Hi,
I was working on adding Google’s BigQuery as an Insights data source but have been confused while doing so.
I was expecting the entire flow to be bit modular, but turns out I would have to make a lot more changes than presumed.
It was kind of obvious to make a whole lot of changes given the fact that PostgresDB / MariaDB / Frappe site-db provide username and password based connection. Connection string can also be used to establish the connection in these cases. But when it comes to BigQuery, the connection is established using a service-account based authorization method.
I’m having some issues primarily with the understanding of what are the steps after a connection is established.
Until now, I have created a doctype called custom-data-source which takes a project ID and service-account as parameters and establishes a connection to the database.
I have made the changes to this fork.
I’m looking for a quick review of the progress I’ve made.
I have a few more questions:
- Am I following the right path?
- What are the steps further?
- How do I display the tables of my BigQuery DB to Insights dashboard from here?
- How do query builders work?
- Are query builders generic to all relational databases? Or is there a need to create new ones for BigQuery?
- Once I have the data from a data source, how are they visualized on the dashboard?
- Is there a complete workbook of how things work behind the scene for reference?