def __getitem__()

in dpr/data/biencoder_data.py [0:0]

• 19 lines of code
• 9 McCabe index (conditional complexity)

    def __getitem__(self, index) -> BiEncoderSample:
        json_sample = self.data[index]
        r = BiEncoderSample()
        r.query = self._process_query(json_sample["question"])

        positive_ctxs = json_sample["positive_ctxs"]
        negative_ctxs = json_sample["negative_ctxs"] if "negative_ctxs" in json_sample else []
        hard_negative_ctxs = json_sample["hard_negative_ctxs"] if "hard_negative_ctxs" in json_sample else []

        for ctx in positive_ctxs + negative_ctxs + hard_negative_ctxs:
            if "title" not in ctx:
                ctx["title"] = None

        def create_passage(ctx: dict):
            return BiEncoderPassage(
                normalize_passage(ctx["text"]) if self.normalize else ctx["text"],
                ctx["title"],
            )

        r.positive_passages = [create_passage(ctx) for ctx in positive_ctxs]
        r.negative_passages = [create_passage(ctx) for ctx in negative_ctxs]
        r.hard_negative_passages = [create_passage(ctx) for ctx in hard_negative_ctxs]
        return r