diff options
| author | Karl Williamson <public@khwilliamson.com> | 2011-10-16 12:47:21 -0600 |
|---|---|---|
| committer | Karl Williamson <public@khwilliamson.com> | 2011-10-17 21:52:17 -0600 |
| commit | 77a6d8568e288ad300ad7f0805946559b4ec28d1 (patch) | |
| tree | 7b6281d683ce12dfd390312b6de067c0643d4144 | |
| parent | 227968da0f4293e121ecb5c36b17ddc98da70dc1 (diff) | |
| download | perl-77a6d8568e288ad300ad7f0805946559b4ec28d1.tar.gz | |
regexec.c: Less work in /i matching
If you watch an execution trace of regexec /i, often you will see it
folding the same thing over and over, as it backtracks or searches
ahead. regcomp.c has now been changed to always fold UTF-8 encoded
EXACTF and EXCACTFU nodes. This allows these to not be re-folded each
time.
This commit does it just for find_by_class(). Other commits will expand
this technique for other cases.
| -rw-r--r-- | regexec.c | 6 |
1 files changed, 4 insertions, 2 deletions
@@ -1465,7 +1465,9 @@ S_find_byclass(pTHX_ regexp * prog, const regnode *c, char *s, case EXACTF: if (UTF_PATTERN || utf8_target) { - utf8_fold_flags = 0; + + /* regcomp.c already folded this if pattern is in UTF-8 */ + utf8_fold_flags = (UTF_PATTERN) ? FOLDEQ_S2_ALREADY_FOLDED : 0; goto do_exactf_utf8; } fold_array = PL_fold; @@ -1483,7 +1485,7 @@ S_find_byclass(pTHX_ regexp * prog, const regnode *c, char *s, case EXACTFU: if (UTF_PATTERN || utf8_target) { - utf8_fold_flags = 0; + utf8_fold_flags = (UTF_PATTERN) ? FOLDEQ_S2_ALREADY_FOLDED : 0; goto do_exactf_utf8; } |