-
Notifications
You must be signed in to change notification settings - Fork 0
Open
Labels
Description
Description
Considering replacement of SQL language with one standardised workflow, using python for example.
- This will work fine for synthetic data, but maybe not real data – maybe that is fine at this stage?
Options to consider:
- All SQL (current state)
- SQL database but python code for everything else
- All python (see fast read, fast write functions)
Additional context or further questions:
- Think about the pros and cons of the above options first (e.g. code efficiency, understandably, scalability) before implementing
- See cvd-net-wrangle for context: https://github.com/alan-turing-institute/cvd-net-wrangle