def fetch()

in competitions/leaderboard.py [0:0]

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

    def fetch(self, private=False):
        if private:
            submissions = self._process_private_lb()
        else:
            submissions = self._process_public_lb()

        if len(submissions) == 0:
            return pd.DataFrame()

        df = pd.DataFrame(submissions)

        # convert submission datetime to pandas datetime
        df["submission_datetime"] = pd.to_datetime(df["submission_datetime"], format="%Y-%m-%d %H:%M:%S")

        # only keep submissions before the end date
        df = df[df["submission_datetime"] < self.end_date].reset_index(drop=True)

        # sort by submission datetime
        # sort by public score and submission datetime
        if self.eval_higher_is_better:
            if private:
                df = df.sort_values(
                    by=[self.scoring_metric, "submission_datetime"],
                    ascending=[False, True],
                )
            else:
                df = df.sort_values(
                    by=[self.scoring_metric, "submission_datetime"],
                    ascending=[False, True],
                )
        else:
            if private:
                df = df.sort_values(
                    by=[self.scoring_metric, "submission_datetime"],
                    ascending=[True, True],
                )
            else:
                df = df.sort_values(
                    by=[self.scoring_metric, "submission_datetime"],
                    ascending=[True, True],
                )

        # only keep 4 significant digits in the scores
        for col in df.columns:
            if col in self.non_score_columns:
                continue
            df[col] = df[col].round(4)

        # reset index
        df = df.reset_index(drop=True)
        df["rank"] = df.index + 1

        # convert datetime column to string
        df["submission_datetime"] = df["submission_datetime"].dt.strftime("%Y-%m-%d %H:%M:%S")

        # send submission_datetime to the end
        columns = df.columns.tolist()
        columns.remove("submission_datetime")
        columns.append("submission_datetime")
        df = df[columns]

        # send rank to first position
        columns = df.columns.tolist()
        columns.remove("rank")
        columns = ["rank"] + columns
        df = df[columns]

        team_metadata = hf_hub_download(
            repo_id=self.competition_id,
            filename="teams.json",
            token=self.token,
            repo_type="dataset",
        )
        with open(team_metadata, "r", encoding="utf-8") as f:
            team_metadata = json.load(f)

        df["id"] = df["id"].apply(lambda x: team_metadata[x]["name"])

        return df