/* Unicode support */
U8 *
-Perl_uv_to_utf8(pTHX_ U8 *d, UV uv)
+Perl_uv_to_utf8(pTHX_ U8 *d, UV uv) /* the d must be UTF8_MAXLEN+1 deep */
{
if (uv < 0x80) {
*d++ = uv;
+ *d = 0;
return d;
}
if (uv < 0x800) {
*d++ = (( uv >> 6) | 0xc0);
*d++ = (( uv & 0x3f) | 0x80);
+ *d = 0;
return d;
}
if (uv < 0x10000) {
*d++ = (( uv >> 12) | 0xe0);
*d++ = (((uv >> 6) & 0x3f) | 0x80);
*d++ = (( uv & 0x3f) | 0x80);
+ *d = 0;
return d;
}
if (uv < 0x200000) {
*d++ = (((uv >> 12) & 0x3f) | 0x80);
*d++ = (((uv >> 6) & 0x3f) | 0x80);
*d++ = (( uv & 0x3f) | 0x80);
+ *d = 0;
return d;
}
if (uv < 0x4000000) {
*d++ = (((uv >> 12) & 0x3f) | 0x80);
*d++ = (((uv >> 6) & 0x3f) | 0x80);
*d++ = (( uv & 0x3f) | 0x80);
+ *d = 0;
return d;
}
if (uv < 0x80000000) {
*d++ = (((uv >> 12) & 0x3f) | 0x80);
*d++ = (((uv >> 6) & 0x3f) | 0x80);
*d++ = (( uv & 0x3f) | 0x80);
+ *d = 0;
return d;
}
#ifdef HAS_QUAD
- if (uv < 0x1000000000LL)
+ if (uv < UTF8_QUAD_MAX)
#endif
{
*d++ = 0xfe; /* Can't match U+FEFF! */
*d++ = (((uv >> 12) & 0x3f) | 0x80);
*d++ = (((uv >> 6) & 0x3f) | 0x80);
*d++ = (( uv & 0x3f) | 0x80);
+ *d = 0;
return d;
}
#ifdef HAS_QUAD
*d++ = (((uv >> 12) & 0x3f) | 0x80);
*d++ = (((uv >> 6) & 0x3f) | 0x80);
*d++ = (( uv & 0x3f) | 0x80);
+ *d = 0;
return d;
}
#endif
while (slen--) {
if ((*s & 0xc0) != 0x80)
return 0;
- uv = (uv << 6) | (*s & 0x3f);
+ uv = UTF8_ACCUMULATE(uv, *s);
if (uv < ouv)
return 0;
ouv = uv;
}
/*
-=for apidoc Am|U8* s|utf8_to_uv_chk|STRLEN curlen|I32 *retlen|I32 checking
+=for apidoc Am|U8* s|utf8_to_uv|STRLEN curlen|STRLEN *retlen|U32 flags
Returns the character value of the first character in the string C<s>
which is assumed to be in UTF8 encoding and no longer than C<curlen>;
-C<retlen> will be set to the length, in bytes, of that character,
-and the pointer C<s> will be advanced to the end of the character.
+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
-is dependent on the value of C<checking>: if this is true, it is
-assumed that the caller will raise a warning, and this function will
-set C<retlen> to C<-1> and return. If C<checking> is not true, an optional UTF8
-warning is produced.
+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
+C<flags> does not contain UTF8_CHECK_ONLY, warnings about
+malformations will be given, C<retlen> will be set to the expected
+length of the UTF-8 character in bytes, and zero will be returned.
-=cut
-*/
+The C<flags> can also contain various flags to allow deviations from
+the strict UTF-8 encoding (see F<utf8.h>).
+
+=cut */
UV
-Perl_utf8_to_uv_chk(pTHX_ U8* s, STRLEN curlen, STRLEN* retlen, bool checking)
+Perl_utf8_to_uv(pTHX_ U8* s, STRLEN curlen, STRLEN* retlen, U32 flags)
{
- dTHR;
UV uv = *s, ouv;
STRLEN len = 1;
+#ifdef EBCDIC
+ bool dowarn = 0;
+#else
bool dowarn = ckWARN_d(WARN_UTF8);
+#endif
STRLEN expectlen = 0;
- if (uv <= 0x7f) { /* Pure ASCII. */
+ if (curlen == 0) {
+ if (dowarn)
+ Perl_warner(aTHX_ WARN_UTF8,
+ "Malformed UTF-8 character (an empty string)");
+ goto malformed;
+ }
+
+ if (UTF8_IS_ASCII(uv)) {
if (retlen)
*retlen = 1;
return *s;
}
- if (uv >= 0x80 && uv <= 0xbf) {
+ if (UTF8_IS_CONTINUATION(uv) &&
+ !(flags & UTF8_ALLOW_CONTINUATION)) {
if (dowarn)
Perl_warner(aTHX_ WARN_UTF8,
- "Malformed UTF-8 character (unexpected continuation byte 0x%02x)",
+ "Malformed UTF-8 character (unexpected continuation byte 0x%02"UVxf")",
uv);
goto malformed;
}
- if (uv >= 0xc0 && uv <= 0xfd && curlen > 1 && s[1] < 0x80) {
+ if (UTF8_IS_START(uv) && curlen > 1 && !UTF8_IS_CONTINUATION(s[1]) &&
+ !(flags & UTF8_ALLOW_NON_CONTINUATION)) {
if (dowarn)
Perl_warner(aTHX_ WARN_UTF8,
- "Malformed UTF-8 character (unexpected non-continuation byte 0x%02x after byte 0x%02x)",
- s[1], uv);
+ "Malformed UTF-8 character (unexpected non-continuation byte 0x%02"UVxf" after start byte 0x%02"UVxf")",
+ (UV)s[1], uv);
goto malformed;
}
-
- if ((uv == 0xfe || uv == 0xff) && IN_UTF8){
+
+ if ((uv == 0xfe || uv == 0xff) &&
+ !(flags & UTF8_ALLOW_FE_FF)) {
if (dowarn)
Perl_warner(aTHX_ WARN_UTF8,
- "Malformed UTF-8 character (impossible byte 0x%02x)",
+ "Malformed UTF-8 character (byte 0x%02"UVxf")",
uv);
goto malformed;
}
-
+
if (!(uv & 0x20)) { len = 2; uv &= 0x1f; }
else if (!(uv & 0x10)) { len = 3; uv &= 0x0f; }
else if (!(uv & 0x08)) { len = 4; uv &= 0x07; }
else if (!(uv & 0x02)) { len = 6; uv &= 0x01; }
else if (!(uv & 0x01)) { len = 7; uv = 0; }
else { len = 13; uv = 0; } /* whoa! */
-
+
if (retlen)
*retlen = len;
expectlen = len;
- if (curlen < expectlen) {
+ if ((curlen < expectlen) &&
+ !(flags & UTF8_ALLOW_SHORT)) {
if (dowarn)
Perl_warner(aTHX_ WARN_UTF8,
"Malformed UTF-8 character (%d byte%s, need %d)",
- curlen, curlen > 1 ? "s" : "", expectlen);
+ curlen, curlen == 1 ? "" : "s", expectlen);
goto malformed;
}
ouv = uv;
while (len--) {
- if ((*s & 0xc0) != 0x80) {
+ if (!UTF8_IS_CONTINUATION(*s) &&
+ !(flags & UTF8_ALLOW_NON_CONTINUATION)) {
if (dowarn)
Perl_warner(aTHX_ WARN_UTF8,
- "Malformed UTF-8 character (unexpected continuation byte 0x%02x)",
+ "Malformed UTF-8 character (unexpected non-continuation byte 0x%02x)",
*s);
goto malformed;
}
else
- uv = (uv << 6) | (*s & 0x3f);
+ uv = UTF8_ACCUMULATE(uv, *s);
if (uv < ouv) {
+ /* This cannot be allowed. */
if (dowarn)
Perl_warner(aTHX_ WARN_UTF8,
"Malformed UTF-8 character (overflow at 0x%"UVxf", byte 0x%02x)",
ouv = uv;
}
- if (uv >= 0xd800 && uv <= 0xdfff) {
+ if (UNICODE_IS_SURROGATE(uv) &&
+ !(flags & UTF8_ALLOW_SURROGATE)) {
if (dowarn)
Perl_warner(aTHX_ WARN_UTF8,
"Malformed UTF-8 character (UTF-16 surrogate 0x%04"UVxf")",
uv);
goto malformed;
- } else if (uv == 0xfffe) {
+ } else if (UNICODE_IS_BYTE_ORDER_MARK(uv) &&
+ !(flags & UTF8_ALLOW_BOM)) {
if (dowarn)
Perl_warner(aTHX_ WARN_UTF8,
"Malformed UTF-8 character (byte order mark 0x%04"UVxf")",
uv);
goto malformed;
- } else if (uv == 0xffff && IN_UTF8) {
+ } else if ((expectlen > UNISKIP(uv)) &&
+ !(flags & UTF8_ALLOW_LONG)) {
if (dowarn)
Perl_warner(aTHX_ WARN_UTF8,
- "Malformed UTF-8 character (impossible character 0x%04"UVxf")",
- uv);
+ "Malformed UTF-8 character (%d byte%s, need %d)",
+ expectlen, expectlen == 1 ? "": "s", UNISKIP(uv));
goto malformed;
- } else if (expectlen > UNISKIP(uv)) {
+ } else if (UNICODE_IS_ILLEGAL(uv) &&
+ !(flags & UTF8_ALLOW_FFFF)) {
if (dowarn)
Perl_warner(aTHX_ WARN_UTF8,
- "Malformed UTF-8 character (%d byte%s, need %d)",
- expectlen, expectlen > 1 ? "s": "", UNISKIP(uv));
+ "Malformed UTF-8 character (character 0x%04"UVxf")",
+ uv);
goto malformed;
}
malformed:
- if (checking) {
+ if (flags & UTF8_CHECK_ONLY) {
if (retlen)
- *retlen = len;
+ *retlen = -1;
return 0;
}
if (retlen)
- *retlen = -1;
+ *retlen = expectlen ? expectlen : len;
- return UNICODE_REPLACEMENT_CHARACTER;
+ return 0;
}
/*
-=for apidoc Am|U8* s|utf8_to_uv|STRLEN *retlen
+=for apidoc Am|U8* s|utf8_to_uv_simple|STRLEN *retlen
Returns the 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, and the pointer C<s> will be
-advanced to the end of the character.
+length, in bytes, of that character.
If C<s> does not point to a well-formed UTF8 character, zero is
returned and retlen is set, if possible, to -1.
*/
UV
-Perl_utf8_to_uv(pTHX_ U8* s, STRLEN* retlen)
+Perl_utf8_to_uv_simple(pTHX_ U8* s, STRLEN* retlen)
{
- return Perl_utf8_to_uv_chk(aTHX_ s, (STRLEN)-1, retlen, 0);
+ return Perl_utf8_to_uv(aTHX_ s, UTF8_MAXLEN, retlen, 0);
}
-/* utf8_distance(a,b) returns the number of UTF8 characters between
- the pointers a and b */
+/*
+=for apidoc Am|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
+up past C<e>, croaks.
+
+=cut
+*/
+
+STRLEN
+Perl_utf8_length(pTHX_ U8* s, U8* e)
+{
+ STRLEN len = 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");
+ while (s < e) {
+ U8 t = UTF8SKIP(s);
+
+ if (e - s < t)
+ Perl_croak(aTHX_ "panic: utf8_length: unaligned end");
+ s += t;
+ len++;
+ }
-I32
+ return len;
+}
+
+/*
+=for apidoc Am|IV|utf8_distance|U8 *a|U8 *b
+
+Returns the number of UTF8 characters between the UTF-8 pointers C<a>
+and C<b>.
+
+WARNING: use only if you *know* that the pointers point inside the
+same UTF-8 buffer.
+
+=cut */
+
+IV
Perl_utf8_distance(pTHX_ U8 *a, U8 *b)
{
- I32 off = 0;
+ 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) {
- a += UTF8SKIP(a);
+ U8 c = UTF8SKIP(a);
+
+ if (b - a < c)
+ Perl_croak(aTHX_ "panic: utf8_distance: unaligned end");
+ a += c;
off--;
}
}
else {
while (b < a) {
- b += UTF8SKIP(b);
+ U8 c = UTF8SKIP(b);
+
+ if (a - b < c)
+ Perl_croak(aTHX_ "panic: utf8_distance: unaligned end");
+ b += c;
off++;
}
}
+
return off;
}
-/* WARNING: do not use the following unless you *know* off is within bounds */
+/*
+=for apidoc Am|U8*|utf8_hop|U8 *s|I32 off
+
+Return the UTF-8 pointer C<s> displaced by C<off> characters, either
+forward or backward.
+
+WARNING: do not use the following unless you *know* C<off> is within
+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 */
U8 *
Perl_utf8_hop(pTHX_ U8 *s, I32 off)
{
+ /* 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 (off >= 0) {
while (off--)
s += UTF8SKIP(s);
else {
while (off++) {
s--;
- if (*s & 0x80) {
- while ((*s & 0xc0) == 0x80)
- s--;
- }
+ while (UTF8_IS_CONTINUATION(*s))
+ s--;
}
}
return s;
U8 *
Perl_utf8_to_bytes(pTHX_ U8* s, STRLEN *len)
{
- dTHR;
U8 *send;
U8 *d;
- U8 *save;
-
- send = s + *len;
- d = save = s;
+ U8 *save = s;
/* ensure valid UTF8 and chars < 256 before updating string */
- while (s < send) {
- U8 c = *s++;
+ for (send = s + *len; s < send; ) {
+ U8 c = *s++;
+
if (c >= 0x80 &&
- ( (s >= send) || ((*s++ & 0xc0) != 0x80) || ((c & 0xfe) != 0xc2))) {
- *len = -1;
- return 0;
- }
+ ((s >= send) ||
+ ((*s++ & 0xc0) != 0x80) || ((c & 0xfe) != 0xc2))) {
+ *len = -1;
+ return 0;
+ }
}
- s = save;
+
+ d = s = save;
while (s < send) {
- if (*s < 0x80)
- *d++ = *s++;
- else {
- STRLEN ulen;
- *d++ = (U8)utf8_to_uv(s, &ulen);
- s += ulen;
- }
+ STRLEN ulen;
+ *d++ = (U8)utf8_to_uv_simple(s, &ulen);
+ s += ulen;
}
*d = '\0';
*len = d - save;
U8*
Perl_bytes_to_utf8(pTHX_ U8* s, STRLEN *len)
{
- dTHR;
U8 *send;
U8 *d;
U8 *dst;
continue;
}
if (uv >= 0xd800 && uv < 0xdbff) { /* surrogates */
- dTHR;
UV low = *p++;
if (low < 0xdc00 || low >= 0xdfff)
Perl_croak(aTHX_ "Malformed UTF-16 surrogate");
bool
Perl_is_uni_alnum(pTHX_ U32 c)
{
- U8 tmpbuf[UTF8_MAXLEN];
+ U8 tmpbuf[UTF8_MAXLEN+1];
uv_to_utf8(tmpbuf, (UV)c);
return is_utf8_alnum(tmpbuf);
}
bool
Perl_is_uni_alnumc(pTHX_ U32 c)
{
- U8 tmpbuf[UTF8_MAXLEN];
+ U8 tmpbuf[UTF8_MAXLEN+1];
uv_to_utf8(tmpbuf, (UV)c);
return is_utf8_alnumc(tmpbuf);
}
bool
Perl_is_uni_idfirst(pTHX_ U32 c)
{
- U8 tmpbuf[UTF8_MAXLEN];
+ U8 tmpbuf[UTF8_MAXLEN+1];
uv_to_utf8(tmpbuf, (UV)c);
return is_utf8_idfirst(tmpbuf);
}
bool
Perl_is_uni_alpha(pTHX_ U32 c)
{
- U8 tmpbuf[UTF8_MAXLEN];
+ U8 tmpbuf[UTF8_MAXLEN+1];
uv_to_utf8(tmpbuf, (UV)c);
return is_utf8_alpha(tmpbuf);
}
bool
Perl_is_uni_ascii(pTHX_ U32 c)
{
- U8 tmpbuf[UTF8_MAXLEN];
+ U8 tmpbuf[UTF8_MAXLEN+1];
uv_to_utf8(tmpbuf, (UV)c);
return is_utf8_ascii(tmpbuf);
}
bool
Perl_is_uni_space(pTHX_ U32 c)
{
- U8 tmpbuf[UTF8_MAXLEN];
+ U8 tmpbuf[UTF8_MAXLEN+1];
uv_to_utf8(tmpbuf, (UV)c);
return is_utf8_space(tmpbuf);
}
bool
Perl_is_uni_digit(pTHX_ U32 c)
{
- U8 tmpbuf[UTF8_MAXLEN];
+ U8 tmpbuf[UTF8_MAXLEN+1];
uv_to_utf8(tmpbuf, (UV)c);
return is_utf8_digit(tmpbuf);
}
bool
Perl_is_uni_upper(pTHX_ U32 c)
{
- U8 tmpbuf[UTF8_MAXLEN];
+ U8 tmpbuf[UTF8_MAXLEN+1];
uv_to_utf8(tmpbuf, (UV)c);
return is_utf8_upper(tmpbuf);
}
bool
Perl_is_uni_lower(pTHX_ U32 c)
{
- U8 tmpbuf[UTF8_MAXLEN];
+ U8 tmpbuf[UTF8_MAXLEN+1];
uv_to_utf8(tmpbuf, (UV)c);
return is_utf8_lower(tmpbuf);
}
bool
Perl_is_uni_cntrl(pTHX_ U32 c)
{
- U8 tmpbuf[UTF8_MAXLEN];
+ U8 tmpbuf[UTF8_MAXLEN+1];
uv_to_utf8(tmpbuf, (UV)c);
return is_utf8_cntrl(tmpbuf);
}
bool
Perl_is_uni_graph(pTHX_ U32 c)
{
- U8 tmpbuf[UTF8_MAXLEN];
+ U8 tmpbuf[UTF8_MAXLEN+1];
uv_to_utf8(tmpbuf, (UV)c);
return is_utf8_graph(tmpbuf);
}
bool
Perl_is_uni_print(pTHX_ U32 c)
{
- U8 tmpbuf[UTF8_MAXLEN];
+ U8 tmpbuf[UTF8_MAXLEN+1];
uv_to_utf8(tmpbuf, (UV)c);
return is_utf8_print(tmpbuf);
}
bool
Perl_is_uni_punct(pTHX_ U32 c)
{
- U8 tmpbuf[UTF8_MAXLEN];
+ U8 tmpbuf[UTF8_MAXLEN+1];
uv_to_utf8(tmpbuf, (UV)c);
return is_utf8_punct(tmpbuf);
}
bool
Perl_is_uni_xdigit(pTHX_ U32 c)
{
- U8 tmpbuf[UTF8_MAXLEN];
+ U8 tmpbuf[UTF8_MAXLEN+1];
uv_to_utf8(tmpbuf, (UV)c);
return is_utf8_xdigit(tmpbuf);
}
U32
Perl_to_uni_upper(pTHX_ U32 c)
{
- U8 tmpbuf[UTF8_MAXLEN];
+ U8 tmpbuf[UTF8_MAXLEN+1];
uv_to_utf8(tmpbuf, (UV)c);
return to_utf8_upper(tmpbuf);
}
U32
Perl_to_uni_title(pTHX_ U32 c)
{
- U8 tmpbuf[UTF8_MAXLEN];
+ U8 tmpbuf[UTF8_MAXLEN+1];
uv_to_utf8(tmpbuf, (UV)c);
return to_utf8_title(tmpbuf);
}
U32
Perl_to_uni_lower(pTHX_ U32 c)
{
- U8 tmpbuf[UTF8_MAXLEN];
+ U8 tmpbuf[UTF8_MAXLEN+1];
uv_to_utf8(tmpbuf, (UV)c);
return to_utf8_lower(tmpbuf);
}
if (!is_utf8_char(p))
return FALSE;
if (!PL_utf8_space)
- PL_utf8_space = swash_init("utf8", "IsSpace", &PL_sv_undef, 0, 0);
+ PL_utf8_space = swash_init("utf8", "IsSpacePerl", &PL_sv_undef, 0, 0);
return swash_fetch(PL_utf8_space, p);
}
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 ? uv : utf8_to_uv_chk(p,STRLEN_MAX,0,0);
+ return uv ? uv : utf8_to_uv(p,UTF8_MAXLEN,0,0);
}
UV
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 ? uv : utf8_to_uv_chk(p,STRLEN_MAX,0,0);
+ return uv ? uv : utf8_to_uv(p,UTF8_MAXLEN,0,0);
}
UV
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 ? uv : utf8_to_uv_chk(p,STRLEN_MAX,0,0);
+ return uv ? uv : utf8_to_uv(p,UTF8_MAXLEN,0,0);
}
/* a "swash" is a swatch hash */
PUSHMARK(SP);
EXTEND(SP,3);
PUSHs((SV*)sv);
- PUSHs(sv_2mortal(newSViv(utf8_to_uv_chk(ptr, STRLEN_MAX, 0, 0) & ~(needents - 1))));
+ PUSHs(sv_2mortal(newSViv(utf8_to_uv(ptr, UTF8_MAXLEN, 0, 0) & ~(needents - 1))));
PUSHs(sv_2mortal(newSViv(needents)));
PUTBACK;
if (call_method("SWASHGET", G_SCALAR))
Copy(ptr, PL_last_swash_key, klen, U8);
}
- switch ((slen << 3) / needents) {
+ switch ((int)((slen << 3) / needents)) {
case 1:
bit = 1 << (off & 7);
off >>= 3;