comparison src/syntax.c @ 32087:83ecc6bc4bc2

(prev_char_comstart_first): Remove. (back_comment): Check two-char comment markers more carefully to better handle overlapping cases like *//* or /* */* */ ... Match nestedness of ender/starter.

32086:73d18e25e303

  return val;
}

/* Return the SYNTAX_COMSTART_FIRST of the character before POS, POS_BYTE.  */

static int
prev_char_comstart_first (pos, pos_byte)
     int pos, pos_byte;
{
  int c, val;

  DEC_BOTH (pos, pos_byte);
  UPDATE_SYNTAX_TABLE_BACKWARD (pos);
  c = FETCH_CHAR (pos_byte);
  val = SYNTAX_COMSTART_FIRST (c);
  UPDATE_SYNTAX_TABLE_FORWARD (pos + 1);
  return val;
}

/* Checks whether charpos FROM is at the end of a comment.
   FROM_BYTE is the bytepos corresponding to FROM.
   Do not move back before STOP.

  int comment_lossage = 0;
  int comment_end = from;
  int comment_end_byte = from_byte;
  int comstart_pos = 0;
  int comstart_byte;
  int scanstart = from - 1;
  /* Place where the containing defun starts,
     or 0 if we didn't come across it yet.  */
  int defun_start = 0;
  int defun_start_byte = 0;
  register enum syntaxcode code;
  int nesting = 1;		/* current comment nesting */
  int c;

  /* At beginning of range to scan, we're outside of strings;
     that determines quote parity to the comment-end.  */
  while (from != stop)
    {
      int temp_byte;

      /* Move back and examine a character.  */
      DEC_BOTH (from, from_byte);
      UPDATE_SYNTAX_TABLE_BACKWARD (from);

      c = FETCH_CHAR (from_byte);
      code = SYNTAX (c);

      /* If this char is the second of a 2-char comment end sequence,
	 back up and give the pair the appropriate syntax.  */
      if (from > stop && SYNTAX_COMEND_SECOND (c)
	  && prev_char_comend_first (from, from_byte))
	{
	  code = Sendcomment;
	  DEC_BOTH (from, from_byte);
	  UPDATE_SYNTAX_TABLE_BACKWARD (from);
	  c = FETCH_CHAR (from_byte);
	}
			
      /* If this char starts a 2-char comment start sequence,
	 treat it like a 1-char comment starter.  */
      if (from < scanstart && SYNTAX_COMSTART_FIRST (c))
	{
	  temp_byte = inc_bytepos (from_byte);
	  UPDATE_SYNTAX_TABLE_FORWARD (from + 1);
	  if (SYNTAX_COMSTART_SECOND (FETCH_CHAR (temp_byte))
	      && comstyle == SYNTAX_COMMENT_STYLE (FETCH_CHAR (temp_byte)))
	    code = Scomment;
	  UPDATE_SYNTAX_TABLE_BACKWARD (from);
	}
      else if (code == Scomment && comstyle != SYNTAX_COMMENT_STYLE (c))
	/* Ignore comment starters of a different style.  */
	continue;

      /* Ignore escaped characters, except comment-enders.  */
      if (code != Sendcomment && char_quoted (from, from_byte))
	continue;

	       strings, else the comment-end itself would be inside a string. */
	    goto done;
	  break;

	case Sendcomment:
	  if (SYNTAX_COMMENT_STYLE (FETCH_CHAR (from_byte)) == comstyle)
	    /* This is the same style of comment ender as ours. */
	    {
	      if (comnested)
		nesting++;
	      else

32087:83ecc6bc4bc2

  return val;
}

/* Return the SYNTAX_COMSTART_FIRST of the character before POS, POS_BYTE.  */

/* static int
 * prev_char_comstart_first (pos, pos_byte)
 *      int pos, pos_byte;
 * {
 *   int c, val;
 * 
 *   DEC_BOTH (pos, pos_byte);
 *   UPDATE_SYNTAX_TABLE_BACKWARD (pos);
 *   c = FETCH_CHAR (pos_byte);
 *   val = SYNTAX_COMSTART_FIRST (c);
 *   UPDATE_SYNTAX_TABLE_FORWARD (pos + 1);
 *   return val;
 * } */

/* Checks whether charpos FROM is at the end of a comment.
   FROM_BYTE is the bytepos corresponding to FROM.
   Do not move back before STOP.

  int comment_lossage = 0;
  int comment_end = from;
  int comment_end_byte = from_byte;
  int comstart_pos = 0;
  int comstart_byte;
  /* Place where the containing defun starts,
     or 0 if we didn't come across it yet.  */
  int defun_start = 0;
  int defun_start_byte = 0;
  register enum syntaxcode code;
  int nesting = 1;		/* current comment nesting */
  int c;
  int syntax = 0;

  /* FIXME: A }} comment-ender style leads to incorrect behavior
     in the case of {{ c }}} because we ignore the last two chars which are
     assumed to be comment-enders although they aren't.  */

  /* At beginning of range to scan, we're outside of strings;
     that determines quote parity to the comment-end.  */
  while (from != stop)
    {
      int temp_byte, prev_syntax;
      int com2start, com2end;

      /* Move back and examine a character.  */
      DEC_BOTH (from, from_byte);
      UPDATE_SYNTAX_TABLE_BACKWARD (from);

      prev_syntax = syntax;
      c = FETCH_CHAR (from_byte);
      syntax = SYNTAX_WITH_FLAGS (c);
      code = SYNTAX (c);

      /* Check for 2-char comment markers.  */
      com2start = (SYNTAX_FLAGS_COMSTART_FIRST (syntax)
		   && SYNTAX_FLAGS_COMSTART_SECOND (prev_syntax)
		   && comstyle == SYNTAX_FLAGS_COMMENT_STYLE (prev_syntax)
		   && (SYNTAX_FLAGS_COMMENT_NESTED (prev_syntax)
		       || SYNTAX_FLAGS_COMMENT_NESTED (syntax)) == comnested);
      com2end = (SYNTAX_FLAGS_COMEND_FIRST (syntax)
		 && SYNTAX_FLAGS_COMEND_SECOND (prev_syntax));

      /* Nasty cases with overlapping 2-char comment markers:
	 - snmp-mode: -- c -- foo -- c --
	              --- c --
		      ------ c --
	 - c-mode:    *||*
		      |* *|* *|
		      |*| |* |*|
		      ///   */

      /* If a 2-char comment sequence partly overlaps with another,
	 we don't try to be clever.  */
      if (from > stop && (com2end || com2start))
	{
	  int next = from, next_byte = from_byte, next_c, next_syntax;
	  DEC_BOTH (next, next_byte);
	  UPDATE_SYNTAX_TABLE_BACKWARD (next);
	  next_c = FETCH_CHAR (next_byte);
	  next_syntax = SYNTAX_WITH_FLAGS (next_c);
	  if (((com2start || comnested)
	       && SYNTAX_FLAGS_COMEND_SECOND (syntax)
	       && SYNTAX_FLAGS_COMEND_FIRST (next_syntax))
	      || ((com2end || comnested)
		  && SYNTAX_FLAGS_COMSTART_SECOND (syntax)
		  && comstyle == SYNTAX_FLAGS_COMMENT_STYLE (syntax)
		  && SYNTAX_FLAGS_COMSTART_FIRST (next_syntax)))
	    goto lossage;
	  /* UPDATE_SYNTAX_TABLE_FORWARD (next + 1); */
	}

      if (com2start && comstart_pos == 0)
	/* We're looking at a comment starter.  But it might be a comment
	   ender as well (see snmp-mode).  The first time we see one, we
	   need to consider it as a comment starter,
	   and the subsequent times as a comment ender.  */
	com2end = 0;

      /* Turn a 2-char comment sequences into the appropriate syntax.  */
      if (com2end)
	code = Sendcomment;
      else if (com2start)
	code = Scomment;
      /* Ignore comment starters of a different style.  */
      else if (code == Scomment
	       && (comstyle != SYNTAX_FLAGS_COMMENT_STYLE (syntax)
		   || SYNTAX_FLAGS_COMMENT_NESTED (syntax) != comnested))
	continue;

      /* Ignore escaped characters, except comment-enders.  */
      if (code != Sendcomment && char_quoted (from, from_byte))
	continue;

	       strings, else the comment-end itself would be inside a string. */
	    goto done;
	  break;

	case Sendcomment:
	  if (SYNTAX_FLAGS_COMMENT_STYLE (syntax) == comstyle
	      && (SYNTAX_FLAGS_COMMENT_NESTED (prev_syntax)
		  || SYNTAX_FLAGS_COMMENT_NESTED (syntax)) == comnested)
	    /* This is the same style of comment ender as ours. */
	    {
	      if (comnested)
		nesting++;
	      else