatch = swash_get(swash,
+ /* On EBCDIC & ~(0xA0-1) isn't a useful thing to do */
+ (klen) ? (code_point & ~(needents - 1)) : 0,
+ needents);
+
+ if (IN_PERL_COMPILETIME)
+ CopHINTS_set(PL_curcop, PL_hints);
+
+ svp = hv_store(hv, (const char *)ptr, klen, swatch, 0);
+
+ if (!svp || !(tmps = (U8*)SvPV(*svp, slen))
+ || (slen << 3) < needents)
+ Perl_croak(aTHX_ "panic: swash_fetch got improper swatch");
}
PL_last_swash_hv = hv;
PL_last_swash_klen = klen;
- PL_last_swash_tmps = tmps;
+ /* FIXME change interpvar.h? */
+ PL_last_swash_tmps = (U8 *) tmps;
PL_last_swash_slen = slen;
if (klen)
Copy(ptr, PL_last_swash_key, klen, U8);
off <<= 2;
return (tmps[off] << 24) + (tmps[off+1] << 16) + (tmps[off+2] << 8) + tmps[off + 3] ;
}
- Perl_croak(aTHX_ "panic: swash_fetch");
- return 0;
+ Perl_croak(aTHX_ "panic: swash_fetch got swatch of unexpected bit width");
+ NORETURN_FUNCTION_END;
}
+/* Note:
+ * Returns a swatch (a bit vector string) for a code point sequence
+ * that starts from the value C<start> and comprises the number C<span>.
+ * A C<swash> must be an object created by SWASHNEW (see lib/utf8_heavy.pl).
+ * Should be used via swash_fetch, which will cache the swatch in C<swash>.
+ */
+STATIC SV*
+S_swash_get(pTHX_ SV* swash, UV start, UV span)
+{
+ SV *swatch;
+ U8 *l, *lend, *x, *xend, *s;
+ STRLEN lcur, xcur, scur;
+
+ HV* const hv = (HV*)SvRV(swash);
+ SV** const listsvp = hv_fetchs(hv, "LIST", FALSE);
+ SV** const typesvp = hv_fetchs(hv, "TYPE", FALSE);
+ SV** const bitssvp = hv_fetchs(hv, "BITS", FALSE);
+ SV** const nonesvp = hv_fetchs(hv, "NONE", FALSE);
+ SV** const extssvp = hv_fetchs(hv, "EXTRAS", FALSE);
+ const U8* const typestr = (U8*)SvPV_nolen(*typesvp);
+ const int typeto = typestr[0] == 'T' && typestr[1] == 'o';
+ const STRLEN bits = SvUV(*bitssvp);
+ const STRLEN octets = bits >> 3; /* if bits == 1, then octets == 0 */
+ const UV none = SvUV(*nonesvp);
+ const UV end = start + span;
+
+ if (bits != 1 && bits != 8 && bits != 16 && bits != 32) {
+ Perl_croak(aTHX_ "panic: swash_get doesn't expect bits %"UVuf,
+ (UV)bits);
+ }
+
+ /* create and initialize $swatch */
+ swatch = newSVpvs("");
+ scur = octets ? (span * octets) : (span + 7) / 8;
+ SvGROW(swatch, scur + 1);
+ s = (U8*)SvPVX(swatch);
+ if (octets && none) {
+ const U8* const e = s + scur;
+ while (s < e) {
+ if (bits == 8)
+ *s++ = (U8)(none & 0xff);
+ else if (bits == 16) {
+ *s++ = (U8)((none >> 8) & 0xff);
+ *s++ = (U8)( none & 0xff);
+ }
+ else if (bits == 32) {
+ *s++ = (U8)((none >> 24) & 0xff);
+ *s++ = (U8)((none >> 16) & 0xff);
+ *s++ = (U8)((none >> 8) & 0xff);
+ *s++ = (U8)( none & 0xff);
+ }
+ }
+ *s = '\0';
+ }
+ else {
+ (void)memzero((U8*)s, scur + 1);
+ }
+ SvCUR_set(swatch, scur);
+ s = (U8*)SvPVX(swatch);
+
+ /* read $swash->{LIST} */
+ l = (U8*)SvPV(*listsvp, lcur);
+ lend = l + lcur;
+ while (l < lend) {
+ UV min, max, val;
+ STRLEN numlen;
+ I32 flags = PERL_SCAN_SILENT_ILLDIGIT | PERL_SCAN_DISALLOW_PREFIX;
+
+ U8* const nl = (U8*)memchr(l, '\n', lend - l);
+
+ numlen = lend - l;
+ min = grok_hex((char *)l, &numlen, &flags, NULL);
+ if (numlen)
+ l += numlen;
+ else if (nl) {
+ l = nl + 1; /* 1 is length of "\n" */
+ continue;
+ }
+ else {
+ l = lend; /* to LIST's end at which \n is not found */
+ break;
+ }
+
+ if (isBLANK(*l)) {
+ ++l;
+ flags = PERL_SCAN_SILENT_ILLDIGIT | PERL_SCAN_DISALLOW_PREFIX;
+ numlen = lend - l;
+ max = grok_hex((char *)l, &numlen, &flags, NULL);
+ if (numlen)
+ l += numlen;
+ else
+ max = min;
+
+ if (octets) {
+ if (isBLANK(*l)) {
+ ++l;
+ flags = PERL_SCAN_SILENT_ILLDIGIT |
+ PERL_SCAN_DISALLOW_PREFIX;
+ numlen = lend - l;
+ val = grok_hex((char *)l, &numlen, &flags, NULL);
+ if (numlen)
+ l += numlen;
+ else
+ val = 0;
+ }
+ else {
+ val = 0;
+ if (typeto) {
+ Perl_croak(aTHX_ "%s: illegal mapping '%s'",
+ typestr, l);
+ }
+ }
+ }
+ else
+ val = 0; /* bits == 1, then val should be ignored */
+ }
+ else {
+ max = min;
+ if (octets) {
+ val = 0;
+ if (typeto) {
+ Perl_croak(aTHX_ "%s: illegal mapping '%s'", typestr, l);
+ }
+ }
+ else
+ val = 0; /* bits == 1, then val should be ignored */
+ }
+
+ if (nl)
+ l = nl + 1;
+ else
+ l = lend;
+
+ if (max < start)
+ continue;
+
+ if (octets) {
+ UV key;
+ if (min < start) {
+ if (!none || val < none) {
+ val += start - min;
+ }
+ min = start;
+ }
+ for (key = min; key <= max; key++) {
+ STRLEN offset;
+ if (key >= end)
+ goto go_out_list;
+ /* offset must be non-negative (start <= min <= key < end) */
+ offset = octets * (key - start);
+ if (bits == 8)
+ s[offset] = (U8)(val & 0xff);
+ else if (bits == 16) {
+ s[offset ] = (U8)((val >> 8) & 0xff);
+ s[offset + 1] = (U8)( val & 0xff);
+ }
+ else if (bits == 32) {
+ s[offset ] = (U8)((val >> 24) & 0xff);
+ s[offset + 1] = (U8)((val >> 16) & 0xff);
+ s[offset + 2] = (U8)((val >> 8) & 0xff);
+ s[offset + 3] = (U8)( val & 0xff);
+ }
+
+ if (!none || val < none)
+ ++val;
+ }
+ }
+ else { /* bits == 1, then val should be ignored */
+ UV key;
+ if (min < start)
+ min = start;
+ for (key = min; key <= max; key++) {
+ const STRLEN offset = (STRLEN)(key - start);
+ if (key >= end)
+ goto go_out_list;
+ s[offset >> 3] |= 1 << (offset & 7);
+ }
+ }
+ } /* while */
+ go_out_list:
+
+ /* read $swash->{EXTRAS} */
+ x = (U8*)SvPV(*extssvp, xcur);
+ xend = x + xcur;
+ while (x < xend) {
+ STRLEN namelen;
+ U8 *namestr;
+ SV** othersvp;
+ HV* otherhv;
+ STRLEN otherbits;
+ SV **otherbitssvp, *other;
+ U8 *s, *o, *nl;
+ STRLEN slen, olen;
+
+ const U8 opc = *x++;
+ if (opc == '\n')
+ continue;
+
+ nl = (U8*)memchr(x, '\n', xend - x);
+
+ if (opc != '-' && opc != '+' && opc != '!' && opc != '&') {
+ if (nl) {
+ x = nl + 1; /* 1 is length of "\n" */
+ continue;
+ }
+ else {
+ x = xend; /* to EXTRAS' end at which \n is not found */
+ break;
+ }
+ }
+
+ namestr = x;
+ if (nl) {
+ namelen = nl - namestr;
+ x = nl + 1;
+ }
+ else {
+ namelen = xend - namestr;
+ x = xend;
+ }
+
+ othersvp = hv_fetch(hv, (char *)namestr, namelen, FALSE);
+ otherhv = (HV*)SvRV(*othersvp);
+ otherbitssvp = hv_fetchs(otherhv, "BITS", FALSE);
+ otherbits = (STRLEN)SvUV(*otherbitssvp);
+ if (bits < otherbits)
+ Perl_croak(aTHX_ "panic: swash_get found swatch size mismatch");
+
+ /* The "other" swatch must be destroyed after. */
+ other = swash_get(*othersvp, start, span);
+ o = (U8*)SvPV(other, olen);
+
+ if (!olen)
+ Perl_croak(aTHX_ "panic: swash_get got improper swatch");
+
+ s = (U8*)SvPV(swatch, slen);
+ if (bits == 1 && otherbits == 1) {
+ if (slen != olen)
+ Perl_croak(aTHX_ "panic: swash_get found swatch length mismatch");
+
+ switch (opc) {
+ case '+':
+ while (slen--)
+ *s++ |= *o++;
+ break;
+ case '!':
+ while (slen--)
+ *s++ |= ~*o++;
+ break;
+ case '-':
+ while (slen--)
+ *s++ &= ~*o++;
+ break;
+ case '&':
+ while (slen--)
+ *s++ &= *o++;
+ break;
+ default:
+ break;
+ }
+ }
+ else {
+ STRLEN otheroctets = otherbits >> 3;
+ STRLEN offset = 0;
+ U8* const send = s + slen;
+
+ while (s < send) {
+ UV otherval = 0;
+
+ if (otherbits == 1) {
+ otherval = (o[offset >> 3] >> (offset & 7)) & 1;
+ ++offset;
+ }
+ else {
+ STRLEN vlen = otheroctets;
+ otherval = *o++;
+ while (--vlen) {
+ otherval <<= 8;
+ otherval |= *o++;
+ }
+ }
+
+ if (opc == '+' && otherval)
+ NOOP; /* replace with otherval */
+ else if (opc == '!' && !otherval)
+ otherval = 1;
+ else if (opc == '-' && otherval)
+ otherval = 0;
+ else if (opc == '&' && !otherval)
+ otherval = 0;
+ else {
+ s += octets; /* no replacement */
+ continue;
+ }
+
+ if (bits == 8)
+ *s++ = (U8)( otherval & 0xff);
+ else if (bits == 16) {
+ *s++ = (U8)((otherval >> 8) & 0xff);
+ *s++ = (U8)( otherval & 0xff);
+ }
+ else if (bits == 32) {
+ *s++ = (U8)((otherval >> 24) & 0xff);
+ *s++ = (U8)((otherval >> 16) & 0xff);
+ *s++ = (U8)((otherval >> 8) & 0xff);
+ *s++ = (U8)( otherval & 0xff);
+ }
+ }
+ }
+ sv_free(other); /* through with it! */
+ } /* while */
+ return swatch;
+}
/*
=for apidoc A|U8 *|uvchr_to_utf8|U8 *d|UV uv
-Adds the UTF8 representation of the Native codepoint C<uv> to the end
-of the string C<d>; C<d> should be have at least C<UTF8_MAXLEN+1> free
+Adds the UTF-8 representation of the Native codepoint C<uv> to the end
+of the string C<d>; C<d> should be have at least C<UTF8_MAXBYTES+1> free
bytes available. The return value is the pointer to the byte after the
end of the new character. In other words,
/* On ASCII machines this is normally a macro but we want a
real function in case XS code wants it
*/
-#undef Perl_uvchr_to_utf8
U8 *
Perl_uvchr_to_utf8(pTHX_ U8 *d, UV uv)
{
}
/*
-=for apidoc A|UV|utf8n_to_uvchr|U8 *s|STRLEN curlen|STRLEN *retlen|U32 flags
+=for apidoc A|UV|utf8n_to_uvchr|U8 *s|STRLEN curlen|STRLEN *retlen|U32
+flags
-Returns the native character value of the first character in the string C<s>
-which is assumed to be in UTF8 encoding; C<retlen> will be set to the
+Returns the native character value of the first character in the string
+C<s>
+which is assumed to be in UTF-8 encoding; C<retlen> will be set to the
length, in bytes, of that character.
Allows length and flags to be passed to low level routine.
/* On ASCII machines this is normally a macro but we want
a real function in case XS code wants it
*/
-#undef Perl_utf8n_to_uvchr
UV
-Perl_utf8n_to_uvchr(pTHX_ U8 *s, STRLEN curlen, STRLEN *retlen, U32 flags)
+Perl_utf8n_to_uvchr(pTHX_ const U8 *s, STRLEN curlen, STRLEN *retlen,
+U32 flags)
{
- UV uv = Perl_utf8n_to_uvuni(aTHX_ s, curlen, retlen, flags);
+ const UV uv = Perl_utf8n_to_uvuni(aTHX_ s, curlen, retlen, flags);
return UNI_TO_NATIVE(uv);
}
=cut */
char *
-Perl_pv_uni_display(pTHX_ SV *dsv, U8 *spv, STRLEN len, STRLEN pvlim, UV flags)
+Perl_pv_uni_display(pTHX_ SV *dsv, const U8 *spv, STRLEN len, STRLEN pvlim, UV flags)
{
int truncated = 0;
- char *s, *e;
+ const char *s, *e;
sv_setpvn(dsv, "", 0);
- for (s = (char *)spv, e = s + len; s < e; s += UTF8SKIP(s)) {
+ for (s = (const char *)spv, e = s + len; s < e; s += UTF8SKIP(s)) {
UV u;
- bool ok = FALSE;
+ /* This serves double duty as a flag and a character to print after
+ a \ when flags & UNI_DISPLAY_BACKSLASH is true.
+ */
+ char ok = 0;
if (pvlim && SvCUR(dsv) >= pvlim) {
truncated++;
}
u = utf8_to_uvchr((U8*)s, 0);
if (u < 256) {
- if (!ok && (flags & UNI_DISPLAY_BACKSLASH)) {
- switch (u & 0xFF) {
+ const unsigned char c = (unsigned char)u & 0xFF;
+ if (flags & UNI_DISPLAY_BACKSLASH) {
+ switch (c) {
case '\n':
- Perl_sv_catpvf(aTHX_ dsv, "\\n"); ok = TRUE; break;
+ ok = 'n'; break;
case '\r':
- Perl_sv_catpvf(aTHX_ dsv, "\\r"); ok = TRUE; break;
+ ok = 'r'; break;
case '\t':
- Perl_sv_catpvf(aTHX_ dsv, "\\t"); ok = TRUE; break;
+ ok = 't'; break;
case '\f':
- Perl_sv_catpvf(aTHX_ dsv, "\\f"); ok = TRUE; break;
+ ok = 'f'; break;
case '\a':
- Perl_sv_catpvf(aTHX_ dsv, "\\a"); ok = TRUE; break;
+ ok = 'a'; break;
case '\\':
- Perl_sv_catpvf(aTHX_ dsv, "\\" ); ok = TRUE; break;
+ ok = '\\'; break;
default: break;
}
+ if (ok) {
+ Perl_sv_catpvf(aTHX_ dsv, "\\%c", ok);
+ }
}
/* isPRINT() is the locale-blind version. */
- if (!ok && (flags & UNI_DISPLAY_ISPRINT) && isPRINT(u & 0xFF)) {
- Perl_sv_catpvf(aTHX_ dsv, "%c", (char)(u & 0xFF));
- ok = TRUE;
+ if (!ok && (flags & UNI_DISPLAY_ISPRINT) && isPRINT(c)) {
+ Perl_sv_catpvf(aTHX_ dsv, "%c", c);
+ ok = 1;
}
}
if (!ok)
Perl_sv_catpvf(aTHX_ dsv, "\\x{%"UVxf"}", u);
}
if (truncated)
- sv_catpvn(dsv, "...", 3);
+ sv_catpvs(dsv, "...");
return SvPVX(dsv);
}
The pointer to the PV of the dsv is returned.
-=cut */
+=cut
+*/
char *
Perl_sv_uni_display(pTHX_ SV *dsv, SV *ssv, STRLEN pvlim, UV flags)
{
- return Perl_pv_uni_display(aTHX_ dsv, (U8*)SvPVX(ssv), SvCUR(ssv),
- pvlim, flags);
+ return Perl_pv_uni_display(aTHX_ dsv, (const U8*)SvPVX_const(ssv),
+ SvCUR(ssv), pvlim, flags);
}
/*
in there (they will point at the beginning of the I<next> character).
If the pointers behind pe1 or pe2 are non-NULL, they are the end
pointers beyond which scanning will not continue under any
-circustances. If the byte lengths l1 and l2 are non-zero, s1+l1 and
+circumstances. If the byte lengths l1 and l2 are non-zero, s1+l1 and
s2+l2 will be used as goal end pointers that will also stop the scan,
and which qualify towards defining a successful match: all the scans
that define an explicit length must reach their goal pointers for
I32
Perl_ibcmp_utf8(pTHX_ const char *s1, char **pe1, register UV l1, bool u1, const char *s2, char **pe2, register UV l2, bool u2)
{
- register U8 *p1 = (U8*)s1;
- register U8 *p2 = (U8*)s2;
- register U8 *e1 = 0, *f1 = 0, *q1 = 0;
- register U8 *e2 = 0, *f2 = 0, *q2 = 0;
+ dVAR;
+ register const U8 *p1 = (const U8*)s1;
+ register const U8 *p2 = (const U8*)s2;
+ register const U8 *f1 = NULL;
+ register const U8 *f2 = NULL;
+ register U8 *e1 = NULL;
+ register U8 *q1 = NULL;
+ register U8 *e2 = NULL;
+ register U8 *q2 = NULL;
STRLEN n1 = 0, n2 = 0;
- U8 foldbuf1[UTF8_MAXLEN_FOLD+1];
- U8 foldbuf2[UTF8_MAXLEN_FOLD+1];
+ U8 foldbuf1[UTF8_MAXBYTES_CASE+1];
+ U8 foldbuf2[UTF8_MAXBYTES_CASE+1];
U8 natbuf[1+1];
STRLEN foldlen1, foldlen2;
bool match;
if (pe1)
e1 = *(U8**)pe1;
- if (e1 == 0 || (l1 && l1 < e1 - (U8*)s1))
- f1 = (U8*)s1 + l1;
+ if (e1 == 0 || (l1 && l1 < (UV)(e1 - (const U8*)s1)))
+ f1 = (const U8*)s1 + l1;
if (pe2)
e2 = *(U8**)pe2;
- if (e2 == 0 || (l2 && l2 < e2 - (U8*)s2))
- f2 = (U8*)s2 + l2;
+ if (e2 == 0 || (l2 && l2 < (UV)(e2 - (const U8*)s2)))
+ f2 = (const U8*)s2 + l2;
if ((e1 == 0 && f1 == 0) || (e2 == 0 && f2 == 0) || (f1 == 0 && f2 == 0))
return 1; /* mismatch; possible infinite loop or false positive */
if (u1)
to_utf8_fold(p1, foldbuf1, &foldlen1);
else {
- natbuf[0] = *p1;
+ uvuni_to_utf8(natbuf, (UV) NATIVE_TO_UNI(((UV)*p1)));
to_utf8_fold(natbuf, foldbuf1, &foldlen1);
}
q1 = foldbuf1;
if (u2)
to_utf8_fold(p2, foldbuf2, &foldlen2);
else {
- natbuf[0] = *p2;
+ uvuni_to_utf8(natbuf, (UV) NATIVE_TO_UNI(((UV)*p2)));
to_utf8_fold(natbuf, foldbuf2, &foldlen2);
}
q2 = foldbuf2;
return match ? 0 : 1; /* 0 match, 1 mismatch */
}
+/*
+ * Local variables:
+ * c-indentation-style: bsd
+ * c-basic-offset: 4
+ * indent-tabs-mode: t
+ * End:
+ *
+ * ex: set ts=8 sts=4 sw=4 noet:
+ */