Exemple #1
0
def sdmx_report(sql, data_frame=None):
    """
    Build the sdmx file.

    :param sql: Sql text query.
    :param data_frame: The Pandas dataframe
    :return: Sdmx.
    """
    col_dict = build_column_dict(data_frame, sql)

    out_stream = StringIO.StringIO()
    generic_data = sdmx.GenericDataType()
    header = sdmx.HeaderType()
    header.set_Test(False)
    header.set_Truncated(False)
    ts = time.time()
    prepared = datetime.fromtimestamp(ts).strftime('%Y-%m-%dT%H:%M:%S')
    header.set_Prepared(prepared)
    sender = sdmx.PartyType()
    sender.set_id(SENDER)
    name = sdmx.TextType()
    name.set_lang(LANGUAGE_CODE)
    name.set_valueOf_(SENDER_NAME)
    sender.add_Name(name)
    header.add_Sender(sender)
    generic_data.set_Header(header)
    data_set = sdmx.DataSetType()
    data_set.set_keyFamilyURI(SENDER_NAME)
    data = get_data_from_data_frame(data_frame)
    for r, row in enumerate(data):
        series = sdmx.SeriesType()
        series_key = sdmx.SeriesKeyType()
        obs = sdmx.ObsType()
        for c in col_dict:
            v = row[c]
            val = "%s" % v
            value = sdmx.ValueType()
            concept = get_concept(col_dict, c)
            if concept == 'obsValue':
                if not val.startswith('*'):
                    obs_value = sdmx.ObsValueType()
                    obs_value.set_value(val)
                    obs.set_ObsValue(obs_value)
            elif concept is not None:
                value.set_concept(concept)
                value.set_value(val.strip())
                series_key.add_Value(value)
        series.set_SeriesKey(series_key)
        series.add_Obs(obs)
        data_set.add_Series(series)

    generic_data.set_DataSet(data_set)
    generic_data.export(out_stream, 0)

    value = out_stream.getvalue()
    value = add_xml_header(value)

    return value
Exemple #2
0
def rdf_report(sql,
               title,
               description,
               data_frame=None,
               rdf_format=None):
    """
    Get the rdf report.

    :param sql: The query.
    :param title: The query title.
    :param description: Query description.
    :param data_frame: The Pandas data frame.
    :param rdf_format: The desired rdf format.
    :return: The serialized Rdf.
    """
    g = Graph()
    data = get_data_from_data_frame(data_frame)
    title = title.decode('utf-8')
    dataset = "dataset-%s" % title
    subject_t_ref = term.URIRef(my_ns[dataset])
    source_tables = get_source_tables(sql)

    g = add_table_triples(description, g, source_tables, title, subject_t_ref)
    col_dict = build_column_dict(data_frame, sql)
    slice_t_ref = term.URIRef(my_ns['slice'])

    g = add_slice_triples(g, subject_t_ref,
                          data_frame, col_dict)
    g = add_observations_triples(g,
                                 data,
                                 title,
                                 slice_t_ref,
                                 col_dict)
    ser = g.serialize(format=rdf_format)

    if rdf_format == "xml":
        ser = add_xml_header(ser)
    elif rdf_format == "turtle":
        ser = add_sh_like_header(ser)
    return ser