/* Unicode support */
/*
-=for apidoc A|U8*|uvuni_to_utf8|U8 *d|UV uv
+=for apidoc A|U8 *|uvuni_to_utf8|U8 *d|UV uv
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
U8 *
Perl_uvuni_to_utf8(pTHX_ U8 *d, UV uv)
{
- if (UTF8_IS_INVARIANT(uv)) {
- *d++ = uv;
+ if (UNI_IS_INVARIANT(uv)) {
+ *d++ = UTF_TO_NATIVE(uv);
return d;
}
-#if defined(EBCDIC) || 1 /* always for testing */
+#if defined(EBCDIC)
else {
STRLEN len = UNISKIP(uv);
U8 *p = d+len-1;
#endif /* Loop style */
}
-/*
-=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
-bytes available. The return value is the pointer to the byte after the
-end of the new character. In other words,
-
- d = uvchr_to_utf8(d, uv);
-
-is the recommended wide native character-aware way of saying
-
- *(d++) = uv;
-
-=cut
-*/
-
-U8 *
-Perl_uvchr_to_utf8(pTHX_ U8 *d, UV uv)
-{
- if (uv < 0x100)
- uv = NATIVE_TO_ASCII(uv);
- return Perl_uvuni_to_utf8(aTHX_ d, uv);
-}
/*
The actual number of bytes in the UTF-8 character will be returned if
it is valid, otherwise 0.
-=cut */
+=cut
+*/
STRLEN
Perl_is_utf8_char(pTHX_ U8 *s)
{
slen = len - 1;
s++;
- /* The initial value is dubious */
+ u &= UTF_START_MASK(len);
uv = u;
ouv = uv;
while (slen--) {
if (!UTF8_IS_CONTINUATION(*s))
return 0;
uv = UTF8_ACCUMULATE(uv, *s);
- if (uv < ouv)
+ if (uv < ouv)
return 0;
ouv = uv;
s++;
Most code should use utf8_to_uvchr() rather than call this directly.
-=cut */
+=cut
+*/
UV
-Perl_utf8n_to_uvuni(pTHX_ U8* s, STRLEN curlen, STRLEN* retlen, U32 flags)
+Perl_utf8n_to_uvuni(pTHX_ U8 *s, STRLEN curlen, STRLEN *retlen, U32 flags)
{
- UV uv = *s, ouv;
+ UV uv = *s, ouv = 0;
STRLEN len = 1;
bool dowarn = ckWARN_d(WARN_UTF8);
STRLEN expectlen = 0;
if (UTF8_IS_INVARIANT(uv)) {
if (retlen)
*retlen = 1;
- return (UV) (*s);
+ return (UV) (NATIVE_TO_UTF(*s));
}
if (UTF8_IS_CONTINUATION(uv) &&
case UTF8_WARN_SHORT:
Perl_sv_catpvf(aTHX_ sv, "(%d byte%s, need %d)",
curlen, curlen == 1 ? "" : "s", expectlen);
+ expectlen = curlen; /* distance for caller to skip */
break;
case UTF8_WARN_OVERFLOW:
Perl_sv_catpvf(aTHX_ sv, "(overflow at 0x%"UVxf", byte 0x%02x)",
if (PL_op)
Perl_warner(aTHX_ WARN_UTF8,
- "%s in %s", s, PL_op_desc[PL_op->op_type]);
+ "%s in %s", s, OP_DESC(PL_op));
else
Perl_warner(aTHX_ WARN_UTF8, "%s", s);
}
}
/*
-=for apidoc A|U8* s|utf8n_to_uvchr|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
-length, in bytes, of that character.
-
-Allows length and flags to be passed to low level routine.
-
-=cut
-*/
-
-UV
-Perl_utf8n_to_uvchr(pTHX_ U8* s, STRLEN curlen, STRLEN* retlen, U32 flags)
-{
- UV uv = Perl_utf8n_to_uvuni(aTHX_ s, curlen, retlen, flags);
- if (uv < 0x100)
- return (UV) ASCII_TO_NATIVE(uv);
- return uv;
-}
-
-/*
-=for apidoc A|U8* s|utf8_to_uvchr|STRLEN *retlen
+=for apidoc A|UV|utf8_to_uvchr|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
*/
UV
-Perl_utf8_to_uvchr(pTHX_ U8* s, STRLEN* retlen)
+Perl_utf8_to_uvchr(pTHX_ U8 *s, STRLEN *retlen)
{
return Perl_utf8n_to_uvchr(aTHX_ s, UTF8_MAXLEN, retlen, 0);
}
/*
-=for apidoc A|U8* s|utf8_to_uvuni|STRLEN *retlen
+=for apidoc A|UV|utf8_to_uvuni|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
*/
UV
-Perl_utf8_to_uvuni(pTHX_ U8* s, STRLEN* retlen)
+Perl_utf8_to_uvuni(pTHX_ U8 *s, STRLEN *retlen)
{
/* Call the low level routine asking for checks */
return Perl_utf8n_to_uvuni(aTHX_ s, UTF8_MAXLEN, retlen, 0);
}
/*
-=for apidoc A|STRLEN|utf8_length|U8* s|U8 *e
+=for apidoc A|STRLEN|utf8_length|U8 *s|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_ U8 *s, U8 *e)
{
STRLEN len = 0;
WARNING: use only if you *know* that the pointers point inside the
same UTF-8 buffer.
-=cut */
+=cut
+*/
IV
Perl_utf8_distance(pTHX_ U8 *a, U8 *b)
}
/*
-=for apidoc A|U8*|utf8_hop|U8 *s|I32 off
+=for apidoc A|U8 *|utf8_hop|U8 *s|I32 off
Return the UTF-8 pointer C<s> displaced by C<off> characters, either
forward or backward.
the UTF-8 data pointed to by C<s> *and* that on entry C<s> is aligned
on the first byte of character or just after the last byte of a character.
-=cut */
+=cut
+*/
U8 *
Perl_utf8_hop(pTHX_ U8 *s, I32 off)
*/
U8 *
-Perl_utf8_to_bytes(pTHX_ U8* s, STRLEN *len)
+Perl_utf8_to_bytes(pTHX_ U8 *s, STRLEN *len)
{
U8 *send;
U8 *d;
is unchanged. Do nothing if C<is_utf8> points to 0. Sets C<is_utf8> to
0 if C<s> is converted or contains all 7bit characters.
-=cut */
+=cut
+*/
U8 *
-Perl_bytes_from_utf8(pTHX_ U8* s, STRLEN *len, bool *is_utf8)
+Perl_bytes_from_utf8(pTHX_ U8 *s, STRLEN *len, bool *is_utf8)
{
U8 *d;
U8 *start = s;
*is_utf8 = 0;
-#ifndef EBCDIC
- /* Can use as-is if no high chars */
- if (!count)
- return start;
-#endif
-
Newz(801, d, (*len) - count + 1, U8);
s = start; start = d;
while (s < send) {
U8 c = *s++;
- if (!UTF8_IS_INVARIANT(c))
- c = UTF8_ACCUMULATE(c, *s++);
- *d++ = ASCII_TO_NATIVE(c);
+ if (!UTF8_IS_INVARIANT(c)) {
+ /* Then it is two-byte encoded */
+ c = UTF8_ACCUMULATE(NATIVE_TO_UTF(c), *s++);
+ c = ASCII_TO_NATIVE(c);
+ }
+ *d++ = c;
}
*d = '\0';
*len = d - start;
*/
U8*
-Perl_bytes_to_utf8(pTHX_ U8* s, STRLEN *len)
+Perl_bytes_to_utf8(pTHX_ U8 *s, STRLEN *len)
{
U8 *send;
U8 *d;
while (s < send) {
UV uv = NATIVE_TO_ASCII(*s++);
- if (UTF8_IS_INVARIANT(uv))
- *d++ = uv;
+ if (UNI_IS_INVARIANT(uv))
+ *d++ = UTF_TO_NATIVE(uv);
else {
*d++ = UTF8_EIGHT_BIT_HI(uv);
*d++ = UTF8_EIGHT_BIT_LO(uv);
/* for now these are all defined (inefficiently) in terms of the utf8 versions */
bool
-Perl_is_uni_alnum(pTHX_ U32 c)
+Perl_is_uni_alnum(pTHX_ UV c)
{
U8 tmpbuf[UTF8_MAXLEN+1];
- uvuni_to_utf8(tmpbuf, (UV)c);
+ uvchr_to_utf8(tmpbuf, (UV)c);
return is_utf8_alnum(tmpbuf);
}
bool
-Perl_is_uni_alnumc(pTHX_ U32 c)
+Perl_is_uni_alnumc(pTHX_ UV c)
{
U8 tmpbuf[UTF8_MAXLEN+1];
- uvuni_to_utf8(tmpbuf, (UV)c);
+ uvchr_to_utf8(tmpbuf, (UV)c);
return is_utf8_alnumc(tmpbuf);
}
bool
-Perl_is_uni_idfirst(pTHX_ U32 c)
+Perl_is_uni_idfirst(pTHX_ UV c)
{
U8 tmpbuf[UTF8_MAXLEN+1];
- uvuni_to_utf8(tmpbuf, (UV)c);
+ uvchr_to_utf8(tmpbuf, (UV)c);
return is_utf8_idfirst(tmpbuf);
}
bool
-Perl_is_uni_alpha(pTHX_ U32 c)
+Perl_is_uni_alpha(pTHX_ UV c)
{
U8 tmpbuf[UTF8_MAXLEN+1];
- uvuni_to_utf8(tmpbuf, (UV)c);
+ uvchr_to_utf8(tmpbuf, (UV)c);
return is_utf8_alpha(tmpbuf);
}
bool
-Perl_is_uni_ascii(pTHX_ U32 c)
+Perl_is_uni_ascii(pTHX_ UV c)
{
U8 tmpbuf[UTF8_MAXLEN+1];
- uvuni_to_utf8(tmpbuf, (UV)c);
+ uvchr_to_utf8(tmpbuf, (UV)c);
return is_utf8_ascii(tmpbuf);
}
bool
-Perl_is_uni_space(pTHX_ U32 c)
+Perl_is_uni_space(pTHX_ UV c)
{
U8 tmpbuf[UTF8_MAXLEN+1];
- uvuni_to_utf8(tmpbuf, (UV)c);
+ uvchr_to_utf8(tmpbuf, (UV)c);
return is_utf8_space(tmpbuf);
}
bool
-Perl_is_uni_digit(pTHX_ U32 c)
+Perl_is_uni_digit(pTHX_ UV c)
{
U8 tmpbuf[UTF8_MAXLEN+1];
- uvuni_to_utf8(tmpbuf, (UV)c);
+ uvchr_to_utf8(tmpbuf, (UV)c);
return is_utf8_digit(tmpbuf);
}
bool
-Perl_is_uni_upper(pTHX_ U32 c)
+Perl_is_uni_upper(pTHX_ UV c)
{
U8 tmpbuf[UTF8_MAXLEN+1];
- uvuni_to_utf8(tmpbuf, (UV)c);
+ uvchr_to_utf8(tmpbuf, (UV)c);
return is_utf8_upper(tmpbuf);
}
bool
-Perl_is_uni_lower(pTHX_ U32 c)
+Perl_is_uni_lower(pTHX_ UV c)
{
U8 tmpbuf[UTF8_MAXLEN+1];
- uvuni_to_utf8(tmpbuf, (UV)c);
+ uvchr_to_utf8(tmpbuf, (UV)c);
return is_utf8_lower(tmpbuf);
}
bool
-Perl_is_uni_cntrl(pTHX_ U32 c)
+Perl_is_uni_cntrl(pTHX_ UV c)
{
U8 tmpbuf[UTF8_MAXLEN+1];
- uvuni_to_utf8(tmpbuf, (UV)c);
+ uvchr_to_utf8(tmpbuf, (UV)c);
return is_utf8_cntrl(tmpbuf);
}
bool
-Perl_is_uni_graph(pTHX_ U32 c)
+Perl_is_uni_graph(pTHX_ UV c)
{
U8 tmpbuf[UTF8_MAXLEN+1];
- uvuni_to_utf8(tmpbuf, (UV)c);
+ uvchr_to_utf8(tmpbuf, (UV)c);
return is_utf8_graph(tmpbuf);
}
bool
-Perl_is_uni_print(pTHX_ U32 c)
+Perl_is_uni_print(pTHX_ UV c)
{
U8 tmpbuf[UTF8_MAXLEN+1];
- uvuni_to_utf8(tmpbuf, (UV)c);
+ uvchr_to_utf8(tmpbuf, (UV)c);
return is_utf8_print(tmpbuf);
}
bool
-Perl_is_uni_punct(pTHX_ U32 c)
+Perl_is_uni_punct(pTHX_ UV c)
{
U8 tmpbuf[UTF8_MAXLEN+1];
- uvuni_to_utf8(tmpbuf, (UV)c);
+ uvchr_to_utf8(tmpbuf, (UV)c);
return is_utf8_punct(tmpbuf);
}
bool
-Perl_is_uni_xdigit(pTHX_ U32 c)
+Perl_is_uni_xdigit(pTHX_ UV c)
{
- U8 tmpbuf[UTF8_MAXLEN+1];
- uvuni_to_utf8(tmpbuf, (UV)c);
+ U8 tmpbuf[UTF8_MAXLEN_UCLC+1];
+ uvchr_to_utf8(tmpbuf, (UV)c);
return is_utf8_xdigit(tmpbuf);
}
-U32
-Perl_to_uni_upper(pTHX_ U32 c)
+UV
+Perl_to_uni_upper(pTHX_ UV c, U8* p, STRLEN *lenp)
{
- U8 tmpbuf[UTF8_MAXLEN+1];
- uvuni_to_utf8(tmpbuf, (UV)c);
- return to_utf8_upper(tmpbuf);
+ U8 tmpbuf[UTF8_MAXLEN_UCLC+1];
+ uvchr_to_utf8(tmpbuf, (UV)c);
+ return to_utf8_upper(tmpbuf, p, lenp);
}
-U32
-Perl_to_uni_title(pTHX_ U32 c)
+UV
+Perl_to_uni_title(pTHX_ UV c, U8* p, STRLEN *lenp)
{
- U8 tmpbuf[UTF8_MAXLEN+1];
- uvuni_to_utf8(tmpbuf, (UV)c);
- return to_utf8_title(tmpbuf);
+ U8 tmpbuf[UTF8_MAXLEN_UCLC+1];
+ uvchr_to_utf8(tmpbuf, (UV)c);
+ return to_utf8_title(tmpbuf, p, lenp);
}
-U32
-Perl_to_uni_lower(pTHX_ U32 c)
+UV
+Perl_to_uni_lower(pTHX_ UV c, U8* p, STRLEN *lenp)
{
- U8 tmpbuf[UTF8_MAXLEN+1];
- uvuni_to_utf8(tmpbuf, (UV)c);
- return to_utf8_lower(tmpbuf);
+ U8 tmpbuf[UTF8_MAXLEN_UCLC+1];
+ uvchr_to_utf8(tmpbuf, (UV)c);
+ return to_utf8_lower(tmpbuf, p, lenp);
+}
+
+UV
+Perl_to_uni_fold(pTHX_ UV c, U8* p, STRLEN *lenp)
+{
+ U8 tmpbuf[UTF8_MAXLEN_FOLD+1];
+ uvchr_to_utf8(tmpbuf, (UV)c);
+ return to_utf8_fold(tmpbuf, p, lenp);
}
/* for now these all assume no locale info available for Unicode > 255 */
bool
-Perl_is_uni_alnum_lc(pTHX_ U32 c)
+Perl_is_uni_alnum_lc(pTHX_ UV c)
{
return is_uni_alnum(c); /* XXX no locale support yet */
}
bool
-Perl_is_uni_alnumc_lc(pTHX_ U32 c)
+Perl_is_uni_alnumc_lc(pTHX_ UV c)
{
return is_uni_alnumc(c); /* XXX no locale support yet */
}
bool
-Perl_is_uni_idfirst_lc(pTHX_ U32 c)
+Perl_is_uni_idfirst_lc(pTHX_ UV c)
{
return is_uni_idfirst(c); /* XXX no locale support yet */
}
bool
-Perl_is_uni_alpha_lc(pTHX_ U32 c)
+Perl_is_uni_alpha_lc(pTHX_ UV c)
{
return is_uni_alpha(c); /* XXX no locale support yet */
}
bool
-Perl_is_uni_ascii_lc(pTHX_ U32 c)
+Perl_is_uni_ascii_lc(pTHX_ UV c)
{
return is_uni_ascii(c); /* XXX no locale support yet */
}
bool
-Perl_is_uni_space_lc(pTHX_ U32 c)
+Perl_is_uni_space_lc(pTHX_ UV c)
{
return is_uni_space(c); /* XXX no locale support yet */
}
bool
-Perl_is_uni_digit_lc(pTHX_ U32 c)
+Perl_is_uni_digit_lc(pTHX_ UV c)
{
return is_uni_digit(c); /* XXX no locale support yet */
}
bool
-Perl_is_uni_upper_lc(pTHX_ U32 c)
+Perl_is_uni_upper_lc(pTHX_ UV c)
{
return is_uni_upper(c); /* XXX no locale support yet */
}
bool
-Perl_is_uni_lower_lc(pTHX_ U32 c)
+Perl_is_uni_lower_lc(pTHX_ UV c)
{
return is_uni_lower(c); /* XXX no locale support yet */
}
bool
-Perl_is_uni_cntrl_lc(pTHX_ U32 c)
+Perl_is_uni_cntrl_lc(pTHX_ UV c)
{
return is_uni_cntrl(c); /* XXX no locale support yet */
}
bool
-Perl_is_uni_graph_lc(pTHX_ U32 c)
+Perl_is_uni_graph_lc(pTHX_ UV c)
{
return is_uni_graph(c); /* XXX no locale support yet */
}
bool
-Perl_is_uni_print_lc(pTHX_ U32 c)
+Perl_is_uni_print_lc(pTHX_ UV c)
{
return is_uni_print(c); /* XXX no locale support yet */
}
bool
-Perl_is_uni_punct_lc(pTHX_ U32 c)
+Perl_is_uni_punct_lc(pTHX_ UV c)
{
return is_uni_punct(c); /* XXX no locale support yet */
}
bool
-Perl_is_uni_xdigit_lc(pTHX_ U32 c)
+Perl_is_uni_xdigit_lc(pTHX_ UV c)
{
return is_uni_xdigit(c); /* XXX no locale support yet */
}
-U32
-Perl_to_uni_upper_lc(pTHX_ U32 c)
-{
- return to_uni_upper(c); /* XXX no locale support yet */
-}
-
-U32
-Perl_to_uni_title_lc(pTHX_ U32 c)
-{
- return to_uni_title(c); /* XXX no locale support yet */
-}
-
-U32
-Perl_to_uni_lower_lc(pTHX_ U32 c)
-{
- return to_uni_lower(c); /* XXX no locale support yet */
-}
-
bool
Perl_is_utf8_alnum(pTHX_ U8 *p)
{
* 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);
+ 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);
+ return swash_fetch(PL_utf8_alnum, p, TRUE);
#endif
}
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);
+ 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);
+ return swash_fetch(PL_utf8_alnum, p, TRUE);
#endif
}
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);
+ return swash_fetch(PL_utf8_alpha, p, TRUE);
}
bool
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);
+ return swash_fetch(PL_utf8_ascii, p, TRUE);
}
bool
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);
+ return swash_fetch(PL_utf8_space, p, TRUE);
}
bool
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);
+ return swash_fetch(PL_utf8_digit, p, TRUE);
}
bool
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);
+ return swash_fetch(PL_utf8_upper, p, TRUE);
}
bool
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);
+ return swash_fetch(PL_utf8_lower, p, TRUE);
}
bool
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);
+ return swash_fetch(PL_utf8_cntrl, p, TRUE);
}
bool
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);
+ return swash_fetch(PL_utf8_graph, p, TRUE);
}
bool
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);
+ return swash_fetch(PL_utf8_print, p, TRUE);
}
bool
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);
+ return swash_fetch(PL_utf8_punct, p, TRUE);
}
bool
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);
+ return swash_fetch(PL_utf8_xdigit, p, TRUE);
}
bool
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);
+ return swash_fetch(PL_utf8_mark, p, TRUE);
}
+/*
+=for apidoc A|UV|to_utf8_case|U8 *p|U8* ustrp|STRLEN *lenp|SV **swash|char *normal|char *special
+
+The "p" contains the pointer to the UTF-8 string encoding
+the character that is being converted.
+
+The "ustrp" is a pointer to the character buffer to put the
+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 "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.
+
+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().
+
+=cut
+ */
+
UV
-Perl_to_utf8_upper(pTHX_ U8 *p)
+Perl_to_utf8_case(pTHX_ U8 *p, U8* ustrp, STRLEN *lenp, SV **swashp,char *normal, char *special)
{
UV uv;
- if (!PL_utf8_toupper)
- PL_utf8_toupper = swash_init("utf8", "ToUpper", &PL_sv_undef, 4, 0);
- uv = swash_fetch(PL_utf8_toupper, p);
- return uv ? UNI_TO_NATIVE(uv) : utf8_to_uvchr(p,0);
+ if (!*swashp)
+ *swashp = swash_init("utf8", normal, &PL_sv_undef, 4, 0);
+ uv = swash_fetch(*swashp, p, TRUE);
+ if (uv)
+ uv = UNI_TO_NATIVE(uv);
+ else {
+ HV *hv;
+ SV *keysv;
+ HE *he;
+
+ uv = utf8_to_uvchr(p, 0);
+
+ if ((hv = get_hv(special, FALSE)) &&
+ (keysv = sv_2mortal(Perl_newSVpvf(aTHX_ "%04"UVXf, uv))) &&
+ (he = hv_fetch_ent(hv, keysv, FALSE, 0))) {
+ SV *val = HeVAL(he);
+ char *s = SvPV(val, *lenp);
+ U8 c = *(U8*)s;
+ if (*lenp > 1 || UNI_IS_INVARIANT(c))
+ Copy(s, ustrp, *lenp, U8);
+ else {
+ /* something in the 0x80..0xFF range */
+ ustrp[0] = UTF8_EIGHT_BIT_HI(c);
+ ustrp[1] = UTF8_EIGHT_BIT_LO(c);
+ *lenp = 2;
+ }
+ return 0;
+ }
+ }
+ if (lenp)
+ *lenp = UNISKIP(uv);
+ uvuni_to_utf8(ustrp, uv);
+ return uv;
}
UV
-Perl_to_utf8_title(pTHX_ U8 *p)
+Perl_to_utf8_upper(pTHX_ U8 *p, U8* ustrp, STRLEN *lenp)
{
- UV uv;
+ return Perl_to_utf8_case(aTHX_ p, ustrp, lenp,
+ &PL_utf8_toupper, "ToUpper", "utf8::ToSpecUpper");
+}
- if (!PL_utf8_totitle)
- PL_utf8_totitle = swash_init("utf8", "ToTitle", &PL_sv_undef, 4, 0);
- uv = swash_fetch(PL_utf8_totitle, p);
- return uv ? UNI_TO_NATIVE(uv) : utf8_to_uvchr(p,0);
+UV
+Perl_to_utf8_title(pTHX_ U8 *p, U8* ustrp, STRLEN *lenp)
+{
+ return Perl_to_utf8_case(aTHX_ p, ustrp, lenp,
+ &PL_utf8_totitle, "ToTitle", "utf8::ToSpecTitle");
}
UV
-Perl_to_utf8_lower(pTHX_ U8 *p)
+Perl_to_utf8_lower(pTHX_ U8 *p, U8* ustrp, STRLEN *lenp)
{
- UV uv;
+ return Perl_to_utf8_case(aTHX_ p, ustrp, lenp,
+ &PL_utf8_tolower, "ToLower", "utf8::ToSpecLower");
+}
- if (!PL_utf8_tolower)
- PL_utf8_tolower = swash_init("utf8", "ToLower", &PL_sv_undef, 4, 0);
- uv = swash_fetch(PL_utf8_tolower, p);
- return uv ? UNI_TO_NATIVE(uv) : utf8_to_uvchr(p,0);
+UV
+Perl_to_utf8_fold(pTHX_ U8 *p, U8* ustrp, STRLEN *lenp)
+{
+ return Perl_to_utf8_case(aTHX_ p, ustrp, lenp,
+ &PL_utf8_tofold, "ToFold", "utf8::ToSpecFold");
}
/* a "swash" is a swatch hash */
SV* retval;
SV* tokenbufsv = sv_2mortal(NEWSV(0,0));
dSP;
+ HV *stash = gv_stashpvn(pkg, strlen(pkg), FALSE);
+ SV* errsv_save;
- if (!gv_stashpv(pkg, 0)) { /* demand load utf8 */
+ 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);
+ if (!SvTRUE(ERRSV))
+ sv_setsv(ERRSV, errsv_save);
+ SvREFCNT_dec(errsv_save);
LEAVE;
}
SPAGAIN;
if (PL_curcop == &PL_compiling)
/* XXX ought to be handled by lex_start */
sv_setpv(tokenbufsv, PL_tokenbuf);
+ errsv_save = newSVsv(ERRSV);
if (call_method("SWASHNEW", G_SCALAR))
retval = newSVsv(*PL_stack_sp--);
else
retval = &PL_sv_undef;
+ if (!SvTRUE(ERRSV))
+ sv_setsv(ERRSV, errsv_save);
+ SvREFCNT_dec(errsv_save);
LEAVE;
POPSTACK;
if (PL_curcop == &PL_compiling) {
return retval;
}
+
+/* This API is wrong for special case conversions since we may need to
+ * return several Unicode characters for a single Unicode character
+ * (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 */
UV
-Perl_swash_fetch(pTHX_ SV *sv, U8 *ptr)
+Perl_swash_fetch(pTHX_ SV *sv, U8 *ptr, bool do_utf8)
{
HV* hv = (HV*)SvRV(sv);
- U32 klen = UTF8SKIP(ptr) - 1;
- U32 off = ptr[klen] & 127; /* NB: 64 bit always 0 when len > 1 */
+ U32 klen;
+ U32 off;
STRLEN slen;
- STRLEN needents = (klen ? 64 : 128);
- U8 *tmps;
+ STRLEN needents;
+ U8 *tmps = NULL;
U32 bit;
SV *retval;
+ U8 tmputf8[2];
+ 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;
+ }
+ /* Given a UTF-X encoded char 0xAA..0xYY,0xZZ
+ * then the "swatch" is a vec() for al the chars which start
+ * with 0xAA..0xYY
+ * So the key in the hash (klen) is length of encoded char -1
+ */
+ klen = UTF8SKIP(ptr) - 1;
+ off = ptr[klen];
+
+ 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
+ */
+ 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;
+ }
/*
* This single-entry cache saves about 1/3 of the utf8 overhead in test
* NB: this code assumes that swatches are never modified, once generated!
*/
- if (hv == PL_last_swash_hv &&
+ if (hv == PL_last_swash_hv &&
klen == PL_last_swash_klen &&
- (!klen || memEQ((char *)ptr,(char *)PL_last_swash_key,klen)) )
+ (!klen || memEQ((char *)ptr, (char *)PL_last_swash_key, klen)) )
{
tmps = PL_last_swash_tmps;
slen = PL_last_swash_slen;
/* If not cached, generate it via utf8::SWASHGET */
if (!svp || !SvPOK(*svp) || !(tmps = (U8*)SvPV(*svp, slen))) {
dSP;
+ /* 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();
PUSHMARK(SP);
EXTEND(SP,3);
PUSHs((SV*)sv);
- /* We call utf8_to_uni as we want and index into Unicode tables,
- not a native character number.
- */
- PUSHs(sv_2mortal(newSViv(utf8_to_uvuni(ptr, 0) & ~(needents - 1))));
+ /* 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;
svp = hv_store(hv, (char*)ptr, klen, retval, 0);
- if (!svp || !(tmps = (U8*)SvPV(*svp, slen)) || slen < 8)
+ if (!svp || !(tmps = (U8*)SvPV(*svp, slen)) || (slen << 3) < needents)
Perl_croak(aTHX_ "SWASHGET didn't return result of proper length");
}
Perl_croak(aTHX_ "panic: swash_fetch");
return 0;
}
+
+
+/*
+=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
+bytes available. The return value is the pointer to the byte after the
+end of the new character. In other words,
+
+ d = uvchr_to_utf8(d, uv);
+
+is the recommended wide native character-aware way of saying
+
+ *(d++) = uv;
+
+=cut
+*/
+
+/* 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)
+{
+ return Perl_uvuni_to_utf8(aTHX_ d, NATIVE_TO_UNI(uv));
+}
+
+
+/*
+=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
+length, in bytes, of that character.
+
+Allows length and flags to be passed to low level routine.
+
+=cut
+*/
+/* 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)
+{
+ UV uv = Perl_utf8n_to_uvuni(aTHX_ s, curlen, retlen, flags);
+ return UNI_TO_NATIVE(uv);
+}
+
+/*
+=for apidoc A|char *|pv_uni_display|SV *dsv|U8 *spv|STRLEN len|STRLEN pvlim|UV flags
+
+Build to the scalar dsv a displayable version of the string spv,
+length len, the displayable version being at most pvlim bytes long
+(if longer, the rest is truncated and "..." will be appended).
+The flags argument is currently unused but available for future extensions.
+The pointer to the PV of the dsv is returned.
+
+=cut */
+char *
+Perl_pv_uni_display(pTHX_ SV *dsv, U8 *spv, STRLEN len, STRLEN pvlim, UV flags)
+{
+ int truncated = 0;
+ char *s, *e;
+
+ sv_setpvn(dsv, "", 0);
+ for (s = (char *)spv, e = s + len; s < e; s += UTF8SKIP(s)) {
+ UV u;
+ if (pvlim && SvCUR(dsv) >= pvlim) {
+ truncated++;
+ break;
+ }
+ u = utf8_to_uvchr((U8*)s, 0);
+ Perl_sv_catpvf(aTHX_ dsv, "\\x{%"UVxf"}", u);
+ }
+ if (truncated)
+ sv_catpvn(dsv, "...", 3);
+
+ return SvPVX(dsv);
+}
+
+/*
+=for apidoc A|char *|sv_uni_display|SV *dsv|SV *ssv|STRLEN pvlim|UV flags
+
+Build to the scalar dsv a displayable version of the scalar sv,
+he displayable version being at most pvlim bytes long
+(if longer, the rest is truncated and "..." will be appended).
+The flags argument is currently unused but available for future extensions.
+The pointer to the PV of the dsv is returned.
+
+=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);
+}
+
+/*
+=for apidoc A|I32|ibcmp_utf8|const char *s1|bool u1|const char *s2|bool u2|register I32 len
+
+Return true if the strings s1 and s2 differ case-insensitively, false
+if not (if they are equal case-insensitively). If u1 is true, the
+string s1 is assumed to be in UTF-8-encoded Unicode. If u2 is true,
+the string s2 is assumed to be in UTF-8-encoded Unicode.
+
+For case-insensitiveness, the "casefolding" of Unicode is used
+instead of upper/lowercasing both the characters, see
+http://www.unicode.org/unicode/reports/tr21/ (Case Mappings).
+
+=cut */
+I32
+Perl_ibcmp_utf8(pTHX_ const char *s1, bool u1, register I32 len1, const char *s2, bool u2, register I32 len2)
+{
+ register U8 *a = (U8*)s1;
+ register U8 *b = (U8*)s2;
+ register U8 *ae = b + len1;
+ register U8 *be = b + len2;
+ STRLEN la, lb;
+ UV ca, cb;
+ STRLEN ulen1, ulen2;
+ U8 tmpbuf1[UTF8_MAXLEN_FOLD+1];
+ U8 tmpbuf2[UTF8_MAXLEN_FOLD+1];
+
+ while (a < ae && b < be) {
+ if (u1) {
+ if (a + UTF8SKIP(a) > ae)
+ break;
+ ca = utf8_to_uvchr((U8*)a, &la);
+ } else {
+ ca = *a;
+ la = 1;
+ }
+ if (u2) {
+ if (b + UTF8SKIP(b) > be)
+ break;
+ cb = utf8_to_uvchr((U8*)b, &lb);
+ } else {
+ cb = *b;
+ lb = 1;
+ }
+ if (ca != cb) {
+ if (u1)
+ to_uni_fold(NATIVE_TO_UNI(ca), tmpbuf1, &ulen1);
+ else
+ ulen1 = 1;
+ if (u2)
+ to_uni_fold(NATIVE_TO_UNI(cb), tmpbuf2, &ulen2);
+ else
+ ulen2 = 1;
+ if (ulen1 != ulen2
+ || (ulen1 == 1 && PL_fold[ca] != PL_fold[cb])
+ || memNE((char *)tmpbuf1, (char *)tmpbuf2, ulen1))
+ return 1; /* mismatch */
+ }
+ a += la;
+ b += lb;
+ }
+ return a == ae && b == be ? 0 : 1; /* 0 match, 1 mismatch */
+}
+