minor readline fixes for isearch, insert-char,quoted-insert, and do-lowercase-version; fix buffer overflow with printf %b; work around issues with isblank in UTF-8 locales on macOS; fixes for the enable builtin

CWRU/CWRU.chlog

@@ -6442,3 +6442,55 @@ eval.c
 	  back to 0 if we should not exit. This can occur when EOF delimits
 	  a simple command (simple_command_terminator)
 	  From a report by Grisha Levit <grishalevit@gmail.com>
+
+				   5/27
+				   ----
+lib/readline/mbutil.c
+	- _rl_get_char_len: if SRC is "", make sure to return 0
+
+lib/readline/isearch.c
+	- _rl_isearch_callback: if _rl_search_getchar returns EOF, return 1
+	  to abort search; if it returns with RL_STATE_ISEARCH unset, return
+	  1 (handling a signal can turn it off via _rl_isearch_cleanup)
+
+				   5/29
+				   ----
+lib/readline/text.c
+	- _rl_insert_char, rl_quoted_insert: minor fixes from Grisha Levit
+
+builtins/printf.def
+	- bexpand: use the same calculation for the size of the return buffer
+	  as ansicstr().
+	  From a report by Grisha Levit <grishalevit@gmail.com>
+
+locale.c
+	- locale_setblanks: work around problem with macOS where isblank(x)
+	  incorrectly returns true for characters 0x80-0xff in a UTF-8 locale
+	  From a report by Grisha Levit <grishalevit@gmail.com>
+	- set_default_locale_vars,set_locale_vars,reset_locale_vars: make sure
+	  we call locale_setblanks after setting/resetting locale_utf8locale,
+	  since we use that in locale_setblanks
+
+lib/sh/smatch.c
+	- charseqcmp,is_cclass,FOLD: make sure we don't treat chars >= 0x80
+	  as valid in a UTF-8 locale, since they are multibyte characters.
+	  From a report by Grisha Levit <grishalevit@gmail.com> with a full
+	  explanation in http://www.openradar.me/FB9973780
+
+				   5/30
+				   ----
+builtins/enable.def
+	- enable_builtin: don't call dlopen with a null pathname, it will
+	  succeed. Report from Wiley Young <wyeth2485@gmail.com>
+
+lib/readline/readline.c
+	- _rl_dispatch_subseq: if a user binds do-lowercase-version to something
+	  that isn't an uppercase character, flag an error rather than recurse.
+	  From a report by Grisha Levit <grishalevit@gmail.com>
+
+				   5/31
+				   ----
+builtins/enable.def
+	- enable_builtin: don't try to load a builtin that's not found if
+	  the -n flag is supplied.
+	  From a report from Emanuele Torre <torreemanuele6@gmail.com>

builtins/enable.def

@@ -215,20 +215,23 @@ enable_builtin (WORD_LIST *list)
     {
       while (list)
 	{
-	  opt = enable_shell_command (list->word->word, flags & NFLAG);
+	  char *command;
+
+	  command = list->word->word;
+	  opt = enable_shell_command (command, flags & NFLAG);
 	  next = list->next;
 
 #if defined (HAVE_DLOPEN) && defined (HAVE_DLSYM)
 	  /* If we try to enable a non-existent builtin, and we have dynamic
 	     loading, try the equivalent of `enable -f name name'. */
-	  if (opt == EX_NOTFOUND)
+	  if (*command && (flags & NFLAG) == 0 && opt == EX_NOTFOUND)
 	    {
 	      int dflags, r;
 	      
 	      dflags = ENABLED|SILENT|((flags & SFLAG) ? SPECIAL : 0);
 
 	      list->next = 0;
-	      r = dyn_load_builtin (list, dflags, list->word->word);
+	      r = dyn_load_builtin (list, dflags, command);
 	      list->next = next;
 	      if (r == EXECUTION_SUCCESS)
 		opt = r;
@@ -240,7 +243,7 @@ enable_builtin (WORD_LIST *list)
 
 	  if (opt == EX_NOTFOUND)
 	    {
-	      sh_notbuiltin (list->word->word);
+	      sh_notbuiltin (command);
 	      result = EXECUTION_FAILURE;
 	    }
 	  else if (opt != EXECUTION_SUCCESS)

builtins/printf.def

@@ -1152,7 +1152,16 @@ bexpand (char *string, int len, int *sawc, int *lenp)
       return (ret);
     }
 
+#if defined (HANDLE_MULTIBYTE)
+  /* same logic as lib/sh/strtrans.c:ansicstr() */
+  temp = 4*len + 4;
+  if (temp < 12)
+    temp = 12;			/* ensure enough for eventual u32cesc */
+  ret = (char *)xmalloc (temp);
+#else
   ret = (char *)xmalloc (len + 1);
+#endif
+  temp = 0;
   for (r = ret, s = string; s && *s; )
     {
       c = *s++;

include/shmbutil.h

@@ -55,6 +55,9 @@ extern int locale_utf8locale;	/* XXX */
 #define UTF8_MBFIRSTCHAR(c)	(((c) & 0xc0) == 0xc0)
 #define UTF8_MBCHAR(c)		(((c) & 0xc0) == 0x80)
 
+/* Is an eight-bit quantity a valid character in the current locale? */
+#define VALID_SINGLEBYTE_CHAR(c)  (locale_utf8locale == 0 || ((c) & 0x80) == 0)
+
 #else /* !HANDLE_MULTIBYTE */
 
 #undef MB_LEN_MAX
@@ -83,6 +86,8 @@ extern int locale_utf8locale;	/* XXX */
 #define UTF8_SINGLEBYTE(c)	(1)
 #define UTF8_MBFIRSTCHAR(c)	(0)
 
+#defined VALID_SINGLEBYTE_CHAR(c)  (1)
+
 #endif /* !HANDLE_MULTIBYTE */
 
 /* Declare and initialize a multibyte state.  Call must be terminated

lib/glob/smatch.c

@@ -141,6 +141,10 @@ rangecmp (int c1, int c2, int forcecoll)
 static int
 collseqcmp (int c, int equiv)
 {
+  /* Make sure characters >= 0x80 are compared as bytes in a UTF-8 locale. */
+  if (locale_utf8locale && (UTF8_SINGLEBYTE (c) == 0 || UTF8_SINGLEBYTE (equiv) == 0))
+    return (c == equiv);
+
   if (charcmp (c, equiv, 1) == 0)
     return 1;
 
@@ -281,6 +285,9 @@ is_cclass (int c, const char *name)
   enum char_class char_class;
   int result;
 
+  if (locale_utf8locale && UTF8_SINGLEBYTE (c) == 0)
+    return -1;
+
   char_class = is_valid_cclass (name);
   if (char_class == CC_NO_CLASS)
     return -1;
@@ -291,9 +298,10 @@ is_cclass (int c, const char *name)
 
 /* Now include `sm_loop.c' for single-byte characters. */
 /* The result of FOLD is an `unsigned char' */
-# define FOLD(c) ((flags & FNM_CASEFOLD) \
-	? TOLOWER ((unsigned char)c) \
-	: ((unsigned char)c))
+# define FOLD(c) \
+  (((flags & FNM_CASEFOLD) && (locale_utf8locale == 0 || UTF8_SINGLEBYTE (c))) \
+    ? TOLOWER ((unsigned char)c) \
+    : ((unsigned char)c))
 
 #if !defined (__CYGWIN__)
 #  define ISDIRSEP(c)	((c) == '/')

lib/readline/isearch.c

@@ -919,6 +919,13 @@ _rl_isearch_callback (_rl_search_cxt *cxt)
   int c, r;
 
   c = _rl_search_getchar (cxt);
+
+  if (c < 0)					/* EOF */
+    return 1;
+
+  if (RL_ISSTATE (RL_STATE_ISEARCH) == 0)	/* signal could turn it off */
+    return 1;
+
   /* We might want to handle EOF here */
   r = _rl_isearch_dispatch (cxt, cxt->lastc);
 

lib/readline/mbutil.c

@@ -361,8 +361,8 @@ _rl_get_char_len (const char *src, mbstate_t *ps)
   int mb_cur_max;
 
   /* Look at no more than MB_CUR_MAX characters */
-  l = (size_t)strlen (src);
-  if (_rl_utf8locale && l > 0 && UTF8_SINGLEBYTE(*src))
+  l = strlen (src);
+  if (_rl_utf8locale && l >= 0 && UTF8_SINGLEBYTE(*src))
     tmp = (*src != 0) ? 1 : 0;
   else
     {

lib/readline/readline.c

@@ -905,8 +905,17 @@ _rl_dispatch_subseq (register int key, Keymap map, int got_subseq)
 	{
 	  /* Special case rl_do_lowercase_version (). */
 	  if (func == rl_do_lowercase_version)
-	    /* Should we do anything special if key == ANYOTHERKEY? */
-	    return (_rl_dispatch (_rl_to_lower ((unsigned char)key), map));
+	    {
+	      /* Should we do anything special if key == ANYOTHERKEY? */
+	      newkey = _rl_to_lower ((unsigned char)key);
+	      if (newkey != key)
+		return (_rl_dispatch (newkey, map));
+	      else
+		{
+		  rl_ding ();		/* gentle failure */
+		  return 0;
+		}
+	    }
 
 	  rl_executing_keymap = map;
 	  rl_executing_key = key;

lib/readline/text.c

@@ -853,7 +853,11 @@ _rl_insert_char (int count, int c)
       rl_insert_text (string);
       xfree (string);
 
+#if defined (HANDLE_MULTIBYTE)
       return (pending_bytes_length != 0);
+#else
+      return 0;
+#endif
     }
 
   if (count > TEXT_COUNT_MAX)
@@ -1112,6 +1116,8 @@ rl_quoted_insert (int count, int key)
 	r = _rl_insert_next (1);
       while (r == 0 && ++count < 0);
     }
+  else
+    r = _rl_insert_next (count);
 
   if (r == 1)
     _rl_insert_char (0, 0);	/* insert partial multibyte character */

locale.c

@@ -112,9 +112,9 @@ set_default_locale_vars (void)
   if (val == 0 && lc_all && *lc_all)
     {
       setlocale (LC_CTYPE, lc_all);
-      locale_setblanks ();
       locale_mb_cur_max = MB_CUR_MAX;
       locale_utf8locale = locale_isutf8 (lc_all);
+      locale_setblanks ();
 
 #    if defined (HANDLE_MULTIBYTE)
       locale_shiftstates = mblen ((char *)NULL, 0);
@@ -219,11 +219,11 @@ set_locale_var (const char *var, const char *value)
 	  else
 	    internal_warning(_("setlocale: LC_ALL: cannot change locale (%s): %s"), lc_all, strerror (errno));
 	}
-      locale_setblanks ();
       locale_mb_cur_max = MB_CUR_MAX;
       /* if LC_ALL == "", reset_locale_vars has already called this */
       if (*lc_all && x)
 	locale_utf8locale = locale_isutf8 (lc_all);
+      locale_setblanks ();
 #  if defined (HANDLE_MULTIBYTE)
       locale_shiftstates = mblen ((char *)NULL, 0);
 #  else
@@ -243,11 +243,11 @@ set_locale_var (const char *var, const char *value)
       if (lc_all == 0 || *lc_all == '\0')
 	{
 	  x = setlocale (LC_CTYPE, get_locale_var ("LC_CTYPE"));
-	  locale_setblanks ();
 	  locale_mb_cur_max = MB_CUR_MAX;
 	  /* if setlocale() returns NULL, the locale is not changed */
 	  if (x)
 	    locale_utf8locale = locale_isutf8 (x);
+	  locale_setblanks ();
 #if defined (HANDLE_MULTIBYTE)
 	  locale_shiftstates = mblen ((char *)NULL, 0);
 #else
@@ -396,10 +396,10 @@ reset_locale_vars (void)
     retval = 0;
 #  endif
 
-  locale_setblanks ();  
   locale_mb_cur_max = MB_CUR_MAX;
   if (x)
     locale_utf8locale = locale_isutf8 (x);
+  locale_setblanks ();  
 #  if defined (HANDLE_MULTIBYTE)
   locale_shiftstates = mblen ((char *)NULL, 0);
 #  else
@@ -575,6 +575,12 @@ locale_expand (const char *string, int start, int end, int lineno, size_t *lenp)
 }
 #endif
 
+/* Is character C in the [:blank:] class in the current locale? Work around
+   systems like macOS that treat some characters 0x80-0xff as blanks even in
+   a UTF-8 locale where they are multibyte characters. */
+#define locale_isblank(c) \
+  ((locale_utf8locale == 0 || ((c) & 0x80) == 0) && isblank ((unsigned char)(c)))
+
 /* Set every character in the <blank> character class to be a shell break
    character for the lexical analyzer when the locale changes. */
 static void
@@ -584,7 +590,7 @@ locale_setblanks (void)
 
   for (x = 0; x < sh_syntabsiz; x++)
     {
-      if (isblank ((unsigned char)x))
+      if (locale_isblank (x))
 	sh_syntaxtab[x] |= CSHBRK|CBLANK;
       else if (member (x, shell_break_chars))
 	{

po/hr.po

@@ -9,7 +9,7 @@ msgstr ""
 "Project-Id-Version: bash-5.2-rc1\n"
 "Report-Msgid-Bugs-To: \n"
 "POT-Creation-Date: 2022-01-11 14:50-0500\n"
-"PO-Revision-Date: 2023-05-01 16:11-0700\n"
+"PO-Revision-Date: 2023-05-23 19:54-0700\n"
 "Last-Translator: Božidar Putanec <bozidarp@yahoo.com>\n"
 "Language-Team: Croatian <lokalizacija@linux.hr>\n"
 "Language: hr\n"
@@ -4064,7 +4064,7 @@ msgstr ""
 "\n"
 "    Opcije:\n"
 "      -f   navedena IMENA se odnose samo na funkcije\n"
-"      -n   ukloni izvezeni atribut iz svakog IMENA \n"
+"      -n   ukloni izvezeni atribut iz svakog IMENA\n"
 "      -p   izlista popis svih izvezenih varijabli i funkcija\n"
 "\n"
 "    Argument „--“ spriječi daljnje procesiranje opcija.\n"