Closed ajparsons closed 3 months ago
I've added a venv creation process to deploy.bash
.
Added some packages to the packages file for future docker use but these are all present on the server anyway so should be fine.
Issues above resolved, I've also just quickly updated the packages now we're not stuck on an old pandas version because of 3.7 - https://github.com/mysociety/theyworkforyou/pull/1765/commits/49053830b6d01c0baa9156abddfc7bf693fa82a4
This PR adds
division_io.py
- a click CLI to handle importing and exporting division data.This handles bulk importing updates to division titles (e.g. from Parliament's commons votes api), and exporting parquet tables of divisions as a basic api to feed twfy-votes.
There is an adjustment to the division table schema - adding a
title_priority
field. This is to keep track of the origin of the current title, and stop manual updates to titles being overridden by automated ones. Existing titles with ayes_text
are retrospectively updated in priority.For the moment, this uses a requirements.txt file for the needed python packages. When the server is updated, I'll do a new PR to tidy up the python-tooling in general. Unsure for the moment if this needs the morningupdate commands to run in a venv, or if system packages can just be updated in the short term.