cdplib/import_process_instances/CleanRs70.py

#!/usr/bin/env python3
# -*- coding: utf-8 -*-
"""
Created on Mon Sep 30 10:11:55 2019

@author: tanya
"""
import pandas as pd
import os
import sys

sys.path.append(os.getcwd())

from libraries.import_process_instances.CleanProcessTable import CleanProcessTable


class CleanRs70(CleanProcessTable):
    '''
    '''
    def __init__(self):
        '''
        '''
        super().__init__(
                mapping_path=os.path.join(".", "migration_mappings",
                                          "rs70_mapping.json"),
                inconsist_report_table="inconsist_rs70",
                sort_columns=["radsatznummer", "eingabe_datum"],
                index_columns=["radsatznummer", "eingabe_datum"],
                log_name="CleanRs70")

    def filter_invalid_schadcode(self, data: pd.DataFrame) -> pd.DataFrame:
        '''
        In the configuration we store a list of schadcodes assignment
         of which means that the product is scrap. No more schadcodes after
         this schadcode should be assigned.
        '''
        from libraries.configuration import default as cfg

        self.error_column_abscence(columns=["radsatznummer", "schadcode"],
                                   data=data)

        data.sort_values(by=self._sort_columns, inplace=True)

        is_last_schadcode = (data["radsatznummer"] !=
                             data["radsatznummer"].shift(-1))

        is_invalid = (~is_last_schadcode &
                      data["schadcode"].isin(cfg.schrott_schadcodes))

        data = self._filter_invalid_data(
                    data=data,
                    invalid_mask=is_invalid,
                    reason="invalid schadcode")

        # XXX temporary here
        # data["eingabe_datum"] = pd.to_datetime(data["eingabe_datum"])

        return data