Associate a multibyte conversion state object with each stream. Reset it

to the initial state when a stream is opened or seeked upon. Use the
stream's conversion state object instead of a freshly-zeroed one in
fgetwc(), fputwc() and ungetwc().

This is only a performance improvement for now, but it would also be
required in order to support state-dependent encodings.

lib/libc/stdio/fgetwc.c

@@ -1,5 +1,5 @@
 /*-
- * Copyright (c) 2002 Tim J. Robbins.
+ * Copyright (c) 2002-2004 Tim J. Robbins.
  * All rights reserved.
  *
  * Redistribution and use in source and binary forms, with or without
@@ -77,34 +77,31 @@ fgetwc(FILE *fp)
 static __inline wint_t
 __fgetwc_nbf(FILE *fp)
 {
-	static const mbstate_t initial;
-	mbstate_t mbs;
-	char buf[MB_LEN_MAX];
 	size_t n, nconv;
 	int c;
+	char cc;
 	wchar_t wc;
 
 	n = 0;
-	while (n < MB_CUR_MAX) {
+	for (;;) {
 		if ((c = __sgetc(fp)) == EOF) {
 			if (n == 0)
 				return (WEOF);
 			break;
 		}
-		buf[n++] = (char)c;
-		mbs = initial;
-		nconv = mbrtowc(&wc, buf, n, &mbs);
-		if (nconv == n)
-			return (wc);
-		else if (nconv == 0)
-			return (L'\0');
+		n++;
+		cc = (char)c;
+		nconv = mbrtowc(&wc, &cc, 1, &fp->_extra->mbstate);
+		if (nconv == (size_t)-2)
+			continue;
 		else if (nconv == (size_t)-1)
 			break;
+		else if (nconv == 0)
+			return (L'\0');
+		else
+			return (wc);
 	}
-
-	while (n-- != 0)
-		__ungetc((unsigned char)buf[n], fp);
-	errno = EILSEQ;
 	fp->_flags |= __SERR;
+	errno = EILSEQ;
 	return (WEOF);
 }

lib/libc/stdio/findfp.c

@@ -180,11 +180,7 @@ found:
 	fp->_lb._size = 0;
 /*	fp->_lock = NULL; */	/* once set always set (reused) */
 	fp->_extra->orientation = 0;
-#ifdef notdef
-	/* Stateful encoding/decoding is not yet supported. */
-	memset(&fp->_extra->wstate, 0, sizeof(mbstate_t));
-	memset(&fp->_extra->rstate, 0, sizeof(mbstate_t));
-#endif
+	memset(&fp->_extra->mbstate, 0, sizeof(mbstate_t));
 	return (fp);
 }
 

lib/libc/stdio/fputwc.c

@@ -1,5 +1,5 @@
 /*-
- * Copyright (c) 2002 Tim J. Robbins.
+ * Copyright (c) 2002-2004 Tim J. Robbins.
  * All rights reserved.
  *
  * Redistribution and use in source and binary forms, with or without
@@ -43,8 +43,6 @@ __FBSDID("$FreeBSD$");
 wint_t
 __fputwc(wchar_t wc, FILE *fp)
 {
-	static const mbstate_t initial;
-	mbstate_t mbs;
 	char buf[MB_LEN_MAX];
 	size_t i, len;
 
@@ -57,8 +55,8 @@ __fputwc(wchar_t wc, FILE *fp)
 		*buf = (unsigned char)wc;
 		len = 1;
 	} else {
-		mbs = initial;
-		if ((len = wcrtomb(buf, wc, &mbs)) == (size_t)-1) {
+		if ((len = wcrtomb(buf, wc, &fp->_extra->mbstate)) ==
+		    (size_t)-1) {
 			fp->_flags |= __SERR;
 			return (WEOF);
 		}

lib/libc/stdio/freopen.c

@@ -188,6 +188,7 @@ finish:
 		FREELB(fp);
 	fp->_lb._size = 0;
 	fp->_extra->orientation = 0;
+	memset(&fp->_extra->mbstate, 0, sizeof(mbstate_t));
 
 	if (f < 0) {			/* did not get it after all */
 		fp->_flags = 0;		/* set it free */

lib/libc/stdio/fseek.c

@@ -259,6 +259,7 @@ _fseeko(fp, offset, whence, ltest)
 		if (HASUB(fp))
 			FREEUB(fp);
 		fp->_flags &= ~__SEOF;
+		memset(&fp->_extra->mbstate, 0, sizeof(mbstate_t));
 		return (0);
 	}
 
@@ -303,6 +304,7 @@ dumb:
 	fp->_r = 0;
 	/* fp->_w = 0; */	/* unnecessary (I think...) */
 	fp->_flags &= ~__SEOF;
+	memset(&fp->_extra->mbstate, 0, sizeof(mbstate_t));
 	if (ltest && ret > LONG_MAX) {
 		fp->_flags |= __SERR;
 		errno = EOVERFLOW;

lib/libc/stdio/local.h

@@ -89,15 +89,7 @@ struct __sFILEX {
 	pthread_t	fl_owner;	/* current owner */
 	int		fl_count;	/* recursive lock count */
 	int		orientation;	/* orientation for fwide() */
-#ifdef notdef
-	/*
-	 * XXX These are not used yet -- they will be used to store the
-	 * multibyte conversion state for writing and reading when
-	 * stateful encodings are supported by the locale framework.
-	 */
-	mbstate_t	wstate;		/* write conversion state */
-	mbstate_t	rstate;		/* read conversion state */
-#endif
+	mbstate_t	mbstate;	/* multibyte conversion state */
 };
 
 /*
@@ -134,8 +126,7 @@ struct __sFILEX {
 	(fp)->_extra->fl_owner = NULL; \
 	(fp)->_extra->fl_count = 0; \
 	(fp)->_extra->orientation = 0; \
-	/* memset(&(fp)->_extra->wstate, 0, sizeof(mbstate_t)); */ \
-	/* memset(&(fp)->_extra->rstate, 0, sizeof(mbstate_t)); */ \
+	memset(&(fp)->_extra->mbstate, 0, sizeof(mbstate_t)); \
 }
 
 /*

lib/libc/stdio/ungetwc.c

@@ -1,5 +1,5 @@
 /*-
- * Copyright (c) 2002 Tim J. Robbins.
+ * Copyright (c) 2002-2004 Tim J. Robbins.
  * All rights reserved.
  *
  * Redistribution and use in source and binary forms, with or without
@@ -42,15 +42,12 @@ __FBSDID("$FreeBSD$");
 wint_t
 __ungetwc(wint_t wc, FILE *fp)
 {
-	static const mbstate_t initial;
-	mbstate_t mbs;
 	char buf[MB_LEN_MAX];
 	size_t len;
 
 	if (wc == WEOF)
 		return (WEOF);
-	mbs = initial;
-	if ((len = wcrtomb(buf, wc, &mbs)) == (size_t)-1) {
+	if ((len = wcrtomb(buf, wc, &fp->_extra->mbstate)) == (size_t)-1) {
 		fp->_flags |= __SERR;
 		return (WEOF);
 	}