Class: Gherkin::CLexer::It

Inherits:

Object

• Object
• Gherkin::CLexer::It

Defined in:

ext/gherkin_lexer_it/gherkin_lexer_it.c

Instance Method Summary

• #initialize(listener) ⇒ Object constructor
• #scan(input) ⇒ Object

Constructor Details

#initialize(listener) ⇒ Object

# File 'ext/gherkin_lexer_it/gherkin_lexer_it.c', line 907

static VALUE CLexer_init(VALUE self, VALUE listener)
{
  rb_iv_set(self, "@listener", listener);
  
  lexer_state *lxr = NULL;
  DATA_GET(self, lexer_state, lxr);
  lexer_init(lxr);
  
  return self;
}

Instance Method Details

#scan(input) ⇒ Object

# File 'ext/gherkin_lexer_it/gherkin_lexer_it.c', line 918

static VALUE CLexer_scan(VALUE self, VALUE input)
{
  VALUE listener = rb_iv_get(self, "@listener");

  lexer_state *lexer = NULL;
  DATA_GET(self, lexer_state, lexer);

  VALUE input_copy = rb_str_dup(input);

  rb_str_append(input_copy, rb_str_new2("\n%_FEATURE_END_%"));
  char *data = RSTRING_PTR(input_copy);
  size_t len = RSTRING_LEN(input_copy);
  
  if (count_char('\r', data) > (count_char('\n', data) / 2)) {
    lexer->eol = CRLF_FLAG;
  }
  
  if (len == 0) { 
    rb_raise(rb_eGherkinLexingError, "No content to lex.");
  } else {

    const char *p, *pe, *eof;
    int cs = 0;
    
    VALUE current_row = Qnil;

    p = data;
    pe = data + len;
    eof = pe;
    
    assert(*pe == '\0' && "pointer does not end on NULL");
    
    
#line 952 "ext/gherkin_lexer_it/gherkin_lexer_it.c"
	{
	cs = lexer_start;
	}

#line 430 "/Users/aslakhellesoy/scm/gherkin/tasks/../ragel/i18n/it.c.rl"
    
#line 959 "ext/gherkin_lexer_it/gherkin_lexer_it.c"
	{
	int _klen;
	unsigned int _trans;
	const char *_acts;
	unsigned int _nacts;
	const char *_keys;

	if ( p == pe )
		goto _test_eof;
	if ( cs == 0 )
		goto _out;
_resume:
	_keys = _lexer_trans_keys + _lexer_key_offsets[cs];
	_trans = _lexer_index_offsets[cs];

	_klen = _lexer_single_lengths[cs];
	if ( _klen > 0 ) {
		const char *_lower = _keys;
		const char *_mid;
		const char *_upper = _keys + _klen - 1;
		while (1) {
			if ( _upper < _lower )
				break;

			_mid = _lower + ((_upper-_lower) >> 1);
			if ( (*p) < *_mid )
				_upper = _mid - 1;
			else if ( (*p) > *_mid )
				_lower = _mid + 1;
			else {
				_trans += (_mid - _keys);
				goto _match;
			}
		}
		_keys += _klen;
		_trans += _klen;
	}

	_klen = _lexer_range_lengths[cs];
	if ( _klen > 0 ) {
		const char *_lower = _keys;
		const char *_mid;
		const char *_upper = _keys + (_klen<<1) - 2;
		while (1) {
			if ( _upper < _lower )
				break;

			_mid = _lower + (((_upper-_lower) >> 1) & ~1);
			if ( (*p) < _mid[0] )
				_upper = _mid - 2;
			else if ( (*p) > _mid[1] )
				_lower = _mid + 2;
			else {
				_trans += ((_mid - _keys)>>1);
				goto _match;
			}
		}
		_trans += _klen;
	}

_match:
	cs = _lexer_trans_targs[_trans];

	if ( _lexer_trans_actions[_trans] == 0 )
		goto _again;

	_acts = _lexer_actions + _lexer_trans_actions[_trans];
	_nacts = (unsigned int) *_acts++;
	while ( _nacts-- > 0 )
	{
		switch ( *_acts++ )
		{
	case 0:
#line 87 "/Users/aslakhellesoy/scm/gherkin/tasks/../ragel/i18n/it.c.rl"
	{
		MARK(content_start, p);
    lexer->current_line = lexer->line_number;
  }
	break;
	case 1:
#line 92 "/Users/aslakhellesoy/scm/gherkin/tasks/../ragel/i18n/it.c.rl"
	{
    MARK(content_start, p);
  }
	break;
	case 2:
#line 96 "/Users/aslakhellesoy/scm/gherkin/tasks/../ragel/i18n/it.c.rl"
	{
    lexer->current_line = lexer->line_number;
    lexer->start_col = p - data - lexer->last_newline;
  }
	break;
	case 3:
#line 101 "/Users/aslakhellesoy/scm/gherkin/tasks/../ragel/i18n/it.c.rl"
	{
    int len = LEN(content_start, PTR_TO(final_newline));

    if (len < 0) len = 0;

    store_pystring_content(listener, lexer->start_col, PTR_TO(content_start), len, lexer->current_line);
  }
	break;
	case 4:
#line 109 "/Users/aslakhellesoy/scm/gherkin/tasks/../ragel/i18n/it.c.rl"
	{
    STORE_KW_END_CON(feature);
  }
	break;
	case 5:
#line 113 "/Users/aslakhellesoy/scm/gherkin/tasks/../ragel/i18n/it.c.rl"
	{
    STORE_KW_END_CON(background);
  }
	break;
	case 6:
#line 117 "/Users/aslakhellesoy/scm/gherkin/tasks/../ragel/i18n/it.c.rl"
	{
    STORE_KW_END_CON(scenario);
  }
	break;
	case 7:
#line 121 "/Users/aslakhellesoy/scm/gherkin/tasks/../ragel/i18n/it.c.rl"
	{
    STORE_KW_END_CON(scenario_outline);
  }
	break;
	case 8:
#line 125 "/Users/aslakhellesoy/scm/gherkin/tasks/../ragel/i18n/it.c.rl"
	{
    STORE_KW_END_CON(examples);
  }
	break;
	case 9:
#line 129 "/Users/aslakhellesoy/scm/gherkin/tasks/../ragel/i18n/it.c.rl"
	{
    store_kw_con(listener, "step",
      PTR_TO(keyword_start), LEN(keyword_start, PTR_TO(keyword_end)),
      PTR_TO(content_start), LEN(content_start, p), 
      lexer->current_line);
  }
	break;
	case 10:
#line 136 "/Users/aslakhellesoy/scm/gherkin/tasks/../ragel/i18n/it.c.rl"
	{
    STORE_ATTR(comment);
    lexer->mark = 0;
  }
	break;
	case 11:
#line 141 "/Users/aslakhellesoy/scm/gherkin/tasks/../ragel/i18n/it.c.rl"
	{
    STORE_ATTR(tag);
    lexer->mark = 0;
  }
	break;
	case 12:
#line 146 "/Users/aslakhellesoy/scm/gherkin/tasks/../ragel/i18n/it.c.rl"
	{
    lexer->line_number += 1;
    MARK(final_newline, p);
  }
	break;
	case 13:
#line 151 "/Users/aslakhellesoy/scm/gherkin/tasks/../ragel/i18n/it.c.rl"
	{
    MARK(last_newline, p + 1);
  }
	break;
	case 14:
#line 155 "/Users/aslakhellesoy/scm/gherkin/tasks/../ragel/i18n/it.c.rl"
	{
    if (lexer->mark == 0) {
      MARK(mark, p);
    }
  }
	break;
	case 15:
#line 161 "/Users/aslakhellesoy/scm/gherkin/tasks/../ragel/i18n/it.c.rl"
	{
    MARK(keyword_end, p);
    MARK(keyword_start, PTR_TO(mark));
    MARK(content_start, p + 1);
    lexer->mark = 0;
  }
	break;
	case 16:
#line 168 "/Users/aslakhellesoy/scm/gherkin/tasks/../ragel/i18n/it.c.rl"
	{
    MARK(content_end, p);
  }
	break;
	case 17:
#line 172 "/Users/aslakhellesoy/scm/gherkin/tasks/../ragel/i18n/it.c.rl"
	{
    p = p - 1;
    lexer->current_line = lexer->line_number;
    current_row = rb_ary_new();
  }
	break;
	case 18:
#line 178 "/Users/aslakhellesoy/scm/gherkin/tasks/../ragel/i18n/it.c.rl"
	{
		MARK(content_start, p);
  }
	break;
	case 19:
#line 182 "/Users/aslakhellesoy/scm/gherkin/tasks/../ragel/i18n/it.c.rl"
	{
    VALUE con = ENCODED_STR_NEW(PTR_TO(content_start), LEN(content_start, p));
    rb_funcall(con, rb_intern("strip!"), 0);
    VALUE re_pipe      = rb_reg_regcomp(rb_str_new2("\\\\\\|"));
    VALUE re_backslash = rb_reg_regcomp(rb_str_new2("\\\\\\\\"));
    rb_funcall(con, rb_intern("gsub!"), 2, re_pipe,      rb_str_new2("|"));
    rb_funcall(con, rb_intern("gsub!"), 2, re_backslash, rb_str_new2("\\"));

    rb_ary_push(current_row, con);
  }
	break;
	case 20:
#line 193 "/Users/aslakhellesoy/scm/gherkin/tasks/../ragel/i18n/it.c.rl"
	{
    rb_funcall(listener, rb_intern("row"), 2, current_row, INT2FIX(lexer->current_line));
  }
	break;
	case 21:
#line 197 "/Users/aslakhellesoy/scm/gherkin/tasks/../ragel/i18n/it.c.rl"
	{
    if (cs < lexer_first_final) {
      if (raise_lexer_error != NULL) {
        size_t count = 0;
    	VALUE newstr_val;
        char *newstr;
        int newstr_count = 0;        
        size_t len;
        const char *buff;
        if (lexer->last_newline != 0) {
          len = LEN(last_newline, eof);
          buff = PTR_TO(last_newline);
        } else {
          len = strlen(data);
          buff = data;
        }

        // Allocate as a ruby string so that it gets cleaned up by GC
        newstr_val = rb_str_new(buff, len);
        newstr = RSTRING_PTR(newstr_val);

        for (count = 0; count < len; count++) {
          if(buff[count] == 10) {
            newstr[newstr_count] = '\0'; // terminate new string at first newline found
            break;
          } else {
            if (buff[count] == '%') {
              newstr[newstr_count++] = buff[count];
              newstr[newstr_count] = buff[count];
            } else {
              newstr[newstr_count] = buff[count];
            }
          }
          newstr_count++;
        }

        int line = lexer->line_number;
        lexer_init(lexer); // Re-initialize so we can scan again with the same lexer
        raise_lexer_error(newstr, line);
      }
    } else {
      rb_funcall(listener, rb_intern("eof"), 0);
    }
  }
	break;
#line 1231 "ext/gherkin_lexer_it/gherkin_lexer_it.c"
		}
	}

_again:
	if ( cs == 0 )
		goto _out;
	if ( ++p != pe )
		goto _resume;
	_test_eof: {}
	if ( p == eof )
	{
	const char *__acts = _lexer_actions + _lexer_eof_actions[cs];
	unsigned int __nacts = (unsigned int) *__acts++;
	while ( __nacts-- > 0 ) {
		switch ( *__acts++ ) {
	case 21:
#line 197 "/Users/aslakhellesoy/scm/gherkin/tasks/../ragel/i18n/it.c.rl"
	{
    if (cs < lexer_first_final) {
      if (raise_lexer_error != NULL) {
        size_t count = 0;
    	VALUE newstr_val;
        char *newstr;
        int newstr_count = 0;        
        size_t len;
        const char *buff;
        if (lexer->last_newline != 0) {
          len = LEN(last_newline, eof);
          buff = PTR_TO(last_newline);
        } else {
          len = strlen(data);
          buff = data;
        }

        // Allocate as a ruby string so that it gets cleaned up by GC
        newstr_val = rb_str_new(buff, len);
        newstr = RSTRING_PTR(newstr_val);

        for (count = 0; count < len; count++) {
          if(buff[count] == 10) {
            newstr[newstr_count] = '\0'; // terminate new string at first newline found
            break;
          } else {
            if (buff[count] == '%') {
              newstr[newstr_count++] = buff[count];
              newstr[newstr_count] = buff[count];
            } else {
              newstr[newstr_count] = buff[count];
            }
          }
          newstr_count++;
        }

        int line = lexer->line_number;
        lexer_init(lexer); // Re-initialize so we can scan again with the same lexer
        raise_lexer_error(newstr, line);
      }
    } else {
      rb_funcall(listener, rb_intern("eof"), 0);
    }
  }
	break;
#line 1294 "ext/gherkin_lexer_it/gherkin_lexer_it.c"
		}
	}
	}

	_out: {}
	}

#line 431 "/Users/aslakhellesoy/scm/gherkin/tasks/../ragel/i18n/it.c.rl"

    assert(p <= pe && "data overflow after parsing execute");
    assert(lexer->content_start <= len && "content starts after data end");
    assert(lexer->mark < len && "mark is after data end");
    
    // Reset lexer by re-initializing the whole thing
    lexer_init(lexer);

    if (cs == lexer_error) {
      rb_raise(rb_eGherkinLexingError, "Invalid format, lexing fails.");
    } else {
      return Qtrue;
    }
  }
}