#endif
#endif /* Loop style */
}
-
-U8 *
-Perl_uvuni_to_utf8(pTHX_ U8 *d, UV uv)
-{
- return Perl_uvuni_to_utf8_flags(aTHX_ d, uv, 0);
-}
/*
UV
Perl_utf8_to_uvchr(pTHX_ const U8 *s, STRLEN *retlen)
{
- return Perl_utf8n_to_uvchr(aTHX_ s, UTF8_MAXBYTES, retlen,
- ckWARN(WARN_UTF8) ? 0 : UTF8_ALLOW_ANY);
+ return utf8n_to_uvchr(s, UTF8_MAXBYTES, retlen,
+ ckWARN(WARN_UTF8) ? 0 : UTF8_ALLOW_ANY);
}
/*
return (U32)to_uni_lower(c, tmpbuf, &len);
}
-bool
-Perl_is_utf8_alnum(pTHX_ const U8 *p)
+static bool
+S_is_utf8_common(pTHX_ const U8 *const p, SV **swash,
+ const char *const swashname)
{
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) != 0;
-/* 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) != 0;
-#endif
+ if (!*swash)
+ *swash = swash_init("utf8", swashname, &PL_sv_undef, 1, 0);
+ return swash_fetch(*swash, p, TRUE) != 0;
+}
+
+bool
+Perl_is_utf8_alnum(pTHX_ const U8 *p)
+{
+ /* 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");
}
bool
Perl_is_utf8_alnumc(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) != 0;
-/* 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) != 0;
-#endif
+ return S_is_utf8_common(aTHX_ p, &PL_utf8_alnumc, "IsAlnumC");
}
bool
{
if (*p == '_')
return TRUE;
- if (!is_utf8_char(p))
- return FALSE;
- if (!PL_utf8_idstart) /* is_utf8_idstart would be more logical. */
- PL_utf8_idstart = swash_init("utf8", "IdStart", &PL_sv_undef, 0, 0);
- return swash_fetch(PL_utf8_idstart, p, TRUE) != 0;
+ /* is_utf8_idstart would be more logical. */
+ return S_is_utf8_common(aTHX_ p, &PL_utf8_idstart, "IdStart");
}
bool
{
if (*p == '_')
return TRUE;
- if (!is_utf8_char(p))
- return FALSE;
- if (!PL_utf8_idcont)
- PL_utf8_idcont = swash_init("utf8", "IdContinue", &PL_sv_undef, 0, 0);
- return swash_fetch(PL_utf8_idcont, p, TRUE) != 0;
+ return S_is_utf8_common(aTHX_ p, &PL_utf8_idcont, "IdContinue");
}
bool
Perl_is_utf8_alpha(pTHX_ const U8 *p)
{
- 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) != 0;
+ return S_is_utf8_common(aTHX_ p, &PL_utf8_alpha, "IsAlpha");
}
bool
Perl_is_utf8_ascii(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) != 0;
+ return S_is_utf8_common(aTHX_ p, &PL_utf8_ascii, "IsAscii");
}
bool
Perl_is_utf8_space(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) != 0;
+ return S_is_utf8_common(aTHX_ p, &PL_utf8_space, "IsSpacePerl");
}
bool
Perl_is_utf8_digit(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) != 0;
+ return S_is_utf8_common(aTHX_ p, &PL_utf8_digit, "IsDigit");
}
bool
Perl_is_utf8_upper(pTHX_ const U8 *p)
{
- if (!is_utf8_char(p))
- return FALSE;
- if (!PL_utf8_upper)
- PL_utf8_upper = swash_init("utf8", "IsUppercase", &PL_sv_undef, 0, 0);
- return swash_fetch(PL_utf8_upper, p, TRUE) != 0;
+ return S_is_utf8_common(aTHX_ p, &PL_utf8_upper, "IsUppercase");
}
bool
Perl_is_utf8_lower(pTHX_ const U8 *p)
{
- if (!is_utf8_char(p))
- return FALSE;
- if (!PL_utf8_lower)
- PL_utf8_lower = swash_init("utf8", "IsLowercase", &PL_sv_undef, 0, 0);
- return swash_fetch(PL_utf8_lower, p, TRUE) != 0;
+ return S_is_utf8_common(aTHX_ p, &PL_utf8_lower, "IsLowercase");
}
bool
Perl_is_utf8_cntrl(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) != 0;
+ return S_is_utf8_common(aTHX_ p, &PL_utf8_cntrl, "IsCntrl");
}
bool
Perl_is_utf8_graph(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) != 0;
+ return S_is_utf8_common(aTHX_ p, &PL_utf8_graph, "IsGraph");
}
bool
Perl_is_utf8_print(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) != 0;
+ return S_is_utf8_common(aTHX_ p, &PL_utf8_print, "IsPrint");
}
bool
Perl_is_utf8_punct(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) != 0;
+ return S_is_utf8_common(aTHX_ p, &PL_utf8_punct, "IsPunct");
}
bool
Perl_is_utf8_xdigit(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) != 0;
+ return S_is_utf8_common(aTHX_ p, &PL_utf8_xdigit, "Isxdigit");
}
bool
Perl_is_utf8_mark(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) != 0;
+ return S_is_utf8_common(aTHX_ p, &PL_utf8_mark, "IsM");
}
/*
The "swashp" is a pointer to the swash to use.
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,
+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
&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_ const char* pkg, const char* name, SV *listsv, I32 minbits, I32 none)
{
* (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. */
UV
Perl_swash_fetch(pTHX_ SV *sv, const U8 *ptr, bool do_utf8)
{
STRLEN needents;
const U8 *tmps = NULL;
U32 bit;
- SV *retval;
+ SV *swatch;
U8 tmputf8[2];
UV c = NATIVE_TO_ASCII(*ptr);
if (!do_utf8 && !UNI_IS_INVARIANT(c)) {
- tmputf8[0] = (U8)UTF8_EIGHT_BIT_HI(c);
- tmputf8[1] = (U8)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 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
/* Try our second-level swatch cache, kept in a hash. */
SV** svp = hv_fetch(hv, (const char*)ptr, klen, FALSE);
- /* If not cached, generate it via utf8::SWASHGET */
- if (!svp || !SvPOK(*svp) || !(tmps = (const U8*)SvPV_const(*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.
*/
const UV code_point = utf8n_to_uvuni(ptr, UTF8_MAXBYTES, 0,
ckWARN(WARN_UTF8) ?
0 : UTF8_ALLOW_ANY);
- 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;
+ swatch = swash_get(sv,
+ /* On EBCDIC & ~(0xA0-1) isn't a useful thing to do */
+ (klen) ? (code_point & ~(needents - 1)) : 0,
+ needents);
+
if (IN_PERL_COMPILETIME)
PL_curcop->op_private = (U8)(PL_hints & HINT_PRIVATE_MASK);
- svp = hv_store(hv, (const char *)ptr, klen, retval, 0);
+ svp = hv_store(hv, (const char *)ptr, klen, swatch, 0);
- if (!svp || !(tmps = (U8*)SvPV(*svp, slen)) || (slen << 3) < needents)
- Perl_croak(aTHX_ "SWASHGET didn't return result of proper length");
+ if (!svp || !(tmps = (U8*)SvPV(*svp, slen))
+ || (slen << 3) < needents)
+ Perl_croak(aTHX_ "The swatch does not have proper length");
}
PL_last_swash_hv = hv;
return 0;
}
+/* 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** listsvp = hv_fetch(hv, "LIST", 4, FALSE);
+ SV** typesvp = hv_fetch(hv, "TYPE", 4, FALSE);
+ SV** bitssvp = hv_fetch(hv, "BITS", 4, FALSE);
+ SV** nonesvp = hv_fetch(hv, "NONE", 4, FALSE);
+ SV** extssvp = hv_fetch(hv, "EXTRAS", 6, FALSE);
+ U8* typestr = (U8*)SvPV_nolen(*typesvp);
+ int typeto = typestr[0] == 'T' && typestr[1] == 'o';
+ STRLEN bits = SvUV(*bitssvp);
+ STRLEN octets = bits >> 3; /* if bits == 1, then octets == 0 */
+ UV none = SvUV(*nonesvp);
+ UV end = start + span;
+
+ if (bits != 1 && bits != 8 && bits != 16 && bits != 32) {
+ Perl_croak(aTHX_ "swash_get: unknown bits %"UVuf, (UV) bits);
+ }
+
+ /* create and initialize $swatch */
+ swatch = newSVpvn("",0);
+ scur = octets ? (span * octets) : (span + 7) / 8;
+ SvGROW(swatch, scur + 1);
+ s = (U8*)SvPVX(swatch);
+ if (octets && none) {
+ const U8* 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, key;
+ STRLEN numlen;
+ I32 flags = PERL_SCAN_SILENT_ILLDIGIT | PERL_SCAN_DISALLOW_PREFIX;
+
+ U8* 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) {
+ 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 */
+ if (min < start)
+ min = start;
+ for (key = min; key <= max; key++) {
+ 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;
+
+ 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);
+ if (*othersvp && SvROK(*othersvp) &&
+ SvTYPE(SvRV(*othersvp))==SVt_PVHV)
+ otherhv = (HV*)SvRV(*othersvp);
+ else
+ Perl_croak(aTHX_ "otherhv is not a hash reference");
+
+ otherbitssvp = hv_fetch(otherhv, "BITS", 4, FALSE);
+ otherbits = (STRLEN)SvUV(*otherbitssvp);
+ if (bits < otherbits)
+ Perl_croak(aTHX_ "swash_get: 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_ "swash_get didn't return valid swatch for other");
+
+ s = (U8*)SvPV(swatch, slen);
+ if (bits == 1 && otherbits == 1) {
+ if (slen != olen)
+ Perl_croak(aTHX_ "swash_get: 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* 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)
+ ; /* 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 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,
+
+ 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
+*/
+U8 *
+Perl_uvchr_to_utf8(pTHX_ U8 *d, UV uv)
+{
+ return Perl_uvuni_to_utf8_flags(aTHX_ d, NATIVE_TO_UNI(uv), 0);
+}
+
U8 *
Perl_uvchr_to_utf8_flags(pTHX_ U8 *d, UV uv, UV 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 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.
+
+=cut
+*/
+/* On ASCII machines this is normally a macro but we want
+ a real function in case XS code wants it
+*/
+UV
+Perl_utf8n_to_uvchr(pTHX_ const U8 *s, STRLEN curlen, STRLEN *retlen,
+U32 flags)
+{
+ const 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,
projects / p5sagit/p5-mst-13.2.git / blobdiff