allow raw (ASCII) matching via REG_RAW flag, so we no longer need a compile-time flag

You no longer need to build two libraries just for supporting raw/ASCII and UTF-8
patterns.
This makes using the library a lot easier and reduces the total binary size.
Since strings are always `unsigned char *` now internally and the raw vs. UTF-8
decision is important only in a few select places, it would make no sense to
use meta-programming techniques.

The test suite has been extended and is now fixed for cases with embedded
non-printable characters.

1 files changed, 14 insertions, 14 deletions

diff --git a/regc_lex.c b/regc_lex.c
index 2033913..f7b4c95 100644
--- a/regc_lex.c
+++ b/regc_lex.c
@@ -47,15 +47,15 @@
 #define	LASTTYPE(t)	(v->lasttype == (t))
 
 /* return and skip the next (unicode) character */
-#ifdef REGEX_UTF8
-#define SKIPCHR(x)	do { \
-	wchar_t __c; \
-	v->now += mbtowc(&__c, (const char *)v->now, v->stop - v->now); \
-	x = __c; \
-} while (0)
-#else
-#define SKIPCHR(x)	do x = *v->now++; while (0)
-#endif
+static inline pchr
+skipchr(struct vars *v)
+{
+	if (v->cflags & REG_RAW)
+		return *v->now++;
+	wchar_t c;
+	v->now += mbtowc(&c, (const char *)v->now, v->stop - v->now);
+	return c;
+}
 
 /* lexical contexts */
 #define	L_ERE	1	/* mainline ERE/ARE */
@@ -383,7 +383,7 @@ next(
      * Okay, time to actually get a character.
      */
 
-    SKIPCHR(c);
+    c = skipchr(v);
 
     /*
      * Deal with the easy contexts, punt EREs to code below.
@@ -709,7 +709,7 @@ next(
 
     assert(!ATEOS());
     if (!(v->cflags&REG_ADVF)) {/* only AREs have non-trivial escapes */
-	SKIPCHR(c);
+	c = skipchr(v);
 	if (iscalnum(c)) {
 	    NOTE(REG_UBSALNUM);
 	    NOTE(REG_UUNSPEC);
@@ -766,7 +766,7 @@ lexescape(
     assert(v->cflags&REG_ADVF);
 
     assert(!ATEOS());
-    SKIPCHR(c);
+    c = skipchr(v);
     if (!iscalnum(c)) {
 	RETV(PLAIN, c);
     }
@@ -790,7 +790,7 @@ lexescape(
 	if (ATEOS()) {
 	    FAILW(REG_EESCAPE);
 	}
-	SKIPCHR(c);
+	c = skipchr(v);
 	RETV(PLAIN, c & 037);
 	break;
     case CHR('d'):
@@ -1055,7 +1055,7 @@ brenext(
 	FAILW(REG_EESCAPE);
     }
 
-    SKIPCHR(c);
+    c = skipchr(v);
     switch (c) {
     case CHR('{'):
 	INTOCON(L_BBND);