itunesTemplate.py

Go to the documentation of this file.

import re
import awswrangler as wr
 
class itunesTemplate: 
    def preprocessing(self, filename, features, session, rel_col):
        """Loads file and do some fixes to fit it to our standard and make it able to be processed
 
        Args:
            filename (str): s3 full path of filename
            features (dict): contains delimiter, skip_rows and encoding required for current file
            session (boto3 obj): AWS client connection
            rel_col (str): column name used to identify upc/release_id
        Returns: df (pandas dataframe)
        """
        delimiter = features["delimiter"]
        skip_rows = features["skip_rows"]
        encoding = features["encoding"]
        df = wr.s3.read_csv(filename, sep=delimiter, skiprows=skip_rows, encoding=encoding, dtype={rel_col:str}, low_memory=False, boto3_session=session)
        # df = read_csv(filename, sep=delimiter, skiprows=skip_rows, encoding=encoding, low_memory=False)
        total_rows = df["End Date"][df["Start Date"]=="Total_Rows"]
        total_rows = int(total_rows)
        df.drop(range(total_rows, df.shape[0], 1), axis=0, inplace=True)
        return df
 
    def preprocessing_music(self, filename, features, session, rel_col):
        """Loads file and do some fixes to fit it to our standard and make it able to be processed.
           It's used for a specific itunes version
 
        Args:
            filename (str): s3 full path of filename
            features (dict): contains delimiter, skip_rows and encoding required for current file
            session (boto3 obj): AWS client connection
            rel_col (str): column name used to identify upc/release_id
        Returns: df (pandas dataframe)
        """
        delimiter = features["delimiter"]
        skip_rows = features["skip_rows"]
        encoding = features["encoding"]
        df = wr.s3.read_csv(filename, sep=delimiter, skiprows=skip_rows, encoding=encoding, dtype={rel_col:str}, low_memory=False, boto3_session=session)
        # df = read_csv(filename, sep=delimiter, skiprows=skip_rows, low_memory=False)
        total_rows = df["Apple Identifier"][df["Storefront Name"]=="Row Count"]
        total_rows = int(total_rows)
        df.drop(range(total_rows, df.shape[0], 1), axis=0, inplace=True)
        return df
 
    def type(self, upc):
        """Maps the sale type with the name specified by the client according their requirements
        Args:
            upc (str): sale type (for products) shown in the original file
        Returns: (sale type in a single one letter)
        """
        if str(upc) == "" or str(upc) == "nan":
            return "T"
        return "A"
 
    def assetType(self, df):
        """Applies type functions to current dataframe
        Args:
            df (pandas dataframe): dataframe where changes will applied
        Returns: df (pandas dataframe)
        """
        df["gettypeitunes"] = df.apply(lambda Row: self.type(Row['upc']), axis=1)
        return df
 
    def date(self, filename):
        """Sets date column given the filename (it contains the date)
           For example S1_89680172_0723_ZZ.txt
        Args:
            filename (str): current filename
        Returns: date_str (str)
        """
        date = re.findall(r'_\d{4}_', filename)[0]
        date = date.replace("_", "")
        month = date[:2] 
        year = "20"+date[2:] 
        date_str = year+month+"01"
        # df["from_file_itune"] = date_str
        return date_str