4e144c9f842d7415d8be5bdbb5912d88ae32cced,pycorrector/seq2seq/corpus_reader.py,CGEDReader,read_tokens,#CGEDReader#Any#Any#,96

Before Change

        with open(path, "r", encoding="utf-8") as f:
            dom_tree = minidom.parse(f)
            docs = dom_tree.documentElement.getElementsByTagName("DOC")
            for doc in docs:
                if is_infer:
                    // Input the error text
                    sentence = doc.getElementsByTagName("TEXT")[0]. \
                        childNodes[0].data.strip()
                else:
                    // Input the correct text
                    sentence = doc.getElementsByTagName("CORRECTION")[0]. \
                        childNodes[0].data.strip()
                yield segment(sentence, cut_type="char")

After Change

    def read_tokens(self, path, is_infer=False):
        i = 0
        with open(path, "r", encoding="utf-8") as f:
            for line in f:
                // Input the correct text, which start with 0
                if i % 2 == 1:
                    if line and len(line) > 5:
                        yield line.lower()[5:].strip().split()
                i += 1

Instances