Release

docker build --platform linux/amd64 -t hub.opensciencegrid.org/chtc/chtc-user-app-api:active_field .

docker push hub.opensciencegrid.org/chtc/chtc-user-app-api:active_field

docker run -it -p 8000:8000 hub.opensciencegrid.org/opensciencegrid/chtc-userapp-api:latest

Run the NGINX server that sits in front of the webhook

docker run -it -p 80:80 -v ${PWD}/srv:/srv  -v ${PWD}/nginx.conf:/etc/nginx/nginx.conf:ro nginx

Query Parser Docs

All list endpoints in the api allow filtering based on url parameters.

The format for a query is:

<column_name>=<comparator>.<value>

So to query a user with netid equal to clock they url would be /users?netid=eq.clock.

All comparators are ["not", "eq", "lt", "le", "gt", "ge", "ne", "like", "ilike", "in", "is"].

In the case of not you must use it in conjunction with another comparator, such as /users?netid=not.like.clock.

like and ilike automatically have their values sandwiched between % so /users?netid=like.clock is equivalent to netid LIKE '%clock%'.

All comparators are automatically combined with AND.

Example:

/users?netid=not.like.clock&username=not.eq.cannonlock is the SQL equivalent to NOT netid LIKE '%clock%' AND NOT username = 'cannonlock'.

Ordering

Ordering format is:

<column_name>=order_by.<asc | desc>

For example:

https://userapp.chtc.wisc.edu/api/users?page=0&page_size=50&date=order_by.desc

Tests

Requires Docker (for a throwaway Postgres) and Python 3.12.

First-time setup — create the virtualenv and install dependencies:

python3 -m venv .venv && .venv/bin/pip install -r requirements.txt

Then run the full suite:

make test

To run against a different interpreter (e.g. an already-activated environment), override PYTHON:

make test PYTHON=python

On Windows, the virtualenv interpreter lives at .venv/Scripts/python.exe rather than .venv/bin/python