1 files changed, 70 insertions, 67 deletions

diff --git a/src/coding.c b/src/coding.c
index b0a9f6ef4cb..e4b52f6db48 100644
--- a/src/coding.c
+++ b/src/coding.c
@@ -642,15 +642,6 @@ static enum coding_category coding_priorities[coding_category_max];
    Nth coding category.  */
 static struct coding_system coding_categories[coding_category_max];
 
-/*** Commonly used macros and functions ***/
-
-#ifndef min
-#define min(a, b) ((a) < (b) ? (a) : (b))
-#endif
-#ifndef max
-#define max(a, b) ((a) > (b) ? (a) : (b))
-#endif
-
 /* Encode a flag that can be nil, something else, or t as -1, 0, 1.  */
 
 static int
@@ -690,6 +681,14 @@ CHECK_NATNUM_CDR (Lisp_Object x)
   XSETCDR (x, tmp);
 }
 
+/* True if CODING's destination can be grown.  */
+
+static bool
+growable_destination (struct coding_system *coding)
+{
+  return STRINGP (coding->dst_object) || BUFFERP (coding->dst_object);
+}
+
 
 /* Safely get one byte from the source text pointed by SRC which ends
    at SRC_END, and set C to that byte.  If there are not enough bytes
@@ -1485,8 +1484,7 @@ decode_coding_utf_8 (struct coding_system *coding)
       src = src_base;
       consumed_chars = consumed_chars_base;
       ONE_MORE_BYTE (c);
-      *charbuf++ = ASCII_BYTE_P (c) ? c : BYTE8_TO_CHAR (c);
-      coding->errors++;
+      *charbuf++ = ASCII_CHAR_P (c) ? c : BYTE8_TO_CHAR (c);
     }
 
  no_more_source:
@@ -1685,7 +1683,6 @@ decode_coding_utf_16 (struct coding_system *coding)
 	  /* The first two bytes are not BOM.  Treat them as bytes
 	     for a normal character.  */
 	  src = src_base;
-	  coding->errors++;
 	}
       CODING_UTF_16_BOM (coding) = utf_without_bom;
     }
@@ -1725,7 +1722,7 @@ decode_coding_utf_16 (struct coding_system *coding)
 	ONE_MORE_BYTE (c2);
       if (c2 < 0)
 	{
-	  *charbuf++ = ASCII_BYTE_P (c1) ? c1 : BYTE8_TO_CHAR (c1);
+	  *charbuf++ = ASCII_CHAR_P (c1) ? c1 : BYTE8_TO_CHAR (c1);
 	  *charbuf++ = -c2;
 	  continue;
 	}
@@ -1742,7 +1739,6 @@ decode_coding_utf_16 (struct coding_system *coding)
 		c1 = surrogate & 0xFF, c2 = surrogate >> 8;
 	      *charbuf++ = c1;
 	      *charbuf++ = c2;
-	      coding->errors++;
 	      if (UTF_16_HIGH_SURROGATE_P (c))
 		CODING_UTF_16_SURROGATE (coding) = surrogate = c;
 	      else
@@ -2108,7 +2104,7 @@ emacs_mule_char (struct coding_system *coding, const unsigned char *src,
 
 	case 1:
 	  code = c;
-	  charset_ID = ASCII_BYTE_P (code) ? charset_ascii : charset_eight_bit;
+	  charset_ID = ASCII_CHAR_P (code) ? charset_ascii : charset_eight_bit;
 	  break;
 
 	default:
@@ -2596,9 +2592,8 @@ decode_coding_emacs_mule (struct coding_system *coding)
       src = src_base;
       consumed_chars = consumed_chars_base;
       ONE_MORE_BYTE (c);
-      *charbuf++ = ASCII_BYTE_P (c) ? c : BYTE8_TO_CHAR (c);
+      *charbuf++ = ASCII_CHAR_P (c) ? c : BYTE8_TO_CHAR (c);
       char_offset++;
-      coding->errors++;
     }
 
  no_more_source:
@@ -3591,7 +3586,7 @@ decode_coding_iso_2022 (struct coding_system *coding)
 
       if (CODING_ISO_EXTSEGMENT_LEN (coding) > 0)
 	{
-	  *charbuf++ = ASCII_BYTE_P (c1) ? c1 : BYTE8_TO_CHAR (c1);
+	  *charbuf++ = ASCII_CHAR_P (c1) ? c1 : BYTE8_TO_CHAR (c1);
 	  char_offset++;
 	  CODING_ISO_EXTSEGMENT_LEN (coding)--;
 	  continue;
@@ -3618,7 +3613,7 @@ decode_coding_iso_2022 (struct coding_system *coding)
 	    }
 	  else
 	    {
-	      *charbuf++ = ASCII_BYTE_P (c1) ? c1 : BYTE8_TO_CHAR (c1);
+	      *charbuf++ = ASCII_CHAR_P (c1) ? c1 : BYTE8_TO_CHAR (c1);
 	      char_offset++;
 	    }
 	  continue;
@@ -3992,7 +3987,7 @@ decode_coding_iso_2022 (struct coding_system *coding)
 	  MAYBE_FINISH_COMPOSITION ();
 	  for (; src_base < src; src_base++, char_offset++)
 	    {
-	      if (ASCII_BYTE_P (*src_base))
+	      if (ASCII_CHAR_P (*src_base))
 		*charbuf++ = *src_base;
 	      else
 		*charbuf++ = BYTE8_TO_CHAR (*src_base);
@@ -4022,9 +4017,8 @@ decode_coding_iso_2022 (struct coding_system *coding)
       src = src_base;
       consumed_chars = consumed_chars_base;
       ONE_MORE_BYTE (c);
-      *charbuf++ = c < 0 ? -c : ASCII_BYTE_P (c) ? c : BYTE8_TO_CHAR (c);
+      *charbuf++ = c < 0 ? -c : ASCII_CHAR_P (c) ? c : BYTE8_TO_CHAR (c);
       char_offset++;
-      coding->errors++;
       /* Reset the invocation and designation status to the safest
 	 one; i.e. designate ASCII to the graphic register 0, and
 	 invoke that register to the graphic plane 0.  This typically
@@ -4855,7 +4849,6 @@ decode_coding_sjis (struct coding_system *coding)
       ONE_MORE_BYTE (c);
       *charbuf++ = c < 0 ? -c : BYTE8_TO_CHAR (c);
       char_offset++;
-      coding->errors++;
     }
 
  no_more_source:
@@ -4951,7 +4944,6 @@ decode_coding_big5 (struct coding_system *coding)
       ONE_MORE_BYTE (c);
       *charbuf++ = c < 0 ? -c : BYTE8_TO_CHAR (c);
       char_offset++;
-      coding->errors++;
     }
 
  no_more_source:
@@ -5658,9 +5650,8 @@ decode_coding_charset (struct coding_system *coding)
       src = src_base;
       consumed_chars = consumed_chars_base;
       ONE_MORE_BYTE (c);
-      *charbuf++ = c < 0 ? -c : ASCII_BYTE_P (c) ? c : BYTE8_TO_CHAR (c);
+      *charbuf++ = c < 0 ? -c : ASCII_CHAR_P (c) ? c : BYTE8_TO_CHAR (c);
       char_offset++;
-      coding->errors++;
     }
 
  no_more_source:
@@ -6893,6 +6884,11 @@ decode_eol (struct coding_system *coding)
 }
 
 
+/* MAX_LOOKUP's maximum value.  MAX_LOOKUP is an int and so cannot
+   exceed INT_MAX.  Also, MAX_LOOKUP is multiplied by sizeof (int) for
+   alloca, so it cannot exceed MAX_ALLOCA / sizeof (int).  */
+enum { MAX_LOOKUP_MAX = min (INT_MAX, MAX_ALLOCA / sizeof (int)) };
+
 /* Return a translation table (or list of them) from coding system
    attribute vector ATTRS for encoding (if ENCODEP) or decoding (if
    not ENCODEP). */
@@ -6945,7 +6941,7 @@ get_translation_table (Lisp_Object attrs, bool encodep, int *max_lookup)
 	{
 	  val = XCHAR_TABLE (translation_table)->extras[1];
 	  if (NATNUMP (val) && *max_lookup < XFASTINT (val))
-	    *max_lookup = XFASTINT (val);
+	    *max_lookup = min (XFASTINT (val), MAX_LOOKUP_MAX);
 	}
       else if (CONSP (translation_table))
 	{
@@ -6957,7 +6953,7 @@ get_translation_table (Lisp_Object attrs, bool encodep, int *max_lookup)
 	      {
 		Lisp_Object tailval = XCHAR_TABLE (XCAR (tail))->extras[1];
 		if (NATNUMP (tailval) && *max_lookup < XFASTINT (tailval))
-		  *max_lookup = XFASTINT (tailval);
+		  *max_lookup = min (XFASTINT (tailval), MAX_LOOKUP_MAX);
 	      }
 	}
     }
@@ -7040,8 +7036,10 @@ produce_chars (struct coding_system *coding, Lisp_Object translation_table,
       int *buf = coding->charbuf;
       int *buf_end = buf + coding->charbuf_used;
 
-      if (EQ (coding->src_object, coding->dst_object))
+      if (EQ (coding->src_object, coding->dst_object)
+	  && ! NILP (coding->dst_object))
 	{
+	  eassert (growable_destination (coding));
 	  coding_set_source (coding);
 	  dst_end = ((unsigned char *) coding->source) + coding->consumed;
 	}
@@ -7080,6 +7078,7 @@ produce_chars (struct coding_system *coding, Lisp_Object translation_table,
 
 	      if ((dst_end - dst) / MAX_MULTIBYTE_LENGTH < to_nchars)
 		{
+		  eassert (growable_destination (coding));
 		  if (((min (PTRDIFF_MAX, SIZE_MAX) - (buf_end - buf))
 		       / MAX_MULTIBYTE_LENGTH)
 		      < to_nchars)
@@ -7124,7 +7123,10 @@ produce_chars (struct coding_system *coding, Lisp_Object translation_table,
       const unsigned char *src_end = src + coding->consumed;
 
       if (EQ (coding->dst_object, coding->src_object))
-	dst_end = (unsigned char *) src;
+	{
+	  eassert (growable_destination (coding));
+	  dst_end = (unsigned char *) src;
+	}
       if (coding->src_multibyte != coding->dst_multibyte)
 	{
 	  if (coding->src_multibyte)
@@ -7140,6 +7142,7 @@ produce_chars (struct coding_system *coding, Lisp_Object translation_table,
 		  ONE_MORE_BYTE (c);
 		  if (dst == dst_end)
 		    {
+		      eassert (growable_destination (coding));
 		      if (EQ (coding->src_object, coding->dst_object))
 			dst_end = (unsigned char *) src;
 		      if (dst == dst_end)
@@ -7170,6 +7173,7 @@ produce_chars (struct coding_system *coding, Lisp_Object translation_table,
 
 		if (dst >= dst_end - 1)
 		  {
+		    eassert (growable_destination (coding));
 		    if (EQ (coding->src_object, coding->dst_object))
 		      dst_end = (unsigned char *) src;
 		    if (dst >= dst_end - 1)
@@ -7283,16 +7287,20 @@ produce_charset (struct coding_system *coding, int *charbuf, ptrdiff_t pos)
 		      coding->dst_object);
 }
 
+#define MAX_CHARBUF_SIZE 0x4000
+/* How many units decoding functions expect in coding->charbuf at
+   most.  Currently, decode_coding_emacs_mule expects the following
+   size, and that is the largest value.  */
+#define MAX_CHARBUF_EXTRA_SIZE ((MAX_ANNOTATION_LENGTH * 3) + 1)
 
-#define CHARBUF_SIZE 0x4000
-
-#define ALLOC_CONVERSION_WORK_AREA(coding)				\
-  do {									\
-    coding->charbuf = SAFE_ALLOCA (CHARBUF_SIZE * sizeof (int));	\
-    coding->charbuf_size = CHARBUF_SIZE;				\
+#define ALLOC_CONVERSION_WORK_AREA(coding, size)		\
+  do {								\
+    ptrdiff_t units = min ((size) + MAX_CHARBUF_EXTRA_SIZE,	\
+			   MAX_CHARBUF_SIZE);			\
+    coding->charbuf = SAFE_ALLOCA (units * sizeof (int));	\
+    coding->charbuf_size = units;				\
   } while (0)
 
-
 static void
 produce_annotation (struct coding_system *coding, ptrdiff_t pos)
 {
@@ -7389,9 +7397,8 @@ decode_coding (struct coding_system *coding)
   coding->produced = coding->produced_char = 0;
   coding->chars_at_source = 0;
   record_conversion_result (coding, CODING_RESULT_SUCCESS);
-  coding->errors = 0;
 
-  ALLOC_CONVERSION_WORK_AREA (coding);
+  ALLOC_CONVERSION_WORK_AREA (coding, coding->src_bytes);
 
   attrs = CODING_ID_ATTRS (coding->id);
   translation_table = get_translation_table (attrs, 0, NULL);
@@ -7785,9 +7792,8 @@ encode_coding (struct coding_system *coding)
   coding->consumed = coding->consumed_char = 0;
   coding->produced = coding->produced_char = 0;
   record_conversion_result (coding, CODING_RESULT_SUCCESS);
-  coding->errors = 0;
 
-  ALLOC_CONVERSION_WORK_AREA (coding);
+  ALLOC_CONVERSION_WORK_AREA (coding, coding->src_chars);
 
   if (coding->encoder == encode_coding_ccl)
     {
@@ -8461,11 +8467,11 @@ from_unicode (Lisp_Object str)
 }
 
 Lisp_Object
-from_unicode_buffer (const wchar_t* wstr)
+from_unicode_buffer (const wchar_t *wstr)
 {
     return from_unicode (
         make_unibyte_string (
-            (char*) wstr,
+            (char *) wstr,
             /* we get one of the two final 0 bytes for free. */
             1 + sizeof (wchar_t) * wcslen (wstr)));
 }
@@ -9049,13 +9055,13 @@ DEFUN ("find-coding-systems-region-internal",
     p = pbeg = BYTE_POS_ADDR (start_byte);
   pend = p + (end_byte - start_byte);
 
-  while (p < pend && ASCII_BYTE_P (*p)) p++;
-  while (p < pend && ASCII_BYTE_P (*(pend - 1))) pend--;
+  while (p < pend && ASCII_CHAR_P (*p)) p++;
+  while (p < pend && ASCII_CHAR_P (*(pend - 1))) pend--;
 
   work_table = Fmake_char_table (Qnil, Qnil);
   while (p < pend)
     {
-      if (ASCII_BYTE_P (*p))
+      if (ASCII_CHAR_P (*p))
 	p++;
       else
 	{
@@ -9109,8 +9115,7 @@ DEFUN ("find-coding-systems-region-internal",
 
 DEFUN ("unencodable-char-position", Funencodable_char_position,
        Sunencodable_char_position, 3, 5, 0,
-       doc: /*
-Return position of first un-encodable character in a region.
+       doc: /* Return position of first un-encodable character in a region.
 START and END specify the region and CODING-SYSTEM specifies the
 encoding to check.  Return nil if CODING-SYSTEM does encode the region.
 
@@ -9120,8 +9125,9 @@ list of positions.
 
 If optional 5th argument STRING is non-nil, it is a string to search
 for un-encodable characters.  In that case, START and END are indexes
-to the string.  */)
-  (Lisp_Object start, Lisp_Object end, Lisp_Object coding_system, Lisp_Object count, Lisp_Object string)
+to the string and treated as in `substring'.  */)
+  (Lisp_Object start, Lisp_Object end, Lisp_Object coding_system,
+   Lisp_Object count, Lisp_Object string)
 {
   EMACS_INT n;
   struct coding_system coding;
@@ -9158,12 +9164,7 @@ to the string.  */)
   else
     {
       CHECK_STRING (string);
-      CHECK_NATNUM (start);
-      CHECK_NATNUM (end);
-      if (! (XINT (start) <= XINT (end) && XINT (end) <= SCHARS (string)))
-	args_out_of_range_3 (string, start, end);
-      from = XINT (start);
-      to = XINT (end);
+      validate_subarray (string, start, end, SCHARS (string), &from, &to);
       if (! STRING_MULTIBYTE (string))
 	return Qnil;
       p = SDATA (string) + string_char_to_byte (string, from);
@@ -9187,7 +9188,7 @@ to the string.  */)
       int c;
 
       if (ascii_compatible)
-	while (p < stop && ASCII_BYTE_P (*p))
+	while (p < stop && ASCII_CHAR_P (*p))
 	  p++, from++;
       if (p >= stop)
 	{
@@ -9303,12 +9304,12 @@ is nil.  */)
     p = pbeg = BYTE_POS_ADDR (start_byte);
   pend = p + (end_byte - start_byte);
 
-  while (p < pend && ASCII_BYTE_P (*p)) p++, pos++;
-  while (p < pend && ASCII_BYTE_P (*(pend - 1))) pend--;
+  while (p < pend && ASCII_CHAR_P (*p)) p++, pos++;
+  while (p < pend && ASCII_CHAR_P (*(pend - 1))) pend--;
 
   while (p < pend)
     {
-      if (ASCII_BYTE_P (*p))
+      if (ASCII_CHAR_P (*p))
 	p++;
       else
 	{
@@ -9616,7 +9617,7 @@ Return the corresponding character.  */)
   CHECK_CODING_SYSTEM_GET_SPEC (Vsjis_coding_system, spec);
   attrs = AREF (spec, 0);
 
-  if (ASCII_BYTE_P (ch)
+  if (ASCII_CHAR_P (ch)
       && ! NILP (CODING_ATTR_ASCII_COMPAT (attrs)))
     return code;
 
@@ -9697,7 +9698,7 @@ Return the corresponding character.  */)
   CHECK_CODING_SYSTEM_GET_SPEC (Vbig5_coding_system, spec);
   attrs = AREF (spec, 0);
 
-  if (ASCII_BYTE_P (ch)
+  if (ASCII_CHAR_P (ch)
       && ! NILP (CODING_ATTR_ASCII_COMPAT (attrs)))
     return code;
 
@@ -9758,7 +9759,7 @@ DEFUN ("set-terminal-coding-system-internal", Fset_terminal_coding_system_intern
        doc: /* Internal use only.  */)
   (Lisp_Object coding_system, Lisp_Object terminal)
 {
-  struct terminal *term = get_terminal (terminal, 1);
+  struct terminal *term = decode_live_terminal (terminal);
   struct coding_system *terminal_coding = TERMINAL_TERMINAL_CODING (term);
   CHECK_SYMBOL (coding_system);
   setup_coding_system (Fcheck_coding_system (coding_system), terminal_coding);
@@ -9799,7 +9800,7 @@ frame's terminal device.  */)
   (Lisp_Object terminal)
 {
   struct coding_system *terminal_coding
-    = TERMINAL_TERMINAL_CODING (get_terminal (terminal, 1));
+    = TERMINAL_TERMINAL_CODING (decode_live_terminal (terminal));
   Lisp_Object coding_system = CODING_ID_NAME (terminal_coding->id);
 
   /* For backward compatibility, return nil if it is `undecided'.  */
@@ -9811,7 +9812,7 @@ DEFUN ("set-keyboard-coding-system-internal", Fset_keyboard_coding_system_intern
        doc: /* Internal use only.  */)
   (Lisp_Object coding_system, Lisp_Object terminal)
 {
-  struct terminal *t = get_terminal (terminal, 1);
+  struct terminal *t = decode_live_terminal (terminal);
   CHECK_SYMBOL (coding_system);
   if (NILP (coding_system))
     coding_system = Qno_conversion;
@@ -9830,7 +9831,7 @@ DEFUN ("keyboard-coding-system",
   (Lisp_Object terminal)
 {
   return CODING_ID_NAME (TERMINAL_KEYBOARD_CODING
-			 (get_terminal (terminal, 1))->id);
+			 (decode_live_terminal (terminal))->id);
 }
 
 
@@ -10040,7 +10041,8 @@ make_subsidiaries (Lisp_Object base)
 {
   Lisp_Object subsidiaries;
   ptrdiff_t base_name_len = SBYTES (SYMBOL_NAME (base));
-  char *buf = alloca (base_name_len + 6);
+  USE_SAFE_ALLOCA;
+  char *buf = SAFE_ALLOCA (base_name_len + 6);
   int i;
 
   memcpy (buf, SDATA (SYMBOL_NAME (base)), base_name_len);
@@ -10050,6 +10052,7 @@ make_subsidiaries (Lisp_Object base)
       strcpy (buf + base_name_len, suffixes[i]);
       ASET (subsidiaries, i, intern (buf));
     }
+  SAFE_FREE ();
   return subsidiaries;
 }