[newlib-cygwin/main] Cygwin: regex: convert wchar_t to wint_t

[Corinna Vinschen <corinna@sourceware.org>]

https://sourceware.org/git/gitweb.cgi?p=newlib-cygwin.git;h=588624da2b0a921c7e72d12d19c2ac3321de0326

commit 588624da2b0a921c7e72d12d19c2ac3321de0326
Author:     Corinna Vinschen <corinna@vinschen.de>
AuthorDate: Tue Feb 14 13:02:15 2023 +0100
Commit:     Corinna Vinschen <corinna@vinschen.de>
CommitDate: Tue Feb 14 13:02:15 2023 +0100

    Cygwin: regex: convert wchar_t to wint_t
    
    - call mbrtowi instead of mbrtowc
    - drop Cygwin-only surrogate handling from wgetnext and xmbrtowc since
      it's encapsulated in mbrtowi.
    
    Signed-off-by: Corinna Vinschen <corinna@vinschen.de>

Diff:
---
 winsup/cygwin/regex/regcomp.c | 25 +++++--------------------
 winsup/cygwin/regex/regexec.c | 22 +++-------------------
 2 files changed, 8 insertions(+), 39 deletions(-)

diff --git a/winsup/cygwin/regex/regcomp.c b/winsup/cygwin/regex/regcomp.c
index a377e56fac22..aef104c1ea45 100644
--- a/winsup/cygwin/regex/regcomp.c
+++ b/winsup/cygwin/regex/regcomp.c
@@ -921,7 +921,7 @@ p_b_coll_elem(struct parse *p,
 	struct cname *cp;
 	int len;
 	mbstate_t mbs;
-	wchar_t wc;
+	wint_t wc;
 	size_t clen;
 
 	while (MORE() && !SEETWO(endc, ']'))
@@ -935,7 +935,7 @@ p_b_coll_elem(struct parse *p,
 		if (strncmp(cp->name, sp, len) == 0 && cp->name[len] == '\0')
 			return(cp->code);	/* known name */
 	memset(&mbs, 0, sizeof(mbs));
-	if ((clen = mbrtowc(&wc, sp, len, &mbs)) == len)
+	if ((clen = mbrtowi(&wc, sp, len, &mbs)) == len)
 		return (wc);			/* single character */
 	else if (clen == (size_t)-1 || clen == (size_t)-2)
 		SETERROR(REG_ILLSEQ);
@@ -1119,8 +1119,7 @@ static wint_t
 wgetnext(struct parse *p)
 {
 	mbstate_t mbs;
-	wchar_t wc;
-	wint_t ret;
+	wint_t wc;
 	size_t n;
 
 #ifdef __CYGWIN__
@@ -1136,29 +1135,15 @@ wgetnext(struct parse *p)
 	  return (wint_t) (unsigned char) *p->next++;
 #endif
 	memset(&mbs, 0, sizeof(mbs));
-	n = mbrtowc(&wc, p->next, p->end - p->next, &mbs);
+	n = mbrtowi(&wc, p->next, p->end - p->next, &mbs);
 	if (n == (size_t)-1 || n == (size_t)-2) {
 		SETERROR(REG_ILLSEQ);
 		return (0);
 	}
-	ret = wc;
 	if (n == 0)
 		n = 1;
-	else if (sizeof (wchar_t) == 2 && wc >= 0xd800 && wc <= 0xdbff) {
-		/* UTF-16 surrogate pair.  Fetch second half and
-		   compute UTF-32 value */
-		size_t n2 = mbrtowc(&wc, p->next + n,
-				    p->end - p->next - n, &mbs);
-		if (n2 == 0 || n2 == (size_t)-1 || n2 == (size_t)-2) {
-			SETERROR(REG_ILLSEQ);
-			return (0);
-		}
-		ret = (((ret & 0x3ff) << 10) | (wc & 0x3ff))
-		      + 0x10000;
-		n += n2;
-	  }
 	p->next += n;
-	return (ret);
+	return (wc);
 }
 
 static size_t
diff --git a/winsup/cygwin/regex/regexec.c b/winsup/cygwin/regex/regexec.c
index c400578b898d..94e95e65abcd 100644
--- a/winsup/cygwin/regex/regexec.c
+++ b/winsup/cygwin/regex/regexec.c
@@ -68,9 +68,9 @@ static __inline size_t
 xmbrtowc(wint_t *wi, const char *s, size_t n, mbstate_t *mbs, wint_t dummy)
 {
 	size_t nr;
-	wchar_t wc;
+	wint_t wc;
 
-	nr = mbrtowc(&wc, s, n, mbs);
+	nr = mbrtowi(&wc, s, n, mbs);
 	if (wi != NULL)
 		*wi = wc;
 	if (nr == 0)
@@ -80,24 +80,8 @@ xmbrtowc(wint_t *wi, const char *s, size_t n, mbstate_t *mbs, wint_t dummy)
 		if (wi != NULL)
 			*wi = dummy;
 		return (1);
-	} else {
-		if (sizeof (wchar_t) == 2 && wc >= 0xd800 && wc <= 0xdbff) {
-			/* UTF-16 surrogate pair.  Fetch second half and
-			   compute UTF-32 value */
-			size_t n2 = mbrtowc(&wc, s + nr, n - nr, mbs);
-			if (n2 == 0 || n2 == (size_t)-1 || n2 == (size_t)-2) {
-				memset(mbs, 0, sizeof(*mbs));
-				if (wi != NULL)
-					*wi = dummy;
-				return (1);
-			}
-			if (wi != NULL)
-				*wi = (((*wi & 0x3ff) << 10) | (wc & 0x3ff))
-				      + 0x10000;
-			nr += n2;
-		  }
+	} else
                 return (nr);
-	}
 }
 
 static __inline size_t