Example #1
0
void
extract_awk (FILE *f,
             const char *real_filename, const char *logical_filename,
             flag_context_list_table_ty *flag_table,
             msgdomain_list_ty *mdlp)
{
  message_list_ty *mlp = mdlp->item[0]->messages;

  fp = f;
  real_file_name = real_filename;
  logical_file_name = xstrdup (logical_filename);
  line_number = 1;

  last_comment_line = -1;
  last_non_comment_line = -1;

  prefer_division_over_regexp = false;

  flag_context_list_table = flag_table;

  init_keywords ();

  /* Eat tokens until eof is seen.  When extract_parenthesized returns
     due to an unbalanced closing parenthesis, just restart it.  */
  while (!extract_parenthesized (mlp, null_context, null_context_list_iterator,
                                 arglist_parser_alloc (mlp, NULL)))
    ;

  fp = NULL;
  real_file_name = NULL;
  logical_file_name = NULL;
  line_number = 0;
}
Example #2
0
void
extract_php (FILE *f,
	     const char *real_filename, const char *logical_filename,
	     flag_context_list_table_ty *flag_table,
	     msgdomain_list_ty *mdlp)
{
  message_list_ty *mlp = mdlp->item[0]->messages;

  fp = f;
  real_file_name = real_filename;
  logical_file_name = xstrdup (logical_filename);
  line_number = 1;

  last_comment_line = -1;
  last_non_comment_line = -1;

  flag_context_list_table = flag_table;

  init_keywords ();

  /* Initial mode is HTML mode, not PHP mode.  */
  skip_html ();

  /* Eat tokens until eof is seen.  When extract_parenthesized returns
     due to an unbalanced closing parenthesis, just restart it.  */
  while (!extract_parenthesized (mlp, null_context, null_context_list_iterator,
				 arglist_parser_alloc (mlp, NULL)))
    ;

  /* Close scanner.  */
  fp = NULL;
  real_file_name = NULL;
  logical_file_name = NULL;
  line_number = 0;
}
Example #3
0
/* Extract messages until the next balanced closing parenthesis.
   Extracted messages are added to MLP.
   Return true upon eof, false upon closing parenthesis.  */
static bool
extract_parenthesized (message_list_ty *mlp,
                       flag_context_ty outer_context,
                       flag_context_list_iterator_ty context_iter,
                       struct arglist_parser *argparser)
{
  /* Current argument number.  */
  int arg = 1;
  /* 0 when no keyword has been seen.  1 right after a keyword is seen.  */
  int state;
  /* Parameters of the keyword just seen.  Defined only in state 1.  */
  const struct callshapes *next_shapes = NULL;
  /* Whether to implicitly assume the next tokens are arguments even without
     a '('.  */
  bool next_is_argument = false;
  /* Context iterator that will be used if the next token is a '('.  */
  flag_context_list_iterator_ty next_context_iter =
    passthrough_context_list_iterator;
  /* Current context.  */
  flag_context_ty inner_context =
    inherited_context (outer_context,
                       flag_context_list_iterator_advance (&context_iter));

  /* Start state is 0.  */
  state = 0;

  for (;;)
    {
      token_ty token;

      x_awk_lex (&token);

      if (next_is_argument && token.type != token_type_lparen)
        {
          /* An argument list starts, even though there is no '('.  */
          context_iter = next_context_iter;
          outer_context = inner_context;
          inner_context =
            inherited_context (outer_context,
                               flag_context_list_iterator_advance (
                                 &context_iter));
        }

      switch (token.type)
        {
        case token_type_symbol:
          {
            void *keyword_value;

            if (hash_find_entry (&keywords, token.string, strlen (token.string),
                                 &keyword_value)
                == 0)
              {
                next_shapes = (const struct callshapes *) keyword_value;
                state = 1;
              }
            else
              state = 0;
          }
          next_is_argument =
            (strcmp (token.string, "print") == 0
             || strcmp (token.string, "printf") == 0);
          next_context_iter =
            flag_context_list_iterator (
              flag_context_list_table_lookup (
                flag_context_list_table,
                token.string, strlen (token.string)));
          free (token.string);
          continue;

        case token_type_lparen:
          if (extract_parenthesized (mlp, inner_context, next_context_iter,
                                     arglist_parser_alloc (mlp,
                                                           state ? next_shapes : NULL)))
            {
              arglist_parser_done (argparser, arg);
              return true;
            }
          next_is_argument = false;
          next_context_iter = null_context_list_iterator;
          state = 0;
          continue;

        case token_type_rparen:
          arglist_parser_done (argparser, arg);
          return false;

        case token_type_comma:
          arg++;
          inner_context =
            inherited_context (outer_context,
                               flag_context_list_iterator_advance (
                                 &context_iter));
          next_is_argument = false;
          next_context_iter = passthrough_context_list_iterator;
          state = 0;
          continue;

        case token_type_string:
          {
            lex_pos_ty pos;
            pos.file_name = logical_file_name;
            pos.line_number = token.line_number;

            if (extract_all)
              remember_a_message (mlp, NULL, token.string, inner_context, &pos,
                                  NULL, savable_comment);
            else
              arglist_parser_remember (argparser, arg, token.string,
                                       inner_context,
                                       pos.file_name, pos.line_number,
                                       savable_comment);
          }
          next_is_argument = false;
          next_context_iter = null_context_list_iterator;
          state = 0;
          continue;

        case token_type_i18nstring:
          {
            lex_pos_ty pos;
            pos.file_name = logical_file_name;
            pos.line_number = token.line_number;

            remember_a_message (mlp, NULL, token.string, inner_context, &pos,
                                NULL, savable_comment);
          }
          next_is_argument = false;
          next_context_iter = null_context_list_iterator;
          state = 0;
          continue;

        case token_type_semicolon:
          /* An argument list ends, and a new statement begins.  */
          /* FIXME: Should handle newline that acts as statement separator
             in the same way.  */
          /* FIXME: Instead of resetting outer_context here, it may be better
             to recurse in the next_is_argument handling above, waiting for
             the next semicolon or other statement terminator.  */
          outer_context = null_context;
          context_iter = null_context_list_iterator;
          next_is_argument = false;
          next_context_iter = passthrough_context_list_iterator;
          inner_context =
            inherited_context (outer_context,
                               flag_context_list_iterator_advance (
                                 &context_iter));
          state = 0;
          continue;

        case token_type_eof:
          arglist_parser_done (argparser, arg);
          return true;

        case token_type_other:
          next_is_argument = false;
          next_context_iter = null_context_list_iterator;
          state = 0;
          continue;

        default:
          abort ();
        }
    }
}
Example #4
0
/* Extract messages until the next balanced closing parenthesis or bracket.
   Extracted messages are added to MLP.
   DELIM can be either token_type_rparen or token_type_rbracket, or
   token_type_eof to accept both.
   Return true upon eof, false upon closing parenthesis.  */
static bool
extract_balanced (message_list_ty *mlp,
		  token_type_ty delim,
		  flag_context_ty outer_context,
		  flag_context_list_iterator_ty context_iter,
		  struct arglist_parser *argparser)
{
  /* Current argument number.  */
  int arg = 1;
  /* 0 when no keyword has been seen.  1 right after a keyword is seen.  */
  int state;
  /* Parameters of the keyword just seen.  Defined only in state 1.  */
  const struct callshapes *next_shapes = NULL;
  /* Context iterator that will be used if the next token is a '('.  */
  flag_context_list_iterator_ty next_context_iter =
    passthrough_context_list_iterator;
  /* Current context.  */
  flag_context_ty inner_context =
    inherited_context (outer_context,
		       flag_context_list_iterator_advance (&context_iter));

  /* Start state is 0.  */
  state = 0;

  for (;;)
    {
      token_ty token;

      x_php_lex (&token);
      switch (token.type)
	{
	case token_type_symbol:
	  {
	    void *keyword_value;

	    if (hash_find_entry (&keywords, token.string, strlen (token.string),
				 &keyword_value)
		== 0)
	      {
		next_shapes = (const struct callshapes *) keyword_value;
		state = 1;
	      }
	    else
	      state = 0;
	  }
	  next_context_iter =
	    flag_context_list_iterator (
	      flag_context_list_table_lookup (
		flag_context_list_table,
		token.string, strlen (token.string)));
	  free (token.string);
	  continue;

	case token_type_lparen:
	  if (extract_balanced (mlp, token_type_rparen,
				inner_context, next_context_iter,
				arglist_parser_alloc (mlp,
						      state ? next_shapes : NULL)))
	    {
	      arglist_parser_done (argparser, arg);
	      return true;
	    }
	  next_context_iter = null_context_list_iterator;
	  state = 0;
	  continue;

	case token_type_rparen:
	  if (delim == token_type_rparen || delim == token_type_eof)
	    {
	      arglist_parser_done (argparser, arg);
	      return false;
	    }
	  next_context_iter = null_context_list_iterator;
	  state = 0;
	  continue;

	case token_type_comma:
	  arg++;
	  inner_context =
	    inherited_context (outer_context,
			       flag_context_list_iterator_advance (
				 &context_iter));
	  next_context_iter = passthrough_context_list_iterator;
	  state = 0;
	  continue;

	case token_type_lbracket:
	  if (extract_balanced (mlp, token_type_rbracket,
				null_context, null_context_list_iterator,
				arglist_parser_alloc (mlp, NULL)))
	    {
	      arglist_parser_done (argparser, arg);
	      return true;
	    }

	case token_type_rbracket:
	  if (delim == token_type_rbracket || delim == token_type_eof)
	    {
	      arglist_parser_done (argparser, arg);
	      return false;
	    }
	  next_context_iter = null_context_list_iterator;
	  state = 0;
	  continue;

	case token_type_string_literal:
	  {
	    lex_pos_ty pos;
	    pos.file_name = logical_file_name;
	    pos.line_number = token.line_number;

	    if (extract_all)
	      remember_a_message (mlp, NULL, token.string, inner_context,
				  &pos, token.comment);
	    else
	      arglist_parser_remember (argparser, arg, token.string,
				       inner_context,
				       pos.file_name, pos.line_number,
				       token.comment);
	    drop_reference (token.comment);
	  }
	  next_context_iter = null_context_list_iterator;
	  state = 0;
	  continue;

	case token_type_dot:
	case token_type_operator1:
	case token_type_operator2:
	case token_type_other:
	  next_context_iter = null_context_list_iterator;
	  state = 0;
	  continue;

	case token_type_eof:
	  arglist_parser_done (argparser, arg);
	  return true;

	default:
	  abort ();
	}
    }
}