SoundtrackTemplate.py

Go to the documentation of this file.

import re
import awswrangler as wr
 
class SoundtrackTemplate:
    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}, keep_default_na=False, low_memory=False, boto3_session=session)
        # df = read_csv(filename, sep=delimiter, skiprows=skip_rows, keep_default_na=False, low_memory=False)
        df["Country_Of_Sale"] = df["Country_Of_Sale"].apply(lambda x: str(x).replace("   ", ""))
        return df
 
    def date(self, filename):
        """Sets date column given the filename (it contains the date)
           For example gyrostream-pty-ltd_soundtrack-your-brand_202203_Monthly-Sales.csv
        Args:
            filename (str): current filename
        Returns: date_str (str)
        """
        # "gyrostream-pty-ltd_soundtrack-your-brand_202203_Monthly-Sales.csv"
        date = re.findall(r'_\d{6}_', filename)[0]
        date = date.replace("_", "")
        date_str = date[:4]+"-"+date[4:]+"-01"
        # df["date_from_file"] = date_str
        return date_str