utils.py 文件源码

def df_multi_reader(filename: str, limit: bool=False) -> PandasDF:
    df = DataFrame()

    try:
        assert isinstance(filename, str), "filename isn't string %s" % filename
        assert isinstance(limit, bool), "limit isn't bool %s" % limit

        if settings.DATA_TYPE == "pickle":
            f = filename + ".mp"
            if isfile(f):
                df = read_pickle(f)
        if settings.DATA_TYPE == "proto2":
            f = filename + ".pr2"
            if isfile(f):
                df = read_pickle(f)
        if settings.DATA_TYPE == "messagepack":
            f = filename + ".pack"
            if isfile(f):
                df = read_msgpack(f)
        if settings.DATA_TYPE == "json":
            f = filename + ".json"
            if isfile(f):
                df = read_json(f)
        if settings.DATA_TYPE == "feather":
            #TODO feather doesn't handle indexes
            f = filename + ".fth"
            if isfile(f):
                df = read_feather(f).reset_index()
        if settings.DATA_TYPE == "hdf":
            f = filename + ".hdf"
            if isfile(f):
                df = read_hdf(f, key=filename)
        if settings.DATA_TYPE == "hdfone":
            f = join(settings.DATA_PATH, "hdfone.hdfone")
            if isfile(f):
                df = read_hdf(f, key=filename, mode='r')

        if limit:
            if len(df.index) > 0:
                df = df.last(settings.LIMIT_MONTHS)
    except Exception as err:
        print(colored.red("MultiReader {}".format(err)))

    return df