blead 25801: Symbian batch of today

[p5sagit/p5-mst-13.2.git] / utf8.c

diff --git a/utf8.c b/utf8.c
index 599edf1..743dea4 100644 (file)
--- a/utf8.c
+++ b/utf8.c
@@ -209,6 +209,9 @@ S_is_utf8_char_slow(pTHX_ const U8 *s, const STRLEN len)
 
     slen = len - 1;
     s++;
+#ifdef EBCDIC
+    u = NATIVE_TO_UTF(u);
+#endif
     u &= UTF_START_MASK(len);
     uv  = u;
     ouv = uv;
@@ -240,7 +243,7 @@ will be returned if it is valid, otherwise 0.
 STRLEN
 Perl_is_utf8_char(pTHX_ const U8 *s)
 {
-    STRLEN len = UTF8SKIP(s);
+    const STRLEN len = UTF8SKIP(s);
 #ifdef IS_UTF8_CHAR
     if (IS_UTF8_CHAR_FAST(len))
         return IS_UTF8_CHAR(s, len) ? len : 0;
@@ -267,7 +270,7 @@ Perl_is_utf8_string(pTHX_ const U8 *s, STRLEN len)
     const U8* x = s;
     const U8* send;
 
-    if (!len && s)
+    if (!len)
        len = strlen((const char *)s);
     send = s + len;
 
@@ -323,7 +326,7 @@ Perl_is_utf8_string_loclen(pTHX_ const U8 *s, STRLEN len, const U8 **ep, STRLEN
     const U8* send;
     STRLEN c;
 
-    if (!len && s)
+    if (!len)
         len = strlen((const char *)s);
     send = s + len;
     if (el)
@@ -543,7 +546,7 @@ malformed:
     }
 
     if (dowarn) {
-       SV* sv = sv_2mortal(newSVpv("Malformed UTF-8 character ", 0));
+       SV* const sv = sv_2mortal(newSVpv("Malformed UTF-8 character ", 0));
 
        switch (warning) {
        case 0: /* Intentionally empty. */ break;
@@ -557,10 +560,12 @@ malformed:
            if (s == s0)
                Perl_sv_catpvf(aTHX_ sv, "(unexpected non-continuation byte 0x%02"UVxf", immediately after start byte 0x%02"UVxf")",
                            (UV)s[1], startbyte);
-           else
+           else {
+               const int len = (int)(s-s0);
                Perl_sv_catpvf(aTHX_ sv, "(unexpected non-continuation byte 0x%02"UVxf", %d byte%s after start byte 0x%02"UVxf", expected %d bytes)",
-                           (UV)s[1], s - s0, s - s0 > 1 ? "s" : "", startbyte, (int)expectlen);
-             
+                           (UV)s[1], len, len > 1 ? "s" : "", startbyte, (int)expectlen);
+           }
+
            break;
        case UTF8_WARN_FE_FF:
            Perl_sv_catpvf(aTHX_ sv, "(byte 0x%02"UVxf")", uv);
@@ -590,7 +595,7 @@ malformed:
        }
        
        if (warning) {
-           const char *s = SvPVX_const(sv);
+           const char * const s = SvPVX_const(sv);
 
            if (PL_op)
                Perl_warner(aTHX_ packWARN(WARN_UTF8),
@@ -1451,12 +1456,13 @@ The "normal" is a string like "ToLower" which means the swash
 =cut */
 
 UV
-Perl_to_utf8_case(pTHX_ const U8 *p, U8* ustrp, STRLEN *lenp, SV **swashp, const char *normal, const char *special)
+Perl_to_utf8_case(pTHX_ const U8 *p, U8* ustrp, STRLEN *lenp,
+                       SV **swashp, const char *normal, const char *special)
 {
     U8 tmpbuf[UTF8_MAXBYTES_CASE+1];
     STRLEN len = 0;
 
-    const UV uv0 = utf8_to_uvchr(p, 0);
+    const UV uv0 = utf8_to_uvchr(p, NULL);
     /* 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. */
@@ -1625,11 +1631,11 @@ Perl_swash_init(pTHX_ const char* pkg, const char* name, SV *listsv, I32 minbits
 {
     dVAR;
     SV* retval;
-    SV* tokenbufsv = sv_newmortal();
+    SV* const tokenbufsv = sv_newmortal();
     dSP;
     const size_t pkg_len = strlen(pkg);
     const size_t name_len = strlen(name);
-    HV *stash = gv_stashpvn(pkg, pkg_len, FALSE);
+    HV * const stash = gv_stashpvn(pkg, pkg_len, FALSE);
     SV* errsv_save;
 
     PUSHSTACKi(PERLSI_MAGIC);
@@ -1674,7 +1680,7 @@ Perl_swash_init(pTHX_ const char* pkg, const char* name, SV *listsv, I32 minbits
     POPSTACK;
     if (IN_PERL_COMPILETIME) {
        STRLEN len;
-        const char* pv = SvPV_const(tokenbufsv, 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);
@@ -1698,7 +1704,7 @@ UV
 Perl_swash_fetch(pTHX_ SV *sv, const U8 *ptr, bool do_utf8)
 {
     dVAR;
-    HV* hv = (HV*)SvRV(sv);
+    HV* const hv = (HV*)SvRV(sv);
     U32 klen;
     U32 off;
     STRLEN slen;
@@ -1763,7 +1769,7 @@ Perl_swash_fetch(pTHX_ SV *sv, const U8 *ptr, bool do_utf8)
            /* 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_MAXBYTES, 0,
+           const UV code_point = utf8n_to_uvuni(ptr, UTF8_MAXBYTES, 0,
                                           ckWARN(WARN_UTF8) ?
                                           0 : UTF8_ALLOW_ANY);
            SV *errsv_save;
@@ -1878,7 +1884,7 @@ Allows length and flags to be passed to low level routine.
 UV
 Perl_utf8n_to_uvchr(pTHX_ const U8 *s, STRLEN curlen, STRLEN *retlen, U32 flags)
 {
-    UV uv = Perl_utf8n_to_uvuni(aTHX_ s, curlen, retlen, flags);
+    const UV uv = Perl_utf8n_to_uvuni(aTHX_ s, curlen, retlen, flags);
     return UNI_TO_NATIVE(uv);
 }