I have to read set of csv files which have 5 columns like name, age, address, type and distance. As these column names are string and I want to iterate them in my pandas df. I have created a class that store name with a variable so that if a name change in these files, I just need to update this in one location and work for all case. class PersonDetailName: name = "name" age = "age" type = "type" address = "address" distance = "distance" Now to access this in iteration I can directly call it like row[PersonDetailName.name"] and it can handle the name automatically. If some files which still have these five fields however instead of age they have personAge and for address they have FullAddress. How can I handle this without changing much in the code. My approaches for this are: Dynamic Column Name Mapping class PaymentDetailName: age = "age" address = "address" dynamic_col_name_mapping = { "type1": {"PersonAge", "PersonAddress"}, "type2": {"PersonAge1", "PersonAddress1"}, "default": {"age", "address"} } @classmethod def update_payment_name(cls, type): cls.age, cls.address = cls.dynamic_col_name_mapping[type] Updating name list in iterator age, address = PaymentDetailName.dynamic_col_name_mapping[type] age = row[age] type = row[PaymentDetailName.type] I want to understand which of the solution is better and is there other method or programming knowledge I can use to make it more dynamic and less amount of code. The PaymentDetailName class should remain and rest I can update to fulfill this requirement. Creating a base class and creating two subclass for this seems overkill to me. Please provide your insights to make better class design. Continue reading...
