+
+
+/*
+=for apidoc A|U8 *|uvchr_to_utf8|U8 *d|UV uv
+
+Adds the UTF8 representation of the Native codepoint C<uv> to the end
+of the string C<d>; C<d> should be have at least C<UTF8_MAXLEN+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
+*/
+#undef Perl_uvchr_to_utf8
+U8 *
+Perl_uvchr_to_utf8(pTHX_ U8 *d, UV uv)
+{
+ return Perl_uvuni_to_utf8(aTHX_ d, NATIVE_TO_UNI(uv));
+}
+
+
+/*
+=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 UTF8 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
+*/
+#undef Perl_utf8n_to_uvchr
+UV
+Perl_utf8n_to_uvchr(pTHX_ U8 *s, STRLEN curlen, STRLEN *retlen, U32 flags)
+{
+ UV uv = Perl_utf8n_to_uvuni(aTHX_ s, curlen, retlen, flags);
+ return UNI_TO_NATIVE(uv);
+}
+
+char *
+Perl_pv_uni_display(pTHX_ SV *dsv, U8 *spv, STRLEN len, STRLEN pvlim, UV flags)
+{
+ int truncated = 0;
+ char *s, *e;
+
+ sv_setpvn(dsv, "", 0);
+ for (s = (char *)spv, e = s + len; s < e; s += UTF8SKIP(s)) {
+ UV u;
+ if (pvlim && SvCUR(dsv) >= pvlim) {
+ truncated++;
+ break;
+ }
+ u = utf8_to_uvchr((U8*)s, 0);
+ Perl_sv_catpvf(aTHX_ dsv, "\\x{%"UVxf"}", u);
+ }
+ if (truncated)
+ sv_catpvn(dsv, "...", 3);
+
+ return SvPVX(dsv);
+}
+
+char *
+Perl_sv_uni_display(pTHX_ SV *dsv, SV *ssv, STRLEN pvlim, UV flags)
+{
+ return Perl_pv_uni_display(aTHX_ dsv, (U8*)SvPVX(ssv), SvCUR(ssv),
+ pvlim, flags);
+}
+
+I32
+Perl_ibcmp_utf8(pTHX_ const char *s1, bool u1, const char *s2, bool u2, register I32 len)
+{
+ register U8 *a = (U8*)s1;
+ register U8 *b = (U8*)s2;
+ STRLEN la, lb;
+ UV ca, cb;
+ STRLEN ulen1, ulen2;
+ U8 tmpbuf1[UTF8_MAXLEN*3+1];
+ U8 tmpbuf2[UTF8_MAXLEN*3+1];
+
+ while (len) {
+ if (u1)
+ ca = utf8_to_uvchr((U8*)a, &la);
+ else {
+ ca = *a;
+ la = 1;
+ }
+ if (u2)
+ cb = utf8_to_uvchr((U8*)b, &lb);
+ else {
+ cb = *b;
+ lb = 1;
+ }
+ if (ca != cb) {
+ if (u1)
+ to_uni_fold(NATIVE_TO_UNI(ca), tmpbuf1, &ulen1);
+ else
+ ulen1 = 1;
+ if (u2)
+ to_uni_fold(NATIVE_TO_UNI(cb), tmpbuf2, &ulen2);
+ else
+ ulen2 = 1;
+ if (ulen1 != ulen2
+ || (ulen1 == 1 && PL_fold[ca] != PL_fold[cb])
+ || memNE((char *)tmpbuf1, (char *)tmpbuf2, ulen1))
+ return 1;
+ }
+ a += la;
+ b += lb;
+ }
+ return 0;
+}
+