Ingest Google Sheets Data - ADR 002
Context
Hackney currently have datasets distributed over a number of different Google Sheets Documents due their data recovery efforts. We need to get this information pulled into the data platform for processing.
Decision
We will use a python based AWS Glue Job in conjunction with the gspread python library to pull the data onto platform
Consequences
Having the code sit with AWS Glue makes the data import easier to keep consistent with the rest of the platform but leaves us dependent on being able to import the gspread library into the AWS Glue job and requires that we access credentials from Google Workspace