+ /* Longjmp back to here if have to switch in midstream to utf8 */
+ if (! RExC_orig_utf8) {
+ JMPENV_PUSH(jump_ret);
+ }
+
+ if (jump_ret == 0) { /* First time through */
+ exp = SvPV(pattern, plen);
+ xend = exp + plen;
+
+ DEBUG_COMPILE_r({
+ SV *dsv= sv_newmortal();
+ RE_PV_QUOTED_DECL(s, RExC_utf8,
+ dsv, exp, plen, 60);
+ PerlIO_printf(Perl_debug_log, "%sCompiling REx%s %s\n",
+ PL_colors[4],PL_colors[5],s);
+ });
+ }
+ else { /* longjumped back */
+ STRLEN len = plen;
+
+ /* If the cause for the longjmp was other than changing to utf8, pop
+ * our own setjmp, and longjmp to the correct handler */
+ if (jump_ret != UTF8_LONGJMP) {
+ JMPENV_POP;
+ JMPENV_JUMP(jump_ret);
+ }
+
+ GET_RE_DEBUG_FLAGS;
+
+ /* It's possible to write a regexp in ascii that represents Unicode
+ codepoints outside of the byte range, such as via \x{100}. If we
+ detect such a sequence we have to convert the entire pattern to utf8
+ and then recompile, as our sizing calculation will have been based
+ on 1 byte == 1 character, but we will need to use utf8 to encode
+ at least some part of the pattern, and therefore must convert the whole
+ thing.
+ -- dmq */
+ DEBUG_PARSE_r(PerlIO_printf(Perl_debug_log,
+ "UTF8 mismatch! Converting to utf8 for resizing and compile\n"));
+ exp = (char*)Perl_bytes_to_utf8(aTHX_ (U8*)SvPV(pattern, plen), &len);
+ xend = exp + len;
+ RExC_orig_utf8 = RExC_utf8 = 1;
+ SAVEFREEPV(exp);
+ }
+
+#ifdef TRIE_STUDY_OPT
+ restudied = 0;
+#endif
+
https://perl5.git.perl.org / perl5.git / blobdiff