corpus.py 文件源码

def docs(self, fileids=None, categories=None):
        """
        Returns the complete JSON document for every file in the corpus.
        Note that I attempted to use the nltk ``CorpusView`` and ``concat``
        methods here, but was not getting memory safe iteration. Instead the
        simple Python generator by far did a better job of ensuring that file
        handles got closed and that not all data was loaded into memory at a
        time. In the future, I will try to re-implement the corpus view.
        """
        # Resolve the fileids and the categories
        fileids = self._resolve(fileids, categories)

        # Create a generator, loading one document into memory at a time.
        for path, enc, fileid in self.abspaths(fileids, True, True):
            with codecs.open(path, 'r', encoding=enc) as f:
                yield json.load(f)