app.c

来自「基于4个mips核的noc设计」· C语言 代码 · 共 1,329 行 · 第 1/3 页

	  if (ch2 == '*')
	    {
	      for (;;)
		{
		  do
		    {
		      ch2 = GET ();
		      if (ch2 != EOF && IS_NEWLINE (ch2))
			add_newlines++;
		    }
		  while (ch2 != EOF && ch2 != '*');

		  while (ch2 == '*')
		    ch2 = GET ();

		  if (ch2 == EOF || ch2 == '/')
		    break;

		  /* This UNGET will ensure that we count newlines
                     correctly.  */
		  UNGET (ch2);
		}

	      if (ch2 == EOF)
		as_warn (_("end of file in multiline comment"));

	      ch = ' ';
	      goto recycle;
	    }
#ifdef DOUBLESLASH_LINE_COMMENTS
	  else if (ch2 == '/')
	    {
	      do
		{
		  ch = GET ();
		}
	      while (ch != EOF && !IS_NEWLINE (ch));
	      if (ch == EOF)
		as_warn ("end of file in comment; newline inserted");
	      state = 0;
	      PUT ('\n');
	      break;
	    }
#endif
	  else
	    {
	      if (ch2 != EOF)
		UNGET (ch2);
	      if (state == 9 || state == 10)
		state = 3;
	      PUT (ch);
	    }
	  break;

	case LEX_IS_STRINGQUOTE:
	  if (state == 10)
	    {
	      /* Preserve the whitespace in foo "bar" */
	      UNGET (ch);
	      state = 3;
	      PUT (' ');

	      /* PUT didn't jump out.  We could just break, but we
                 know what will happen, so optimize a bit.  */
	      ch = GET ();
	      old_state = 3;
	    }
	  else if (state == 9)
	    old_state = 3;
	  else
	    old_state = state;
	  state = 5;
	  PUT (ch);
	  break;

#ifndef IEEE_STYLE
	case LEX_IS_ONECHAR_QUOTE:
	  if (state == 10)
	    {
	      /* Preserve the whitespace in foo 'b' */
	      UNGET (ch);
	      state = 3;
	      PUT (' ');
	      break;
	    }
	  ch = GET ();
	  if (ch == EOF)
	    {
	      as_warn (_("end of file after a one-character quote; \\0 inserted"));
	      ch = 0;
	    }
	  if (ch == '\\')
	    {
	      ch = GET ();
	      if (ch == EOF)
		{
		  as_warn (_("end of file in escape character"));
		  ch = '\\';
		}
	      else
		ch = process_escape (ch);
	    }
	  sprintf (out_buf, "%d", (int) (unsigned char) ch);

	  /* None of these 'x constants for us.  We want 'x'.  */
	  if ((ch = GET ()) != '\'')
	    {
#ifdef REQUIRE_CHAR_CLOSE_QUOTE
	      as_warn (_("Missing close quote: (assumed)"));
#else
	      if (ch != EOF)
		UNGET (ch);
#endif
	    }
	  if (strlen (out_buf) == 1)
	    {
	      PUT (out_buf[0]);
	      break;
	    }
	  if (state == 9)
	    old_state = 3;
	  else
	    old_state = state;
	  state = -1;
	  out_string = out_buf;
	  PUT (*out_string++);
	  break;
#endif

	case LEX_IS_COLON:
#ifdef KEEP_WHITE_AROUND_COLON
	  state = 9;
#else
	  if (state == 9 || state == 10)
	    state = 3;
	  else if (state != 3)
	    state = 1;
#endif
	  PUT (ch);
	  break;

	case LEX_IS_NEWLINE:
	  /* Roll out a bunch of newlines from inside comments, etc.  */
	  if (add_newlines)
	    {
	      --add_newlines;
	      UNGET (ch);
	    }
	  /* Fall through.  */

	case LEX_IS_LINE_SEPARATOR:
	  state = 0;
	  PUT (ch);
	  break;

#ifdef TC_V850
	case LEX_IS_DOUBLEDASH_1ST:
	  ch2 = GET ();
	  if (ch2 != '-')
	    {
	      UNGET (ch2);
	      goto de_fault;
	    }
	  /* Read and skip to end of line.  */
	  do
	    {
	      ch = GET ();
	    }
	  while (ch != EOF && ch != '\n');
	  if (ch == EOF)
	    {
	      as_warn (_("end of file in comment; newline inserted"));
	    }
	  state = 0;
	  PUT ('\n');
	  break;
#endif
#ifdef DOUBLEBAR_PARALLEL
	case LEX_IS_DOUBLEBAR_1ST:
	  ch2 = GET ();
	  if (ch2 != '|')
	    {
	      UNGET (ch2);
	      goto de_fault;
	    }
	  /* Reset back to state 1 and pretend that we are parsing a line from
	     just after the first white space.  */
	  state = 1;
	  PUT ('|');
	  PUT ('|');
	  break;
#endif
	case LEX_IS_LINE_COMMENT_START:
	  /* FIXME-someday: The two character comment stuff was badly
	     thought out.  On i386, we want '/' as line comment start
	     AND we want C style comments.  hence this hack.  The
	     whole lexical process should be reworked.  xoxorich.  */
	  if (ch == '/')
	    {
	      ch2 = GET ();
	      if (ch2 == '*')
		{
		  old_state = 3;
		  state = -2;
		  break;
		}
	      else
		{
		  UNGET (ch2);
		}
	    } /* bad hack */

	  if (state == 0 || state == 1)	/* Only comment at start of line.  */
	    {
	      int startch;

	      startch = ch;

	      do
		{
		  ch = GET ();
		}
	      while (ch != EOF && IS_WHITESPACE (ch));
	      if (ch == EOF)
		{
		  as_warn (_("end of file in comment; newline inserted"));
		  PUT ('\n');
		  break;
		}
	      if (ch < '0' || ch > '9' || state != 0 || startch != '#')
		{
		  /* Not a cpp line.  */
		  while (ch != EOF && !IS_NEWLINE (ch))
		    ch = GET ();
		  if (ch == EOF)
		    as_warn (_("EOF in Comment: Newline inserted"));
		  state = 0;
		  PUT ('\n');
		  break;
		}
	      /* Looks like `# 123 "filename"' from cpp.  */
	      UNGET (ch);
	      old_state = 4;
	      state = -1;
	      if (scrub_m68k_mri)
		out_string = "\tappline ";
	      else
		out_string = "\t.appline ";
	      PUT (*out_string++);
	      break;
	    }

#ifdef TC_D10V
	  /* All insns end in a char for which LEX_IS_SYMBOL_COMPONENT is true.
	     Trap is the only short insn that has a first operand that is
	     neither register nor label.
	     We must prevent exef0f ||trap #1 to degenerate to exef0f ||trap#1 .
	     We can't make '#' LEX_IS_SYMBOL_COMPONENT because it is
	     already LEX_IS_LINE_COMMENT_START.  However, it is the
	     only character in line_comment_chars for d10v, hence we
	     can recognize it as such.  */
	  /* An alternative approach would be to reset the state to 1 when
	     we see '||', '<'- or '->', but that seems to be overkill.  */
	  if (state == 10)
	    PUT (' ');
#endif
	  /* We have a line comment character which is not at the
	     start of a line.  If this is also a normal comment
	     character, fall through.  Otherwise treat it as a default
	     character.  */
	  if (strchr (tc_comment_chars, ch) == NULL
	      && (! scrub_m68k_mri
		  || (ch != '!' && ch != '*')))
	    goto de_fault;
	  if (scrub_m68k_mri
	      && (ch == '!' || ch == '*' || ch == '#')
	      && state != 1
	      && state != 10)
	    goto de_fault;
	  /* Fall through.  */
	case LEX_IS_COMMENT_START:
#if defined TC_ARM && defined OBJ_ELF
	  /* On the ARM, `@' is the comment character.
	     Unfortunately this is also a special character in ELF .symver
	     directives (and .type, though we deal with those another way).
	     So we check if this line is such a directive, and treat
	     the character as default if so.  This is a hack.  */
	  if ((symver_state != NULL) && (*symver_state == 0))
	    goto de_fault;
#endif
#ifdef WARN_COMMENTS
	  if (!found_comment)
	    as_where (&found_comment_file, &found_comment);
#endif
	  do
	    {
	      ch = GET ();
	    }
	  while (ch != EOF && !IS_NEWLINE (ch));
	  if (ch == EOF)
	    as_warn (_("end of file in comment; newline inserted"));
	  state = 0;
	  PUT ('\n');
	  break;

	case LEX_IS_SYMBOL_COMPONENT:
	  if (state == 10)
	    {
	      /* This is a symbol character following another symbol
		 character, with whitespace in between.  We skipped
		 the whitespace earlier, so output it now.  */
	      UNGET (ch);
	      state = 3;
	      PUT (' ');
	      break;
	    }

	  if (state == 3)
	    state = 9;

	  /* This is a common case.  Quickly copy CH and all the
             following symbol component or normal characters.  */
	  if (to + 1 < toend
	      && mri_state == NULL
#if defined TC_ARM && defined OBJ_ELF
	      && symver_state == NULL
#endif
	      )
	    {
	      char *s;
	      int len;

	      for (s = from; s < fromend; s++)
		{
		  int type;

		  ch2 = *(unsigned char *) s;
		  type = lex[ch2];
		  if (type != 0
		      && type != LEX_IS_SYMBOL_COMPONENT)
		    break;
		}
	      if (s > from)
		{
		  /* Handle the last character normally, for
                     simplicity.  */
		  --s;
		}
	      len = s - from;
	      if (len > (toend - to) - 1)
		len = (toend - to) - 1;
	      if (len > 0)
		{
		  PUT (ch);
		  if (len > 8)
		    {
		      memcpy (to, from, len);
		      to += len;
		      from += len;
		    }
		  else
		    {
		      switch (len)
			{
			case 8: *to++ = *from++;
			case 7: *to++ = *from++;
			case 6: *to++ = *from++;
			case 5: *to++ = *from++;
			case 4: *to++ = *from++;
			case 3: *to++ = *from++;
			case 2: *to++ = *from++;
			case 1: *to++ = *from++;
			}
		    }
		  ch = GET ();
		}
	    }

	  /* Fall through.  */
	default:
	de_fault:
	  /* Some relatively `normal' character.  */
	  if (state == 0)
	    {
	      state = 11;	/* Now seeing label definition */
	    }
	  else if (state == 1)
	    {
	      state = 2;	/* Ditto */
	    }
	  else if (state == 9)
	    {
	      if (lex[ch] != LEX_IS_SYMBOL_COMPONENT)
		state = 3;
	    }
	  else if (state == 10)
	    {
	      if (ch == '\\')
		{
		  /* Special handling for backslash: a backslash may
		     be the beginning of a formal parameter (of a
		     macro) following another symbol character, with
		     whitespace in between.  If that is the case, we
		     output a space before the parameter.  Strictly
		     speaking, correct handling depends upon what the
		     macro parameter expands into; if the parameter
		     expands into something which does not start with
		     an operand character, then we don't want to keep
		     the space.  We don't have enough information to
		     make the right choice, so here we are making the
		     choice which is more likely to be correct.  */
		  PUT (' ');
		}

	      state = 3;
	    }
	  PUT (ch);
	  break;
	}
    }

  /*NOTREACHED*/

 fromeof:
  /* We have reached the end of the input.  */
  return to - tostart;

 tofull:
  /* The output buffer is full.  Save any input we have not yet
     processed.  */
  if (fromend > from)
    {
      saved_input = from;
      saved_input_len = fromend - from;
    }
  else
    saved_input = NULL;

  return to - tostart;
}

/* end of app.c */