Create a `.env` file for every release

[chaen]

Whenever we make a release, we should upload a .env file as github artifact that contains a dump of all the environment variables.

[fstagni]

What is the purpose? Debugging? Or is it an information that could be used elsewhere?

[fstagni]

BTW, copilot agent "solves" this with

name: Dump environment on release

on:
  release:
    types: [published]

permissions:
  contents: read

jobs:
  dump-env:
    runs-on: ubuntu-latest
    steps:
      - name: Checkout repository
        uses: actions/checkout@v4

      - name: Dump environment to file
        run: |
          echo "Warning: this will include all environment variables, including secrets if present."
          python - <<'PY'
          import os
          fn = 'env_dump.env'
          with open(fn, 'w', encoding='utf-8') as f:
              for k, v in os.environ.items():
                  if v is None:
                      v = ''
                  # Replace newlines with literal \n and escape backslashes and double quotes
                  v = v.replace('\n', '\\n')
                  v = v.replace('\\', '\\\\').replace('"', '\\"')
                  f.write(f'{k}="{v}"\n')
          print(f'Wrote {fn}')
          PY

      - name: Upload env artifact
        uses: actions/upload-artifact@v4
        with:
          name: env-${{ github.event.release.tag_name }}.env
          path: env_dump.env

But maybe you just want DIRACX_ variables?

[chrisburr]

As I only said this in the build but to make sure it isn't forgotten, copilot is very wrong here and a security issue :)

[ryuwd]

I've added a pixi task in #698 to do this, if you like I can add it as a pre-commit hook which regenerates it upon every commit

[chaen]

OK my bad, I should have been more explicit about this task.
The idea is to give an example of the environment of the containers running on Kubernetes, such that it could potentially be used for running diracx in docker. Given that there are a lot of dynamically generated values, doing it with a pixi task is probably incomplete. For example, it does not contain any of the SQL connection parameter.
As an exercise, I have already sent such a file to @sfayer , and I generated by dumping the environment of the containers of run_demo.sh, which I think is the proper way of doing.
I am still waiting for @sfayer to make sure it's suitable