Quote multibyte characters correctly.

(ISGRAPH): Remove.
(ISPRINT): New macro.
(<wchar.h>): Include if HAVE_MBRTOWC && HAVE_WCHAR_H.
(isprint, mbrtowc, mbsinit, mbstate_t): New macros,
defined if ! (HAVE_MBRTOWC && HAVE_WCHAR_H).
(quotearg_buffer_restyled): New function, with most of the old
quotearg_buffer's contents.
Major rewrite to support multibyte characters.
(quotearg_buffer): Now just calls quotearg_buffer_restyled.

1 files changed, 250 insertions, 113 deletions

diff --git a/lib/quotearg.c b/lib/quotearg.c
index 06f5fea84..9eb05161c 100644
--- a/lib/quotearg.c
+++ b/lib/quotearg.c
@@ -1,5 +1,5 @@
 /* quotearg.c - quote arguments for output
-   Copyright (C) 1998, 1999 Free Software Foundation, Inc.
+   Copyright (C) 1998, 1999, 2000 Free Software Foundation, Inc.
 
    This program is free software; you can redistribute it and/or modify
    it under the terms of the GNU General Public License as published by
@@ -17,8 +17,6 @@
 
 /* Written by Paul Eggert <eggert@twinsun.com> */
 
-/* FIXME: Multibyte characters are not supported yet.  */
-
 #if HAVE_CONFIG_H
 # include <config.h>
 #endif
@@ -33,11 +31,7 @@
 #else
 # define ISASCII(c) isascii (c)
 #endif
-#ifdef isgraph
-# define ISGRAPH(c) (ISASCII (c) && isgraph (c))
-#else
-# define ISGRAPH(c) (ISASCII (c) && isprint (c) && !isspace (c))
-#endif
+#define ISPRINT(c) (ISASCII (c) && isprint (c))
 
 #if ENABLE_NLS
 # include <libintl.h>
@@ -64,6 +58,15 @@
 # include <string.h>
 #endif
 
+#if HAVE_MBRTOWC && HAVE_WCHAR_H
+# include <wchar.h>
+#else
+# define iswprint(wc) 1
+# define mbrtowc(pwc, s, n, ps) 1
+# define mbsinit(ps) 1
+# define mbstate_t int
+#endif
+
 #define INT_BITS (sizeof (int) * CHAR_BIT)
 
 struct quoting_options
@@ -71,7 +74,7 @@ struct quoting_options
   /* Basic quoting style.  */
   enum quoting_style style;
 
-  /* Quote the chararacters indicated by this bit vector even if the
+  /* Quote the characters indicated by this bit vector even if the
      quoting style would not normally require them to be quoted.  */
   int quote_these_too[((UCHAR_MAX + 1) / INT_BITS
 		       + ((UCHAR_MAX + 1) % INT_BITS != 0))];
@@ -89,7 +92,7 @@ char const *const quoting_style_args[] =
   0
 };
 
-/* Correspondances to quoting style names.  */
+/* Correspondences to quoting style names.  */
 enum quoting_style const quoting_style_vals[] =
 {
   literal_quoting_style,
@@ -147,161 +150,274 @@ set_char_quoting (struct quoting_options *o, char c, int i)
 }
 
 /* Place into buffer BUFFER (of size BUFFERSIZE) a quoted version of
-   argument ARG (of size ARGSIZE), using O to control quoting.
-   If O is null, use the default.
+   argument ARG (of size ARGSIZE), using QUOTING_STYLE and the
+   non-quoting-style part of O to control quoting.
    Terminate the output with a null character, and return the written
    size of the output, not counting the terminating null.
    If BUFFERSIZE is too small to store the output string, return the
    value that would have been returned had BUFFERSIZE been large enough.
-   If ARGSIZE is -1, use the string length of the argument for ARGSIZE.  */
-size_t
-quotearg_buffer (char *buffer, size_t buffersize,
-		 char const *arg, size_t argsize,
-		 struct quoting_options const *o)
+   If ARGSIZE is -1, use the string length of the argument for ARGSIZE.
+
+   This function acts like quotearg_buffer (BUFFER, BUFFERSIZE, ARG,
+   ARGSIZE, O), except it uses QUOTING_STYLE instead of the quoting
+   style specified by O, and O may not be null.  */
+
+static size_t
+quotearg_buffer_restyled (char *buffer, size_t buffersize,
+			  char const *arg, size_t argsize,
+			  enum quoting_style quoting_style,
+			  struct quoting_options const *o)
 {
-  unsigned char c;
   size_t i;
   size_t len = 0;
-  char const *quote_string;
-  size_t quote_string_len;
-  struct quoting_options const *p = o ? o : &default_quoting_options;
-  enum quoting_style quoting_style = p->style;
+  char const *quote_string = 0;
+  size_t quote_string_len = 0;
+  int backslash_escapes = 0;
+
 #define STORE(c) \
     do \
       { \
 	if (len < buffersize) \
 	  buffer[len] = (c); \
-	  len++; \
+	len++; \
       } \
     while (0)
 
   switch (quoting_style)
     {
-    case shell_quoting_style:
-      if (! (argsize == (size_t) -1 ? arg[0] == '\0' : argsize == 0))
-	{
-	  switch (arg[0])
-	    {
-	    case '#': case '~':
-	      break;
-
-	    default:
-	      for (i = 0; ; i++)
-		{
-		  if (argsize == (size_t) -1 ? arg[i] == '\0' : i == argsize)
-		    goto done;
-
-		  c = arg[i];
-
-		  switch (c)
-		    {
-		    case '\t': case '\n': case ' ':
-		    case '!': /* special in csh */
-		    case '"': case '$': case '&': case '\'':
-		    case '(': case ')': case '*': case ';':
-		    case '<': case '>': case '?': case '[': case '\\':
-		    case '^': /* special in old /bin/sh, e.g. SunOS 4.1.4 */
-		    case '`': case '|':
-		      goto needs_quoting;
-		    }
-
-		  if (p->quote_these_too[c / INT_BITS] & (1 << (c % INT_BITS)))
-		    goto needs_quoting;
-
-		  STORE (c);
-		}
-	    needs_quoting:;
-
-	      len = 0;
-	      break;
-	    }
-	}
-      /* Fall through.  */
-
-    case shell_always_quoting_style:
-      STORE ('\'');
-      quote_string = "'";
-      quote_string_len = 1;
-      break;
-
     case c_quoting_style:
       STORE ('"');
+      backslash_escapes = 1;
       quote_string = "\"";
       quote_string_len = 1;
       break;
 
+    case escape_quoting_style:
+      backslash_escapes = 1;
+      break;
+
     case locale_quoting_style:
       for (quote_string = _("`"); *quote_string; quote_string++)
 	STORE (*quote_string);
+      backslash_escapes = 1;
       quote_string = _("'");
       quote_string_len = strlen (quote_string);
       break;
 
+    case shell_always_quoting_style:
+      STORE ('\'');
+      quote_string = "'";
+      quote_string_len = 1;
+      break;
+
     default:
-      quote_string = 0;
-      quote_string_len = 0;
       break;
     }
 
   for (i = 0;  ! (argsize == (size_t) -1 ? arg[i] == '\0' : i == argsize);  i++)
     {
-      c = arg[i];
+      unsigned char c;
+      unsigned char esc;
 
-      switch (quoting_style)
+      if (backslash_escapes
+	  && quote_string_len
+	  && i + quote_string_len <= argsize
+	  && memcmp (arg + i, quote_string, quote_string_len) == 0)
+	STORE ('\\');
+
+      c = arg[i];
+      switch (c)
 	{
-	case literal_quoting_style:
+	case '?':
+	  switch (quoting_style)
+	    {
+	    case shell_quoting_style:
+	      goto use_shell_always_quoting_style;
+
+	    case c_quoting_style:
+	      if (i + 2 < argsize && arg[i + 1] == '?')
+		switch (arg[i + 2])
+		  {
+		  case '!': case '\'':
+		  case '(': case ')': case '-': case '/':
+		  case '<': case '=': case '>':
+		    /* Escape the second '?' in what would otherwise be
+		       a trigraph.  */
+		    i += 2;
+		    c = arg[i + 2];
+		    STORE ('?');
+		    STORE ('\\');
+		    STORE ('?');
+		    break;
+		  }
+	      break;
+
+	    default:
+	      break;
+	    }
 	  break;
 
-	case shell_quoting_style:
-	case shell_always_quoting_style:
-	  if (c == '\'')
+#if HAVE_C_BACKSLASH_A
+	case '\a': esc = 'a'; goto c_escape;
+#endif
+	case '\b': esc = 'b'; goto c_escape;
+	case '\f': esc = 'f'; goto c_escape;
+	case '\n': esc = 'n'; goto c_escape;
+	case '\r': esc = 'r'; goto c_escape;
+	case '\t': esc = 't'; goto c_escape;
+	case '\v': esc = 'v'; goto c_escape;
+	case '\\': esc = c; goto c_escape;
+
+	c_escape:
+	  if (backslash_escapes)
 	    {
-	      STORE ('\'');
-	      STORE ('\\');
-	      STORE ('\'');
+	      c = esc;
+	      goto store_escape;
 	    }
+	  if (quoting_style == shell_quoting_style)
+	    goto use_shell_always_quoting_style;
 	  break;
 
-	case c_quoting_style:
-	case escape_quoting_style:
-	case locale_quoting_style:
-	  switch (c)
+	case '#': case '~':
+	  if (i != 0)
+	    break;
+	  /* Fall through.  */
+	case ' ':
+	case '!': /* special in bash */
+	case '"': case '$': case '&':
+	case '(': case ')': case '*': case ';':
+	case '<': case '>': case '[':
+	case '^': /* special in old /bin/sh, e.g. SunOS 4.1.4 */
+	case '`': case '|':
+	  /* A shell special character.  In theory, '$' and '`' could
+	     be the first bytes of multibyte characters, which means
+	     we should check them with mbrtowc, but in practice this
+	     doesn't happen so it's not worth worrying about.  */
+	  if (quoting_style == shell_quoting_style)
+	    goto use_shell_always_quoting_style;
+	  break;
+
+	case '\'':
+	  switch (quoting_style)
 	    {
-	    case '?': /* Do not generate trigraphs.  */
-	    case '\\': goto store_escape;
-	      /* Not all C compilers know what \a means.  */
-	    case   7 : c = 'a'; goto store_escape;
-	    case '\b': c = 'b'; goto store_escape;
-	    case '\f': c = 'f'; goto store_escape;
-	    case '\n': c = 'n'; goto store_escape;
-	    case '\r': c = 'r'; goto store_escape;
-	    case '\t': c = 't'; goto store_escape;
-	    case '\v': c = 'v'; goto store_escape;
-
-	    case ' ': break;
+	    case shell_quoting_style:
+	      goto use_shell_always_quoting_style;
+
+	    case shell_always_quoting_style:
+	      STORE ('\'');
+	      STORE ('\\');
+	      STORE ('\'');
+	      break;
 
 	    default:
-	      if (quote_string_len
-		  && strncmp (arg + i, quote_string, quote_string_len) == 0)
-		goto store_escape;
-	      if (!ISGRAPH (c))
-		{
-		  STORE ('\\');
-		  STORE ('0' + (c >> 6));
-		  STORE ('0' + ((c >> 3) & 7));
-		  c = '0' + (c & 7);
-		  goto store_c;
-		}
 	      break;
 	    }
+	  break;
 
-	  if (! (p->quote_these_too[c / INT_BITS] & (1 << (c % INT_BITS))))
-	    goto store_c;
+	case '%': case '+': case ',': case '-': case '.': case '/':
+	case '0': case '1': case '2': case '3': case '4': case '5':
+	case '6': case '7': case '8': case '9': case ':': case '=':
+	case 'A': case 'B': case 'C': case 'D': case 'E': case 'F':
+	case 'G': case 'H': case 'I': case 'J': case 'K': case 'L':
+	case 'M': case 'N': case 'O': case 'P': case 'Q': case 'R':
+	case 'S': case 'T': case 'U': case 'V': case 'W': case 'X':
+	case 'Y': case 'Z': case ']': case '_': case 'a': case 'b':
+	case 'c': case 'd': case 'e': case 'f': case 'g': case 'h':
+	case 'i': case 'j': case 'k': case 'l': case 'm': case 'n':
+	case 'o': case 'p': case 'q': case 'r': case 's': case 't':
+	case 'u': case 'v': case 'w': case 'x': case 'y': case 'z':
+	case '{': case '}':
+	  /* These characters don't cause problems, no matter what the
+	     quoting style is.  They cannot start multibyte sequences.  */
+	  break;
 
-	store_escape:
-	  STORE ('\\');
+	default:
+	  /* If we have a multibyte sequence, copy it until we reach
+	     its end, find an error, or come back to the initial shift
+	     state.  For C-like styles, if the sequence has
+	     unprintable characters, escape the whole sequence, since
+	     we can't easily escape single characters within it.  */
+	  {
+	    /* Length of multibyte sequence found so far.  */
+	    size_t m = 0;
+
+	    int printable = 1;
+	    mbstate_t mbstate;
+	    memset (&mbstate, 0, sizeof mbstate);
+
+	    if (argsize == (size_t) -1)
+	      argsize = strlen (arg);
+
+	    do
+	      {
+		wchar_t w;
+		size_t bytes = mbrtowc (&w, &arg[i + m],
+					argsize - (i + m), &mbstate);
+		if (bytes == 0)
+		  break;
+		else if (bytes == (size_t) -1)
+		  {
+		    printable = 0;
+		    break;
+		  }
+		else if (bytes == (size_t) -2)
+		  {
+		    printable = 0;
+		    while (i + m < argsize && arg[i + m])
+		      m++;
+		    break;
+		  }
+		else
+		  {
+		    if (! iswprint (w))
+		      printable = 0;
+		    m += bytes;
+		  }
+	      }
+	    while (! mbsinit (&mbstate));
+
+	    if (m <= 1)
+	      {
+		/* Escape a unibyte character like a multibyte
+		   sequence if using backslash escapes, and if the
+		   character is not printable.  */
+		m = backslash_escapes && ! ISPRINT (c);
+		printable = 0;
+	      }
+
+	    if (m)
+	      {
+		/* Output a multibyte sequence, or an escaped
+		   unprintable unibyte character.  */
+		size_t imax = i + m - 1;
+
+		for (;;)
+		  {
+		    if (backslash_escapes && ! printable)
+		      {
+			STORE ('\\');
+			STORE ('0' + (c >> 6));
+			STORE ('0' + ((c >> 3) & 7));
+			c = '0' + (c & 7);
+		      }
+		    if (i == imax)
+		      break;
+		    STORE (c);
+		    c = arg[++i];
+		  }
+
+		goto store_c;
+	      }
+	  }
 	}
 
+      if (! (backslash_escapes
+	     && o->quote_these_too[c / INT_BITS] & (1 << (c % INT_BITS))))
+	goto store_c;
+
+    store_escape:
+      STORE ('\\');
+
     store_c:
       STORE (c);
     }
@@ -310,10 +426,31 @@ quotearg_buffer (char *buffer, size_t buffersize,
     for (; *quote_string; quote_string++)
       STORE (*quote_string);
 
- done:
   if (len < buffersize)
     buffer[len] = '\0';
   return len;
+
+ use_shell_always_quoting_style:
+  return quotearg_buffer_restyled (buffer, buffersize, arg, argsize,
+				   shell_always_quoting_style, o);
+}
+
+/* Place into buffer BUFFER (of size BUFFERSIZE) a quoted version of
+   argument ARG (of size ARGSIZE), using O to control quoting.
+   If O is null, use the default.
+   Terminate the output with a null character, and return the written
+   size of the output, not counting the terminating null.
+   If BUFFERSIZE is too small to store the output string, return the
+   value that would have been returned had BUFFERSIZE been large enough.
+   If ARGSIZE is -1, use the string length of the argument for ARGSIZE.  */
+size_t
+quotearg_buffer (char *buffer, size_t buffersize,
+		 char const *arg, size_t argsize,
+		 struct quoting_options const *o)
+{
+  struct quoting_options const *p = o ? o : &default_quoting_options;
+  return quotearg_buffer_restyled (buffer, buffersize, arg, argsize,
+				   p->style, p);
 }
 
 /* Use storage slot N to return a quoted version of the string ARG.