68fa96c91dae536ac2568f77e6449b19fc5a9109,snorkel/parser.py,XMLDocParser,parse_file,#XMLDocParser#Any#Any#,136

Before Change

            // We store the XML tree as a string due to a serialization bug. It cannot currently be pickled directly
            //TODO: Implement a special dictionary that can handle this automatically (http://docs.sqlalchemy.org/en/latest/orm/extensions/mutable.html)
            attribs = {"root": et.tostring(doc)} if self.keep_xml_tree else {}
            yield Document(name=str(id), file=str(file_name), attribs=attribs), str(text)

    def _can_read(self, fpath):
        return fpath.endswith(".xml")

After Change

            meta = {"file_name": str(file_name)}
            if self.keep_xml_tree:
                meta["root"] = et.tostring(doc)
            yield Document(name=doc_id, stable_id=doc_id, meta=meta), text

    def _can_read(self, fpath):
        return fpath.endswith(".xml")

Instances