Finish RRI changes per points from Paolo Bonzini.

9 files changed, 41 insertions, 19 deletions

diff --git a/ChangeLog b/ChangeLog
index 660b7191..075809d5 100644
--- a/ChangeLog
+++ b/ChangeLog
@@ -1,3 +1,13 @@
+2011-12-26         Arnold D. Robbins     <arnold@skeeve.com>
+
+	Finish Rational Range Interpretation (!)
+
+	* dfa.c (match_mb_charset): Compare wide characters directly
+	instead of using wcscoll().
+	* regexec.c (check_node_accept_byte): Ditto.
+
+	Thanks to Paolo Bonzini for pointing these out.
+
 2011-12-06         John Haque      <j.eh@mchsi.com>
 
 	* debug.c (source_find): Fix misplaced call to efree.
diff --git a/dfa.c b/dfa.c
index 37ddc1bc..acd1a947 100644
--- a/dfa.c
+++ b/dfa.c
@@ -2887,7 +2887,6 @@ match_mb_charset (struct dfa *d, int s, position pos, int idx)
                            with which this operator match.  */
   int op_len;		/* Length of the operator.  */
   char buffer[128];
-  wchar_t wcbuf[6];
 
   /* Pointer to the structure to which we are currently refering.  */
   struct mb_char_classes *work_mbc;
@@ -2964,17 +2963,11 @@ match_mb_charset (struct dfa *d, int s, position pos, int idx)
         }
     }
 
-  wcbuf[0] = wc;
-  wcbuf[1] = wcbuf[3] = wcbuf[5] = '\0';
-
   /* match with a range?  */
   for (i = 0; i<work_mbc->nranges; i++)
     {
-      wcbuf[2] = work_mbc->range_sts[i];
-      wcbuf[4] = work_mbc->range_ends[i];
-
-      if (wcscoll(wcbuf, wcbuf+2) >= 0 &&
-          wcscoll(wcbuf+4, wcbuf) >= 0)
+      if (work_mbc->range_sts[i] <= wc &&
+          wc <= work_mbc->range_ends[i])
         goto charset_matched;
     }
 
diff --git a/regexec.c b/regexec.c
index 97fcba00..dcd325c4 100644
--- a/regexec.c
+++ b/regexec.c
@@ -3963,18 +3963,10 @@ check_node_accept_bytes (const re_dfa_t *dfa, int node_idx,
 # endif /* _LIBC */
 	{
 	  /* match with range expression?  */
-#if __GNUC__ >= 2
-	  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;
diff --git a/test/ChangeLog b/test/ChangeLog
index 1ceb5b35..711d8b20 100644
--- a/test/ChangeLog
+++ b/test/ChangeLog
@@ -1,3 +1,8 @@
+2011-12-26         Arnold D. Robbins     <arnold@skeeve.com>
+
+	* Makefile.am (rri1): New test.
+	* rri1.awk, rri1.in, rri1.ok: New files.
+
 2011-12-06         Arnold D. Robbins     <arnold@skeeve.com>
 
 	* Makefile.am: Rationalize the $(CMP) lines wherever possible.
diff --git a/test/Makefile.am b/test/Makefile.am
index aae30518..f7d50b28 100644
--- a/test/Makefile.am
+++ b/test/Makefile.am
@@ -625,6 +625,9 @@ EXTRA_DIST = \
 	resplit.awk \
 	resplit.in \
 	resplit.ok \
+	rri1.awk \
+	rri1.in \
+	rri1.ok \
 	rs.awk \
 	rs.in \
 	rs.ok \
@@ -803,6 +806,7 @@ BASIC_TESTS = \
 	paramres paramtyp parse1 parsefld parseme pcntplus posix2008sub \
 	prdupval prec printf0 printf1 prmarscl prmreuse prt1eval prtoeval \
 	rand range1 rebt8b1 redfilnm regeq regrange reindops reparse resplit \
+	rri1 \
 	rs rsnul1nl rsnulbig rsnulbig2 rstest1 rstest2 rstest3 rstest4 \
 	rstest5 rswhite scalar sclforin sclifin sortempty splitargv \
 	splitarr splitdef splitvar splitwht strcat1 strnum1 strtod subamp \
@@ -1433,6 +1437,12 @@ exit:
 	@-AWK="$(AWKPROG)" $(srcdir)/$@.sh > _$@ 2>&1
 	@-$(CMP) $(srcdir)/$@.ok _$@ && rm -f _$@
 
+rri1::
+	@echo $@
+	@[ -z "$$GAWKLOCALE" ] && GAWKLOCALE=en_US.UTF-8; \
+	AWKPATH=$(srcdir) $(AWK) -f $@.awk  < $(srcdir)/$@.in >_$@ 2>&1 || echo EXIT CODE: $$? >>_$@
+	@-$(CMP) $(srcdir)/$@.ok _$@ && rm -f _$@
+
 # Targets generated for other tests:
 include Maketests
 
diff --git a/test/Makefile.in b/test/Makefile.in
index 800c55d8..05946f32 100644
--- a/test/Makefile.in
+++ b/test/Makefile.in
@@ -809,6 +809,9 @@ EXTRA_DIST = \
 	resplit.awk \
 	resplit.in \
 	resplit.ok \
+	rri1.awk \
+	rri1.in \
+	rri1.ok \
 	rs.awk \
 	rs.in \
 	rs.ok \
@@ -987,6 +990,7 @@ BASIC_TESTS = \
 	paramres paramtyp parse1 parsefld parseme pcntplus posix2008sub \
 	prdupval prec printf0 printf1 prmarscl prmreuse prt1eval prtoeval \
 	rand range1 rebt8b1 redfilnm regeq regrange reindops reparse resplit \
+	rri1 \
 	rs rsnul1nl rsnulbig rsnulbig2 rstest1 rstest2 rstest3 rstest4 \
 	rstest5 rswhite scalar sclforin sclifin sortempty splitargv \
 	splitarr splitdef splitvar splitwht strcat1 strnum1 strtod subamp \
@@ -1782,6 +1786,12 @@ exit:
 	@echo $@
 	@-AWK="$(AWKPROG)" $(srcdir)/$@.sh > _$@ 2>&1
 	@-$(CMP) $(srcdir)/$@.ok _$@ && rm -f _$@
+
+rri1::
+	@echo $@
+	@[ -z "$$GAWKLOCALE" ] && GAWKLOCALE=en_US.UTF-8; \
+	AWKPATH=$(srcdir) $(AWK) -f $@.awk  < $(srcdir)/$@.in >_$@ 2>&1 || echo EXIT CODE: $$? >>_$@
+	@-$(CMP) $(srcdir)/$@.ok _$@ && rm -f _$@
 Gt-dummy:
 # file Maketests, generated from Makefile.am by the Gentests program
 addcomma:
diff --git a/test/rri1.awk b/test/rri1.awk
new file mode 100644
index 00000000..889dbdcb
--- /dev/null
+++ b/test/rri1.awk
@@ -0,0 +1 @@
+/[d-f]/
diff --git a/test/rri1.in b/test/rri1.in
new file mode 100644
index 00000000..28b6b408
--- /dev/null
+++ b/test/rri1.in
@@ -0,0 +1 @@
+no match: è
diff --git a/test/rri1.ok b/test/rri1.ok
new file mode 100644
index 00000000..e69de29b
--- /dev/null
+++ b/test/rri1.ok