#define PERL_IN_UTF8_C
#include "perl.h"
+static char unees[] = "Malformed UTF-8 character (unexpected end of string)";
+
/*
=head1 Unicode Support
U8 *
Perl_uvuni_to_utf8_flags(pTHX_ U8 *d, UV uv, UV flags)
{
- if (ckWARN_d(WARN_UTF8)) {
+ 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 <= 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)) {
char *s = SvPVX(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);
}
}
* 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");
+ if (e < s) {
+ 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 0;
+ }
while (s < e) {
U8 t = UTF8SKIP(s);
- if (e - s < t)
- Perl_croak(aTHX_ "panic: utf8_length: unaligned end");
+ if (e - s < t) {
+ if (ckWARN_d(WARN_UTF8)) {
+ if (PL_op)
+ Perl_warner(aTHX_ packWARN(WARN_UTF8),
+ unees, OP_DESC(PL_op));
+ else
+ Perl_warner(aTHX_ packWARN(WARN_UTF8), unees);
+ }
+ return len;
+ }
s += t;
len++;
}
while (a < b) {
U8 c = UTF8SKIP(a);
- if (b - a < c)
- Perl_croak(aTHX_ "panic: utf8_distance: unaligned end");
+ if (b - a < c) {
+ 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 off;
+ }
a += c;
off--;
}
while (b < a) {
U8 c = UTF8SKIP(b);
- if (a - b < c)
- Perl_croak(aTHX_ "panic: utf8_distance: unaligned end");
+ if (a - b < c) {
+ 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 off;
+ }
b += c;
off++;
}
Perl_is_uni_alnum(pTHX_ UV c)
{
U8 tmpbuf[UTF8_MAXLEN+1];
- uvchr_to_utf8(tmpbuf, (UV)c);
+ uvchr_to_utf8(tmpbuf, c);
return is_utf8_alnum(tmpbuf);
}
Perl_is_uni_alnumc(pTHX_ UV c)
{
U8 tmpbuf[UTF8_MAXLEN+1];
- uvchr_to_utf8(tmpbuf, (UV)c);
+ uvchr_to_utf8(tmpbuf, c);
return is_utf8_alnumc(tmpbuf);
}
Perl_is_uni_idfirst(pTHX_ UV c)
{
U8 tmpbuf[UTF8_MAXLEN+1];
- uvchr_to_utf8(tmpbuf, (UV)c);
+ uvchr_to_utf8(tmpbuf, c);
return is_utf8_idfirst(tmpbuf);
}
Perl_is_uni_alpha(pTHX_ UV c)
{
U8 tmpbuf[UTF8_MAXLEN+1];
- uvchr_to_utf8(tmpbuf, (UV)c);
+ uvchr_to_utf8(tmpbuf, c);
return is_utf8_alpha(tmpbuf);
}
Perl_is_uni_ascii(pTHX_ UV c)
{
U8 tmpbuf[UTF8_MAXLEN+1];
- uvchr_to_utf8(tmpbuf, (UV)c);
+ uvchr_to_utf8(tmpbuf, c);
return is_utf8_ascii(tmpbuf);
}
Perl_is_uni_space(pTHX_ UV c)
{
U8 tmpbuf[UTF8_MAXLEN+1];
- uvchr_to_utf8(tmpbuf, (UV)c);
+ uvchr_to_utf8(tmpbuf, c);
return is_utf8_space(tmpbuf);
}
Perl_is_uni_digit(pTHX_ UV c)
{
U8 tmpbuf[UTF8_MAXLEN+1];
- uvchr_to_utf8(tmpbuf, (UV)c);
+ uvchr_to_utf8(tmpbuf, c);
return is_utf8_digit(tmpbuf);
}
Perl_is_uni_upper(pTHX_ UV c)
{
U8 tmpbuf[UTF8_MAXLEN+1];
- uvchr_to_utf8(tmpbuf, (UV)c);
+ uvchr_to_utf8(tmpbuf, c);
return is_utf8_upper(tmpbuf);
}
Perl_is_uni_lower(pTHX_ UV c)
{
U8 tmpbuf[UTF8_MAXLEN+1];
- uvchr_to_utf8(tmpbuf, (UV)c);
+ uvchr_to_utf8(tmpbuf, c);
return is_utf8_lower(tmpbuf);
}
Perl_is_uni_cntrl(pTHX_ UV c)
{
U8 tmpbuf[UTF8_MAXLEN+1];
- uvchr_to_utf8(tmpbuf, (UV)c);
+ uvchr_to_utf8(tmpbuf, c);
return is_utf8_cntrl(tmpbuf);
}
Perl_is_uni_graph(pTHX_ UV c)
{
U8 tmpbuf[UTF8_MAXLEN+1];
- uvchr_to_utf8(tmpbuf, (UV)c);
+ uvchr_to_utf8(tmpbuf, c);
return is_utf8_graph(tmpbuf);
}
Perl_is_uni_print(pTHX_ UV c)
{
U8 tmpbuf[UTF8_MAXLEN+1];
- uvchr_to_utf8(tmpbuf, (UV)c);
+ uvchr_to_utf8(tmpbuf, c);
return is_utf8_print(tmpbuf);
}
Perl_is_uni_punct(pTHX_ UV c)
{
U8 tmpbuf[UTF8_MAXLEN+1];
- uvchr_to_utf8(tmpbuf, (UV)c);
+ uvchr_to_utf8(tmpbuf, c);
return is_utf8_punct(tmpbuf);
}
Perl_is_uni_xdigit(pTHX_ UV c)
{
U8 tmpbuf[UTF8_MAXLEN_UCLC+1];
- uvchr_to_utf8(tmpbuf, (UV)c);
+ 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, (UV)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, (UV)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, (UV)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, (UV)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 */
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 SWASHGET, 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_ U8 *p, U8* ustrp, STRLEN *lenp, SV **swashp, char *normal, char *special)
{
- UV uv;
+ UV uv0, uv1;
+ U8 tmpbuf[UTF8_MAXLEN_FOLD+1];
+ STRLEN len = 0;
+
+ uv0 = utf8_to_uvchr(p, 0);
+ /* 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);
+ uvuni_to_utf8(tmpbuf, uv1);
- if (!*swashp)
- *swashp = swash_init("utf8", normal, &PL_sv_undef, 4, 0);
- uv = swash_fetch(*swashp, p, TRUE);
- if (!uv) {
+ if (!*swashp) /* load on-demand */
+ *swashp = swash_init("utf8", normal, &PL_sv_undef, 4, 0);
+
+ if (special) {
+ /* It might be "special" (sometimes, but not always,
+ * a multicharacter mapping) */
HV *hv;
SV *keysv;
HE *he;
-
- uv = utf8_to_uvchr(p, 0);
-
+ SV *val;
+
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);
+ (keysv = sv_2mortal(Perl_newSVpvf(aTHX_ "%04"UVXf, uv1))) &&
+ (he = hv_fetch_ent(hv, keysv, FALSE, 0)) &&
+ (val = HeVAL(he))) {
+ char *s;
+
+ s = SvPV(val, len);
+ if (len == 1)
+ len = uvuni_to_utf8(ustrp, NATIVE_TO_UNI(*(U8*)s)) - ustrp;
else {
- /* something in the 0x80..0xFF range */
- ustrp[0] = UTF8_EIGHT_BIT_HI(c);
- ustrp[1] = UTF8_EIGHT_BIT_LO(c);
- *lenp = 2;
- }
#ifdef EBCDIC
- {
- U8 tmpbuf[UTF8_MAXLEN_FOLD+1];
- U8 *d = tmpbuf;
- U8 *t, *tend;
- STRLEN tlen;
-
- for (t = ustrp, tend = t + *lenp; t < tend; t += tlen) {
- UV c = utf8_to_uvchr(t, &tlen);
- d = uvchr_to_utf8(d, UNI_TO_NATIVE(c));
+ /* 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(val)) {
+ STRLEN tlen = 0;
+
+ while (t < tend) {
+ UV c = utf8_to_uvchr(t, &tlen);
+ if (tlen > 0) {
+ d = uvchr_to_utf8(d, UNI_TO_NATIVE(c));
+ t += tlen;
+ }
+ else
+ break;
+ }
}
- *lenp = d - tmpbuf;
- Copy(tmpbuf, ustrp, *lenp, 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
- return utf8_to_uvchr(ustrp, 0);
+ }
+ }
+ }
+
+ if (!len && *swashp) {
+ UV uv2 = swash_fetch(*swashp, tmpbuf, TRUE);
+
+ if (uv2) {
+ /* It was "normal" (a single character mapping). */
+ UV uv3 = UNI_TO_NATIVE(uv2);
+
+ len = uvchr_to_utf8(ustrp, uv3) - ustrp;
}
}
+
+ if (!len) /* Neither: just copy. */
+ len = uvchr_to_utf8(ustrp, uv0) - ustrp;
+
if (lenp)
- *lenp = UNISKIP(uv);
- uvuni_to_utf8(ustrp, uv);
- return uv;
+ *lenp = len;
+
+ return len ? utf8_to_uvchr(ustrp, 0) : 0;
}
/*
case '\a':
Perl_sv_catpvf(aTHX_ dsv, "\\a"); ok = TRUE; break;
case '\\':
- Perl_sv_catpvf(aTHX_ dsv, "\\" ); ok = TRUE; break;
+ Perl_sv_catpvf(aTHX_ dsv, "\\\\" ); ok = TRUE; break;
default: break;
}
}
/* isPRINT() is the locale-blind version. */
if (!ok && (flags & UNI_DISPLAY_ISPRINT) && isPRINT(u & 0xFF)) {
- Perl_sv_catpvf(aTHX_ dsv, "%c", u);
+ Perl_sv_catpvf(aTHX_ dsv, "%c", (char)(u & 0xFF));
ok = TRUE;
}
}
if ((e1 == 0 && f1 == 0) || (e2 == 0 && f2 == 0) || (f1 == 0 && f2 == 0))
return 1; /* mismatch; possible infinite loop or false positive */
+ if (!u1 || !u2)
+ natbuf[1] = 0; /* Need to terminate the buffer. */
+
while ((e1 == 0 || p1 < e1) &&
(f1 == 0 || p1 < f1) &&
(e2 == 0 || p2 < e2) &&
if (u1)
to_utf8_fold(p1, foldbuf1, &foldlen1);
else {
- natbuf[0] = NATIVE_TO_UNI(*p1);
+ natbuf[0] = *p1;
to_utf8_fold(natbuf, foldbuf1, &foldlen1);
}
q1 = foldbuf1;
if (u2)
to_utf8_fold(p2, foldbuf2, &foldlen2);
else {
- natbuf[0] = NATIVE_TO_UNI(*p2);
+ natbuf[0] = *p2;
to_utf8_fold(natbuf, foldbuf2, &foldlen2);
}
q2 = foldbuf2;