Rational Range Interpretation patches, 3/3

[Aharon Robbins]

>From 8b8d37d4e4bfae93920d17f7f1c81b8db26fcefd Mon Sep 17 00:00:00 2001
From: Arnold D. Robbins <address@hidden>
Date: Mon, 16 Jan 2012 22:17:58 +0200
Subject: [PATCH] Implement Rational Range Interpretation.

* regcomp.c (build_range_exp): Compare the wide characters
  directly instead of using wcscoll.
* regexec.c (check_node_accept_bytes): Ditto.
---
 lib/regcomp.c |   13 +++----------
 lib/regexec.c |   12 ++----------
 2 files changed, 5 insertions(+), 20 deletions(-)

diff --git a/lib/regcomp.c b/lib/regcomp.c
index 3398adb..bf7f9f8 100644
--- a/lib/regcomp.c
+++ b/lib/regcomp.c
@@ -2654,7 +2654,6 @@ build_range_exp (const reg_syntax_t syntax,
     wchar_t wc;
     wint_t start_wc;
     wint_t end_wc;
-    wchar_t cmp_buf[6] = {L'\0', L'\0', L'\0', L'\0', L'\0', L'\0'};
 
     start_ch = ((start_elem->type == SB_CHAR) ? start_elem->opr.ch
                : ((start_elem->type == COLL_SYM) ? start_elem->opr.name[0]
@@ -2668,11 +2667,7 @@ build_range_exp (const reg_syntax_t syntax,
              ? __btowc (end_ch) : end_elem->opr.wch);
     if (start_wc == WEOF || end_wc == WEOF)
       return REG_ECOLLATE;
-    cmp_buf[0] = start_wc;
-    cmp_buf[4] = end_wc;
-
-    if (BE ((syntax & RE_NO_EMPTY_RANGES)
-            && wcscoll (cmp_buf, cmp_buf + 4) > 0, 0))
+    else if ((syntax & RE_NO_EMPTY_RANGES) && start_wc > end_wc)
       return REG_ERANGE;
 
     /* Got valid collation sequence values, add them as a new entry.
@@ -2713,10 +2708,8 @@ build_range_exp (const reg_syntax_t syntax,
     /* Build the table for single byte characters.  */
     for (wc = 0; wc < SBC_MAX; ++wc)
       {
-       cmp_buf[2] = wc;
-       if (wcscoll (cmp_buf, cmp_buf + 2) <= 0
-           && wcscoll (cmp_buf + 2, cmp_buf + 4) <= 0)
-         bitset_set (sbcset, wc);
+         if (start_wc <= wc && wc <= end_wc)
+           bitset_set (sbcset, wc);
       }
   }
 # else /* not RE_ENABLE_I18N */
diff --git a/lib/regexec.c b/lib/regexec.c
index aa8e524..81066c1 100644
--- a/lib/regexec.c
+++ b/lib/regexec.c
@@ -3985,18 +3985,10 @@ check_node_accept_bytes (const re_dfa_t *dfa, Idx 
node_idx,
 # endif /* _LIBC */
        {
          /* match with range expression?  */
-#if __GNUC__ >= 2 && ! (__STDC_VERSION__ < 199901L && defined __STRICT_ANSI__)
-         wchar_t cmp_buf[] = {L'\0', L'\0', wc, L'\0', L'\0', L'\0'};
-#else
-         wchar_t cmp_buf[] = {L'\0', L'\0', L'\0', L'\0', L'\0', L'\0'};
-         cmp_buf[2] = wc;
-#endif
          for (i = 0; i < cset->nranges; ++i)
            {
-             cmp_buf[0] = cset->range_starts[i];
-             cmp_buf[4] = cset->range_ends[i];
-             if (wcscoll (cmp_buf, cmp_buf + 2) <= 0
-                 && wcscoll (cmp_buf + 2, cmp_buf + 4) <= 0)
+              if (cset->range_starts[i] <= wc
+                  && wc <= cset->range_ends[i])
                {
                  match_len = char_len;
                  goto check_node_accept_bytes_match;
-- 
1.7.1