示例#1
0
文件: cfg.py 项目: sfu-natlang/nltk
    def _parse_productions(self):
        """
        Parse the current contents of the textwidget buffer, to create
        a list of productions.
        """
        productions = []

        # Get the text, normalize it, and split it into lines.
        text = self._textwidget.get('1.0', 'end')
        text = re.sub(self.ARROW, '->', text)
        text = re.sub('\t', ' ', text)
        lines = text.split('\n')

        # Convert each line to a CFG production
        for line in lines:
            line = line.strip()
            if line=='': continue
            productions += parse_cfg_production(line)
            #if line.strip() == '': continue
            #if not CFGEditor._PRODUCTION_RE.match(line):
            #    raise ValueError('Bad production string %r' % line)
            #
            #(lhs_str, rhs_str) = line.split('->')
            #lhs = Nonterminal(lhs_str.strip())
            #rhs = []
            #def parse_token(match, rhs=rhs):
            #    token = match.group()
            #    if token[0] in "'\"": rhs.append(token[1:-1])
            #    else: rhs.append(Nonterminal(token))
            #    return ''
            #CFGEditor._TOKEN_RE.sub(parse_token, rhs_str)
            #
            #productions.append(Production(lhs, *rhs))

        return productions
示例#2
0
文件: cfg.py 项目: wrand/tweater
    def _parse_productions(self):
        """
        Parse the current contents of the textwidget buffer, to create
        a list of productions.
        """
        productions = []

        # Get the text, normalize it, and split it into lines.
        text = self._textwidget.get('1.0', 'end')
        text = re.sub(self.ARROW, '->', text)
        text = re.sub('\t', ' ', text)
        lines = text.split('\n')

        # Convert each line to a CFG production
        for line in lines:
            line = line.strip()
            if line == '': continue
            productions += parse_cfg_production(line)
            #if line.strip() == '': continue
            #if not CFGEditor._PRODUCTION_RE.match(line):
            #    raise ValueError('Bad production string %r' % line)
            #
            #(lhs_str, rhs_str) = line.split('->')
            #lhs = Nonterminal(lhs_str.strip())
            #rhs = []
            #def parse_token(match, rhs=rhs):
            #    token = match.group()
            #    if token[0] in "'\"": rhs.append(token[1:-1])
            #    else: rhs.append(Nonterminal(token))
            #    return ''
            #CFGEditor._TOKEN_RE.sub(parse_token, rhs_str)
            #
            #productions.append(Production(lhs, *rhs))

        return productions