rationalise mb<->wc conversions to use restartable form

7 files changed, 56 insertions, 45 deletions

diff --git a/ChangeLog b/ChangeLog
index 5776e62d4..41371caba 100644
--- a/ChangeLog
+++ b/ChangeLog
@@ -1,5 +1,10 @@
 2006-08-03  Peter Stephenson  <pws@csr.com>
 
+	* 22581: Src/prompt.c, Src/utils.c, Src/Zle/complist.c,
+	Src/Zle/zle.h, Src/Zle/zle_tricky.c, Src/Zle/zle_utils.c:
+	rationalise multibyte conversions to and from wide characters
+	to use restartable form.
+
 	* unposted: Completion/Unix/Command/_subversion: import
 	directory is optional, so complete URLs also on first argument.
 
@@ -58,7 +63,7 @@
 	accept-and-menu-complete, perhaps.
 
 	* 22562: Src/glob.c, Test/D07multibyte.ztst: make ${...#...} etc.
-	understand multibyte characters.	
+	understand multibyte characters.
 
 2006-07-29  Barton E. Schaefer  <schaefer@zsh.org>
 
diff --git a/Src/Zle/complist.c b/Src/Zle/complist.c
index 0bea231c5..8c6937b4a 100644
--- a/Src/Zle/complist.c
+++ b/Src/Zle/complist.c
@@ -588,6 +588,7 @@ clnicezputs(Listcols colors, char *s, int ml)
     if (colors)
 	initiscol(colors);
 
+    mb_metacharinit();
     while (umleft > 0) {
 	size_t cnt = eol ? MB_INVALID : mbrtowc(&cc, uptr, umleft, &mbs);
 
@@ -1964,7 +1965,9 @@ msearch(Cmatch **ptr, int ins, int back, int rep, int *wrapp)
 #ifdef MULTIBYTE_SUPPORT
 	if (lastchar_wide_valid)
 	{
-	    int len = wctomb(s, lastchar_wide);
+	    mbstate_t mbs;
+	    memset(&mbs, 0, sizeof(mbs));
+	    int len = wcrtomb(s, lastchar_wide, &mbs);
 	    if (len < 0)
 		len = 0;
 	    s[len] = '\0';
diff --git a/Src/Zle/zle.h b/Src/Zle/zle.h
index f56960734..200804c7d 100644
--- a/Src/Zle/zle.h
+++ b/Src/Zle/zle.h
@@ -71,8 +71,6 @@ typedef wint_t   ZLE_INT_T;
 #define ZC_tolower towlower
 #define ZC_toupper towupper
 
-#define ZC_nicechar(c) wcs_nicechar(c, NULL, NULL)
-
 #define LASTFULLCHAR	lastchar_wide
 
 #else  /* Not MULTIBYTE_SUPPORT: old single-byte code */
@@ -100,8 +98,6 @@ typedef int ZLE_INT_T;
  */
 #define ZMB_nicewidth	niceztrlen
 
-#define ZC_nicechar nicechar
-
 #ifdef __GNUC__
 static inline size_t ZS_strlen(ZLE_STRING_T s)
 { return strlen((char*)s); }
diff --git a/Src/Zle/zle_tricky.c b/Src/Zle/zle_tricky.c
index fb0ebad2a..4b7d3126d 100644
--- a/Src/Zle/zle_tricky.c
+++ b/Src/Zle/zle_tricky.c
@@ -2464,16 +2464,20 @@ magicspace(char **args)
     ZLE_STRING_T bangq;
     ZLE_CHAR_T zlebangchar[1];
     int ret;
+#ifdef MULTIBYTE_SUPPORT
+    mbstate_t mbs;
+#endif
+
     fixmagicspace();
 
 #ifdef MULTIBYTE_SUPPORT
     /*
-     * TODO: bangchar should really be a multibyte string representing
-     * a single character, since there's no fundamental reason why
-     * it shouldn't be a Unicode character.  In practice this is
-     * very minor, however.
+     * Use mbrtowc() here for consistency and to ensure the
+     * state is initialised properly.  bangchar is unsigned char,
+     * but must be ASCII, so we simply cast the pointer.
      */
-    if (mbtowc(zlebangchar, (char *)&bangchar, 1) < 0)
+    memset(&mbs, 0, sizeof(mbs));
+    if (mbrtowc(zlebangchar, (char *)&bangchar, 1, &mbs) < 0)
 	return selfinsert(args);
 #else
     zlebangchar[0] = bangchar;
diff --git a/Src/Zle/zle_utils.c b/Src/Zle/zle_utils.c
index cce162fa0..6275e74d3 100644
--- a/Src/Zle/zle_utils.c
+++ b/Src/Zle/zle_utils.c
@@ -135,17 +135,20 @@ zlelineasstring(ZLE_STRING_T instr, int inll, int incs, int *outllp,
     char *s;
     int i, j;
     size_t mb_len = 0;
+    mbstate_t mbs;
 
     s = zalloc(inll * MB_CUR_MAX + 1);
 
     outcs = 0;
+    memset(&mbs, 0, sizeof(mbs));
     for (i=0; i < inll; i++, incs--) {
 	if (incs == 0)
 	    outcs = mb_len;
-	j = wctomb(s + mb_len, instr[i]);
+	j = wcrtomb(s + mb_len, instr[i], &mbs);
 	if (j == -1) {
 	    /* invalid char; what to do? */
 	    s[mb_len++] = ZWC('?');
+	    memset(&mbs, 0, sizeof(mbs));
 	} else {
 	    mb_len += j;
 	}
@@ -780,6 +783,7 @@ showmsg(char const *msg)
     p = unmetafy(umsg, &ulen);
     memset(&mbs, 0, sizeof mbs);
 
+    mb_metacharinit();
     while (ulen > 0) {
 	char const *n;
 	if (*p == '\n') {
diff --git a/Src/prompt.c b/Src/prompt.c
index facf777f0..21dff16e0 100644
--- a/Src/prompt.c
+++ b/Src/prompt.c
@@ -771,6 +771,7 @@ stradd(char *d)
 	    /* FALL THROUGH */
 	default:
 	    /* Take full wide character in one go */
+	    mb_metacharinit();
 	    pc = wcs_nicechar(cc, NULL, NULL);
 	    break;
 	}
diff --git a/Src/utils.c b/Src/utils.c
index a81c4ed04..21b722f6a 100644
--- a/Src/utils.c
+++ b/Src/utils.c
@@ -279,6 +279,7 @@ zerrmsg(const char *fmt, va_list ap)
 	    case 'c':
 		num = va_arg(ap, int);
 #ifdef MULTIBYTE_SUPPORT
+		mb_metacharinit();
 		zputs(wcs_nicechar(num, NULL, NULL), stderr);
 #else
 		zputs(nicechar(num), stderr);
@@ -402,6 +403,20 @@ nicechar(int c)
 
 /**/
 #ifdef MULTIBYTE_SUPPORT
+static mbstate_t mb_shiftstate;
+
+/*
+ * Initialise multibyte state: called before a sequence of
+ * wcs_nicechar() or mb_metacharlenconv().
+ */
+
+/**/
+mod_export void
+mb_metacharinit(void)
+{
+    memset(&mb_shiftstate, 0, sizeof(mb_shiftstate));
+}
+
 /*
  * The number of bytes we need to allocate for a "nice" representation
  * of a multibyte character.
@@ -430,6 +445,9 @@ nicechar(int c)
  * Either the initial ASCII part or the wide character part may be empty
  * (but not both).  (Note the complication that the wide character
  * part may contain metafied characters.)
+ *
+ * The caller needs to call mb_metacharinit() before the first call, to
+ * set up the multibyte shift state for a range of characters.
  */
 
 /**/
@@ -475,8 +493,11 @@ wcs_nicechar(wchar_t c, size_t *widthp, char **swidep)
 	}
     }
 
-    if (ret == -1 ||
-	(ret = wctomb(mbstr, c)) == -1) {
+    if (ret != -1)
+	ret = wcrtomb(mbstr, c, &mb_shiftstate);
+
+    if (ret == -1) {
+	memset(&mb_shiftstate, 0, sizeof(mb_shiftstate));
 	/*
 	 * Can't or don't want to convert character: use UCS-2 or
 	 * UCS-4 code in print escape format.
@@ -513,18 +534,6 @@ wcs_nicechar(wchar_t c, size_t *widthp, char **swidep)
 /**/
 #endif /* MULTIBYTE_SUPPORT */
 
-
-/* Output a string's visible representation. */
-
-#if 0 /**/
-void
-nicefputs(char *s, FILE *f)
-{
-    for (; *s; s++)
-	zputs(nicechar(*s), f);
-}
-#endif
-
 /* get a symlink-free pathname for s relative to PWD */
 
 /**/
@@ -699,7 +708,7 @@ get_username(void)
 #ifdef HAVE_GETPWUID
     struct passwd *pswd;
     uid_t current_uid;
- 
+
     current_uid = getuid();
     if (current_uid != cached_uid) {
 	cached_uid = current_uid;
@@ -900,12 +909,12 @@ dircmp(char *s, char *t)
 mod_export LinkList prepromptfns;
 
 /* the last time we checked mail */
- 
+
 /**/
 time_t lastmailcheck;
- 
+
 /* the last time we checked the people in the WATCH variable */
- 
+
 /**/
 time_t lastwatch;
 
@@ -2818,6 +2827,7 @@ mod_export int
 wcsitype(wchar_t c, int itype)
 {
     int len;
+    mbstate_t mbs;
     VARARR(char, outstr, MB_CUR_MAX);
 
     if (!isset(MULTIBYTE))
@@ -2830,7 +2840,8 @@ wcsitype(wchar_t c, int itype)
      * If it doesn't, use iswalnum on the original character.
      * If that fails, resort to the appropriate wide character array.
      */
-    len = wctomb(outstr, c);
+    memset(&mbs, 0, sizeof(mbs));
+    len = wcrtomb(outstr, c, &mbs);
 
     if (len == 0) {
 	/* NULL is special */
@@ -3725,6 +3736,7 @@ mb_niceformat(const char *s, FILE *stream, char **outstrp, int heap)
     ptr = unmetafy(ums, &umlen);
 
     memset(&mbs, 0, sizeof mbs);
+    mb_metacharinit();
     while (umlen > 0) {
 	size_t cnt = eol ? MB_INVALID : mbrtowc(&c, ptr, umlen, &mbs);
 
@@ -3853,20 +3865,6 @@ mb_width(const char *s)
     return width;
 }
 
-static mbstate_t mb_shiftstate;
-
-/*
- * Initialise multibyte state: called before a sequence of
- * mb_metacharlenconv().
- */
-
-/**/
-void
-mb_metacharinit(void)
-{
-    memset(&mb_shiftstate, 0, sizeof(mb_shiftstate));
-}
-
 /*
  * Length of metafied string s which contains the next multibyte
  * character; single (possibly metafied) character if string is not null