diff options
| author | Nicholas Clark <nick@ccl4.org> | 2011-06-28 15:20:56 +0200 |
|---|---|---|
| committer | Nicholas Clark <nick@ccl4.org> | 2011-07-01 14:05:41 +0200 |
| commit | 72de20cdcc84ae35e0d8a55c7a92950fece19347 (patch) | |
| tree | fde310b665c99a781df72f0aeab31f0e8ba8225e /regexec.c | |
| parent | b606cf7f37b8b46206c7f521b29167e037397a62 (diff) | |
| download | perl-72de20cdcc84ae35e0d8a55c7a92950fece19347.tar.gz | |
For shorter strings, store C<study>'s data as U8s or U16s, instead of U32s.
The assumption is that most studied strings are fairly short, hence the pain
of the extra code is worth it, given the memory savings.
80 character string, 336 bytes as U8, down from 1344 as U32
800 character string, 2112 bytes as U16, down from 4224 as U32
Diffstat (limited to 'regexec.c')
| -rw-r--r-- | regexec.c | 12 |
1 files changed, 11 insertions, 1 deletions
@@ -696,12 +696,22 @@ Perl_re_intuit_start(pTHX_ REGEXP * const rx, SV *sv, char *strpos, I32 p = -1; /* Internal iterator of scream. */ I32 * const pp = data ? data->scream_pos : &p; const MAGIC *mg; + bool found = FALSE; assert(SvMAGICAL(sv)); mg = mg_find(sv, PERL_MAGIC_study); assert(mg); - if (((U32 *)mg->mg_ptr)[BmRARE(check)] != (U32)~0 + if (mg->mg_private == 1) { + found = ((U8 *)mg->mg_ptr)[BmRARE(check)] != (U8)~0; + } else if (mg->mg_private == 2) { + found = ((U16 *)mg->mg_ptr)[BmRARE(check)] != (U16)~0; + } else { + assert (mg->mg_private == 4); + found = ((U32 *)mg->mg_ptr)[BmRARE(check)] != (U32)~0; + } + + if (found || ( BmRARE(check) == '\n' && (BmPREVIOUS(check) == SvCUR(check) - 1) && SvTAIL(check) )) |