regexec: Do less work on quantified UTF-8

[perl5.git] / utf8.c

diff --git a/utf8.c b/utf8.c
index 88ca041..8ad0478 100644 (file)
--- a/utf8.c
+++ b/utf8.c
@@ -1096,10 +1096,7 @@ Perl_utf8_length(pTHX_ const U8 *s, const U8 *e)
     if (e < s)
        goto warn_and_return;
     while (s < e) {
-       if (!UTF8_IS_INVARIANT(*s))
-           s += UTF8SKIP(s);
-       else
-           s++;
+        s += UTF8SKIP(s);
        len++;
     }
 
@@ -2229,86 +2226,6 @@ Perl_is_utf8_X_extend(pTHX_ const U8 *p)
     return is_utf8_common(p, &PL_utf8_X_extend, "_X_Extend");
 }
 
-/* These constants are for finding GCB=LV and GCB=LVT.  These are for the
- * pre-composed Hangul syllables, which are all in a contiguous block and
- * arranged there in such a way so as to facilitate alorithmic determination of
- * their characteristics.  As such, they don't need a swash, but can be
- * determined by simple arithmetic.  Almost all are GCB=LVT, but every 28th one
- * is a GCB=LV */
-#define SBASE 0xAC00    /* Start of block */
-#define SCount 11172    /* Length of block */
-#define TCount 28
-
-#if 0   /* This routine is not currently used */
-bool
-Perl_is_utf8_X_LV(pTHX_ const U8 *p)
-{
-    /* Unlike most other similarly named routines here, this does not create a
-     * swash, so swash_fetch() cannot be used on PL_utf8_X_LV. */
-
-    dVAR;
-
-    UV cp = valid_utf8_to_uvchr(p, NULL);
-
-    PERL_ARGS_ASSERT_IS_UTF8_X_LV;
-
-    /* The earliest Unicode releases did not have these precomposed Hangul
-     * syllables.  Set to point to undef in that case, so will return false on
-     * every call */
-    if (! PL_utf8_X_LV) {   /* Set up if this is the first time called */
-        PL_utf8_X_LV = swash_init("utf8", "_X_GCB_LV", &PL_sv_undef, 1, 0);
-        if (_invlist_len(_get_swash_invlist(PL_utf8_X_LV)) == 0) {
-            SvREFCNT_dec(PL_utf8_X_LV);
-            PL_utf8_X_LV = &PL_sv_undef;
-        }
-    }
-
-    return (PL_utf8_X_LV != &PL_sv_undef
-            && cp >= SBASE && cp < SBASE + SCount
-            && (cp - SBASE) % TCount == 0); /* Only every TCount one is LV */
-}
-#endif
-
-bool
-Perl_is_utf8_X_LVT(pTHX_ const U8 *p)
-{
-    /* Unlike most other similarly named routines here, this does not create a
-     * swash, so swash_fetch() cannot be used on PL_utf8_X_LVT. */
-
-    dVAR;
-
-    UV cp = valid_utf8_to_uvchr(p, NULL);
-
-    PERL_ARGS_ASSERT_IS_UTF8_X_LVT;
-
-    /* The earliest Unicode releases did not have these precomposed Hangul
-     * syllables.  Set to point to undef in that case, so will return false on
-     * every call */
-    if (! PL_utf8_X_LVT) {   /* Set up if this is the first time called */
-        PL_utf8_X_LVT = swash_init("utf8", "_X_GCB_LVT", &PL_sv_undef, 1, 0);
-        if (_invlist_len(_get_swash_invlist(PL_utf8_X_LVT)) == 0) {
-            SvREFCNT_dec(PL_utf8_X_LVT);
-            PL_utf8_X_LVT = &PL_sv_undef;
-        }
-    }
-
-    return (PL_utf8_X_LVT != &PL_sv_undef
-            && cp >= SBASE && cp < SBASE + SCount
-            && (cp - SBASE) % TCount != 0); /* All but every TCount one is LV */
-}
-
-
-bool
-Perl__is_utf8_quotemeta(pTHX_ const U8 *p)
-{
-    /* For exclusive use of pp_quotemeta() */
-
-    dVAR;
-
-    PERL_ARGS_ASSERT__IS_UTF8_QUOTEMETA;
-
-    return is_utf8_common(p, &PL_utf8_quotemeta, "_Perl_Quotemeta");
-}
 /*
 =for apidoc to_utf8_case
 
@@ -3689,7 +3606,7 @@ HV*
 Perl__swash_inversion_hash(pTHX_ SV* const swash)
 {
 
-   /* Subject to change or removal.  For use only in one place in regcomp.c.
+   /* Subject to change or removal.  For use only in regcomp.c and regexec.c
     * Can't be used on a property that is subject to user override, as it
     * relies on the value of SPECIALS in the swash which would be set by
     * utf8_heavy.pl to the hash in the non-overriden file, and hence is not set