/* utf8.c
*
- * Copyright (C) 2000, 2001, 2002, 2003, 2004, 2005, 2006,
+ * 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
#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)";
=cut */
STATIC STRLEN
-S_is_utf8_char_slow(pTHX_ const U8 *s, const STRLEN len)
+S_is_utf8_char_slow(const U8 *s, const STRLEN len)
{
U8 u = *s;
STRLEN slen;
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;
bool
Perl_is_utf8_string(pTHX_ const U8 *s, STRLEN len)
{
+ const U8* const send = s + (len ? len : strlen((const char *)s));
const U8* x = s;
- const U8* send;
- if (!len)
- len = strlen((const char *)s);
- send = s + len;
+ PERL_UNUSED_CONTEXT;
while (x < send) {
STRLEN c;
c = UTF8SKIP(x);
if (IS_UTF8_CHAR_FAST(c)) {
if (!IS_UTF8_CHAR(x, c))
- goto out;
- } else if (!is_utf8_char_slow(x, c))
- goto out;
+ c = 0;
+ }
+ else
+ c = is_utf8_char_slow(x, c);
#else
c = is_utf8_char(x);
#endif /* #ifdef IS_UTF8_CHAR */
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;
- const U8* send;
STRLEN c;
-
- if (!len)
- len = strlen((const char *)s);
- send = s + len;
- if (el)
- *el = 0;
+ STRLEN outlen = 0;
+ PERL_UNUSED_CONTEXT;
while (x < send) {
/* Inline the easy bits of is_utf8_char() here for speed... */
goto out;
}
x += c;
- if (el)
- (*el)++;
+ outlen++;
}
out:
+ if (el)
+ *el = outlen;
+
if (ep)
*ep = x;
- if (x != send)
- return FALSE;
-
- return TRUE;
+ 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>
+Returns the Unicode code point value of the first character in the string C<s>
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.
UV
Perl_utf8n_to_uvuni(pTHX_ const U8 *s, STRLEN curlen, STRLEN *retlen, U32 flags)
{
- const U8 *s0 = s;
+ dVAR;
+ const U8 * const s0 = s;
UV uv = *s, ouv = 0;
STRLEN len = 1;
const bool dowarn = ckWARN_d(WARN_UTF8);
if (flags & UTF8_CHECK_ONLY) {
if (retlen)
- *retlen = -1;
+ *retlen = ((STRLEN) -1);
return 0;
}
if (dowarn) {
- SV* const 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_catpv(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);
Perl_sv_catpvf(aTHX_ sv, "(character 0x%04"UVxf")", uv);
break;
default:
- Perl_sv_catpv(aTHX_ sv, "(unknown reason)");
+ sv_catpvs(sv, "(unknown reason)");
break;
}
STRLEN
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.
if (e < s)
goto warn_and_return;
while (s < e) {
- const U8 t = UTF8SKIP(s);
+ t = UTF8SKIP(s);
if (e - s < t) {
warn_and_return:
if (ckWARN_d(WARN_UTF8)) {
IV
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) {
- const U8 c = UTF8SKIP(a);
- if (b - a < c)
- goto warn_and_return;
- a += c;
- off--;
- }
- }
- else {
- while (b < a) {
- const U8 c = UTF8SKIP(b);
-
- if (a - b < c) {
- 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 off;
- }
- b += c;
- off++;
- }
- }
-
- return off;
+ return (a < b) ? -1 * (IV) utf8_length(a, b) : (IV) utf8_length(b, a);
}
/*
U8 *
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. */
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 UTF-8 and chars < 256 before updating string */
- for (send = s + *len; s < send; ) {
+ 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;
}
}
const U8 *send;
I32 count = 0;
+ PERL_UNUSED_CONTEXT;
if (!*is_utf8)
return (U8 *)start;
}
}
- *is_utf8 = 0;
+ *is_utf8 = FALSE;
- Newxz(d, (*len) - count + 1, U8);
+ Newx(d, (*len) - count + 1, U8);
s = start; start = d;
while (s < send) {
U8 c = *s++;
const U8 * const send = s + (*len);
U8 *d;
U8 *dst;
+ PERL_UNUSED_CONTEXT;
- Newxz(d, (*len) * 2 + 1, U8);
+ Newx(d, (*len) * 2 + 1, U8);
dst = d;
while (s < send) {
}
if (bytelen & 1)
- Perl_croak(aTHX_ "panic: utf16_to_utf8: odd bytelen %"UVf, (UV)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) {
+#ifdef EBCDIC
+ *d++ = UNI_TO_NATIVE(uv);
+#else
*d++ = (U8)uv;
+#endif
continue;
}
if (uv < 0x800) {
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;
S_is_utf8_common(pTHX_ const U8 *const p, SV **swash,
const char *const swashname)
{
+ dVAR;
if (!is_utf8_char(p))
return FALSE;
if (!*swash)
bool
Perl_is_utf8_alnum(pTHX_ const U8 *p)
{
+ dVAR;
/* NOTE: "IsWord", not "IsAlnum", since Alnum is a true
* descendant of isalnum(3), in other words, it doesn't
* contain the '_'. --jhi */
- return S_is_utf8_common(aTHX_ p, &PL_utf8_alnum, "IsWord");
+ return is_utf8_common(p, &PL_utf8_alnum, "IsWord");
}
bool
Perl_is_utf8_alnumc(pTHX_ const U8 *p)
{
- return S_is_utf8_common(aTHX_ p, &PL_utf8_alnumc, "IsAlnumC");
+ dVAR;
+ return is_utf8_common(p, &PL_utf8_alnumc, "IsAlnumC");
}
bool
Perl_is_utf8_idfirst(pTHX_ const U8 *p) /* The naming is historical. */
{
+ dVAR;
if (*p == '_')
return TRUE;
/* is_utf8_idstart would be more logical. */
- return S_is_utf8_common(aTHX_ p, &PL_utf8_idstart, "IdStart");
+ return is_utf8_common(p, &PL_utf8_idstart, "IdStart");
}
bool
Perl_is_utf8_idcont(pTHX_ const U8 *p)
{
+ dVAR;
if (*p == '_')
return TRUE;
- return S_is_utf8_common(aTHX_ p, &PL_utf8_idcont, "IdContinue");
+ return is_utf8_common(p, &PL_utf8_idcont, "IdContinue");
}
bool
Perl_is_utf8_alpha(pTHX_ const U8 *p)
{
- return S_is_utf8_common(aTHX_ p, &PL_utf8_alpha, "IsAlpha");
+ dVAR;
+ return is_utf8_common(p, &PL_utf8_alpha, "IsAlpha");
}
bool
Perl_is_utf8_ascii(pTHX_ const U8 *p)
{
- return S_is_utf8_common(aTHX_ p, &PL_utf8_ascii, "IsAscii");
+ dVAR;
+ return is_utf8_common(p, &PL_utf8_ascii, "IsAscii");
}
bool
Perl_is_utf8_space(pTHX_ const U8 *p)
{
- return S_is_utf8_common(aTHX_ p, &PL_utf8_space, "IsSpacePerl");
+ dVAR;
+ return is_utf8_common(p, &PL_utf8_space, "IsSpacePerl");
}
bool
Perl_is_utf8_digit(pTHX_ const U8 *p)
{
- return S_is_utf8_common(aTHX_ p, &PL_utf8_digit, "IsDigit");
+ dVAR;
+ return is_utf8_common(p, &PL_utf8_digit, "IsDigit");
}
bool
Perl_is_utf8_upper(pTHX_ const U8 *p)
{
- return S_is_utf8_common(aTHX_ p, &PL_utf8_upper, "IsUppercase");
+ dVAR;
+ return is_utf8_common(p, &PL_utf8_upper, "IsUppercase");
}
bool
Perl_is_utf8_lower(pTHX_ const U8 *p)
{
- return S_is_utf8_common(aTHX_ p, &PL_utf8_lower, "IsLowercase");
+ dVAR;
+ return is_utf8_common(p, &PL_utf8_lower, "IsLowercase");
}
bool
Perl_is_utf8_cntrl(pTHX_ const U8 *p)
{
- return S_is_utf8_common(aTHX_ p, &PL_utf8_cntrl, "IsCntrl");
+ dVAR;
+ return is_utf8_common(p, &PL_utf8_cntrl, "IsCntrl");
}
bool
Perl_is_utf8_graph(pTHX_ const U8 *p)
{
- return S_is_utf8_common(aTHX_ p, &PL_utf8_graph, "IsGraph");
+ dVAR;
+ return is_utf8_common(p, &PL_utf8_graph, "IsGraph");
}
bool
Perl_is_utf8_print(pTHX_ const U8 *p)
{
- return S_is_utf8_common(aTHX_ p, &PL_utf8_print, "IsPrint");
+ dVAR;
+ return is_utf8_common(p, &PL_utf8_print, "IsPrint");
}
bool
Perl_is_utf8_punct(pTHX_ const U8 *p)
{
- return S_is_utf8_common(aTHX_ p, &PL_utf8_punct, "IsPunct");
+ dVAR;
+ return is_utf8_common(p, &PL_utf8_punct, "IsPunct");
}
bool
Perl_is_utf8_xdigit(pTHX_ const U8 *p)
{
- return S_is_utf8_common(aTHX_ p, &PL_utf8_xdigit, "Isxdigit");
+ dVAR;
+ return is_utf8_common(p, &PL_utf8_xdigit, "Isxdigit");
}
bool
Perl_is_utf8_mark(pTHX_ const U8 *p)
{
- return S_is_utf8_common(aTHX_ p, &PL_utf8_mark, "IsM");
+ dVAR;
+ return is_utf8_common(p, &PL_utf8_mark, "IsM");
}
/*
Perl_to_utf8_case(pTHX_ const U8 *p, U8* ustrp, STRLEN *lenp,
SV **swashp, const char *normal, const char *special)
{
+ dVAR;
U8 tmpbuf[UTF8_MAXBYTES_CASE+1];
STRLEN len = 0;
if (special && (uv1 == 0xDF || uv1 > 0xFF)) {
/* It might be "special" (sometimes, but not always,
* a multicharacter mapping) */
- HV *hv;
+ HV * const hv = get_hv(special, FALSE);
SV **svp;
- if ((hv = get_hv(special, FALSE)) &&
+ if (hv &&
(svp = hv_fetch(hv, (const char*)tmpbuf, UNISKIP(uv1), FALSE)) &&
(*svp)) {
const char *s;
STRLEN tlen = 0;
while (t < tend) {
- UV c = utf8_to_uvchr(t, &tlen);
+ const UV c = utf8_to_uvchr(t, &tlen);
if (tlen > 0) {
d = uvchr_to_utf8(d, UNI_TO_NATIVE(c));
t += tlen;
}
if (!len && *swashp) {
- UV uv2 = swash_fetch(*swashp, tmpbuf, TRUE);
-
+ const UV uv2 = swash_fetch(*swashp, tmpbuf, TRUE);
+
if (uv2) {
/* It was "normal" (a single character mapping). */
- UV uv3 = UNI_TO_NATIVE(uv2);
-
+ const UV uv3 = UNI_TO_NATIVE(uv2);
len = uvchr_to_utf8(ustrp, uv3) - ustrp;
}
}
UV
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");
}
UV
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");
}
UV
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");
}
UV
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");
}
{
dVAR;
SV* retval;
- SV* const tokenbufsv = sv_newmortal();
dSP;
const size_t pkg_len = strlen(pkg);
const size_t name_len = strlen(name);
- HV * const stash = gv_stashpvn(pkg, pkg_len, FALSE);
+ HV * const stash = gv_stashpvn(pkg, pkg_len, 0);
SV* errsv_save;
PUSHSTACKi(PERLSI_MAGIC);
if (!gv_fetchmeth(stash, "SWASHNEW", 8, -1)) { /* demand load utf8 */
ENTER;
errsv_save = newSVsv(ERRSV);
+ /* 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),
- Nullsv);
+ NULL);
if (!SvTRUE(ERRSV))
sv_setsv(ERRSV, errsv_save);
SvREFCNT_dec(errsv_save);
PUSHs(sv_2mortal(newSViv(minbits)));
PUSHs(sv_2mortal(newSViv(none)));
PUTBACK;
- 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--);
LEAVE;
POPSTACK;
if (IN_PERL_COMPILETIME) {
- STRLEN len;
- const char* const pv = SvPV_const(tokenbufsv, len);
-
- Copy(pv, PL_tokenbuf, len+1, char);
- PL_curcop->op_private = (U8)(PL_hints & HINT_PRIVATE_MASK);
+ CopHINTS_set(PL_curcop, PL_hints);
}
if (!SvROK(retval) || SvTYPE(SvRV(retval)) != SVt_PVHV) {
if (SvPOK(retval))
Perl_croak(aTHX_ "Can't find Unicode property definition \"%"SVf"\"",
- retval);
+ SVfARG(retval));
Perl_croak(aTHX_ "SWASHNEW didn't return an HV ref");
}
return retval;
U32 bit;
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] = (U8)UTF8_EIGHT_BIT_HI(c);
needents);
if (IN_PERL_COMPILETIME)
- PL_curcop->op_private = (U8)(PL_hints & HINT_PRIVATE_MASK);
+ CopHINTS_set(PL_curcop, PL_hints);
svp = hv_store(hv, (const char *)ptr, klen, swatch, 0);
}
PL_last_swash_hv = hv;
- PL_last_swash_klen = klen;
+ assert(klen <= sizeof(PL_last_swash_key));
+ PL_last_swash_klen = (U8)klen;
/* FIXME change interpvar.h? */
PL_last_swash_tmps = (U8 *) tmps;
PL_last_swash_slen = slen;
return (tmps[off] << 24) + (tmps[off+1] << 16) + (tmps[off+2] << 8) + tmps[off + 3] ;
}
Perl_croak(aTHX_ "panic: swash_fetch got swatch of unexpected bit width");
- return 0;
+ NORETURN_FUNCTION_END;
}
/* Note:
STRLEN lcur, xcur, scur;
HV* const hv = (HV*)SvRV(swash);
- SV** const listsvp = hv_fetch(hv, "LIST", 4, FALSE);
- SV** const typesvp = hv_fetch(hv, "TYPE", 4, FALSE);
- SV** const bitssvp = hv_fetch(hv, "BITS", 4, FALSE);
- SV** const nonesvp = hv_fetch(hv, "NONE", 4, FALSE);
- SV** const extssvp = hv_fetch(hv, "EXTRAS", 6, FALSE);
+ 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);
}
/* create and initialize $swatch */
- swatch = newSVpvn("",0);
+ swatch = newSVpvs("");
scur = octets ? (span * octets) : (span + 7) / 8;
SvGROW(swatch, scur + 1);
s = (U8*)SvPVX(swatch);
l = (U8*)SvPV(*listsvp, lcur);
lend = l + lcur;
while (l < lend) {
- UV min, max, val, key;
+ UV min, max, val;
STRLEN numlen;
I32 flags = PERL_SCAN_SILENT_ILLDIGIT | PERL_SCAN_DISALLOW_PREFIX;
continue;
if (octets) {
+ UV key;
if (min < start) {
if (!none || val < none) {
val += start - min;
}
}
else { /* bits == 1, then val should be ignored */
+ UV key;
if (min < start)
min = start;
for (key = min; key <= max; key++) {
U8 *s, *o, *nl;
STRLEN slen, olen;
- U8 opc = *x++;
+ const U8 opc = *x++;
if (opc == '\n')
continue;
othersvp = hv_fetch(hv, (char *)namestr, namelen, FALSE);
otherhv = (HV*)SvRV(*othersvp);
- otherbitssvp = hv_fetch(otherhv, "BITS", 4, FALSE);
+ otherbitssvp = hv_fetchs(otherhv, "BITS", FALSE);
otherbits = (STRLEN)SvUV(*otherbitssvp);
if (bits < otherbits)
Perl_croak(aTHX_ "panic: swash_get found swatch size mismatch");
else {
STRLEN otheroctets = otherbits >> 3;
STRLEN offset = 0;
- U8* send = s + slen;
+ U8* const send = s + slen;
while (s < send) {
UV otherval = 0;
}
if (opc == '+' && otherval)
- ; /* replace with otherval */
+ NOOP; /* replace with otherval */
else if (opc == '!' && !otherval)
otherval = 1;
else if (opc == '-' && otherval)
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)
{
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)
{
+ dVAR;
register const U8 *p1 = (const U8*)s1;
register const U8 *p2 = (const U8*)s2;
register const U8 *f1 = NULL;
- register U8 *f2 = NULL;
+ register const U8 *f2 = NULL;
register U8 *e1 = NULL;
register U8 *q1 = NULL;
register U8 *e2 = NULL;
https://perl5.git.perl.org / perl5.git / blobdiff