test.yml

name: Test

on:
  pull_request:
  push:
  schedule:
    - cron: '0 6 * * *'

concurrency:
  group: ${{ github.workflow }}-${{ github.event_name }}-${{ github.ref_name }}
  cancel-in-progress: true

defaults:
  run:
    # Use an interactive shell so that ~/.bashrc is sourced:
    shell: 'bash -eio pipefail {0}'

env:
   # Note, that there's "turtle" as well, which is always excluded from running.
   PYTEST_SELECTION: "integration or usecase or slow or network"
   PYTEST_SELECTION_OP: "not "  # so it would be "not (integration or usecase)"
   DATALAD_TESTS_SSH: "1"
   DATALAD_LOG_ENV: GIT_SSH_COMMAND
   # How/which git-annex we install.  conda's build would be the fastest,
   # but it must not get ahead in PATH to not shadow travis' python
   _DL_ANNEX_INSTALL_SCENARIO: "miniconda=py37_23.1.0-1 --python-match minor --batch git-annex=8.20201007 -m conda"
   BOTO_CONFIG: /tmp/nowhere
   DATALAD_DATASETS_TOPURL: https://datasets-tests.datalad.org

jobs:
  filter:
    runs-on: ubuntu-latest
    outputs:
      jobs: ${{ steps.jobs.outputs.matrix }}
    steps:
      - name: Check out repository
        uses: actions/checkout@v4

      - name: Possibly filter out cron-only jobs
        id: jobs
        run: |
          echo 'matrix<<EOT' >> "$GITHUB_OUTPUT"
          if [ "${{ github.event_name == 'schedule' }}" = true ]
          then
              yq '{"include": .} | to_json' tools/ci/test-jobs.yml >> "$GITHUB_OUTPUT"
          else
              yq '{"include": [.[] | select(."cron-only" != true)]} | to_json' tools/ci/test-jobs.yml >> "$GITHUB_OUTPUT"
          fi
          echo 'EOT' >> "$GITHUB_OUTPUT"

  test:
    runs-on: ubuntu-latest
    needs: filter
    strategy:
      fail-fast: false
      matrix: ${{ fromJson(needs.filter.outputs.jobs) }}
    env: ${{ matrix.extra-envs }}
    continue-on-error: ${{ matrix.allow-failure || false }}
    steps:
      - name: Check out repository
        uses: actions/checkout@v4
        with:
          # Do full clone (~10 extra seconds) to fetch all tags.
          # Otherwise we might be missing the tags for maint PRs
          # whenever those maint releases were not yet merged into master.
          fetch-depth: 0

      - name: Set up Python
        uses: actions/setup-python@v5
        with:
          python-version: ${{ matrix.python-version }}

      - name: Show git describe output to ensure that we did fetch all the tags
        run: git describe

      # Just in case we need to check if nfs is there etc
      - run: sudo lsmod

      - name: Install dependencies
        run: |
          # The ultimate one-liner setup for NeuroDebian repository
          bash <(wget -q -O- http://neuro.debian.net/_files/neurodebian-travis.sh)
          sudo apt-get update -qq
          sudo apt-get install eatmydata  # to speedup some installations
          tools/ci/prep-travis-forssh.sh
          tools/ci/debians_disable_outdated_ssl_cert
          # Install various basic dependencies
          sudo eatmydata apt-get install zip pandoc p7zip-full
          # needed for tests of patool compression fall-back solution
          sudo eatmydata apt-get install xz-utils
          sudo eatmydata apt-get install shunit2
          [ -z "$LC_ALL" ] || sudo eatmydata apt-get install locales-all

      - name: Configure _DL_TMPDIR before installing git-annex
        run: |
          if [[ "${_DL_TMPDIR:-}" =~ .*/sym\ link ]]
          then echo "Symlinking $_DL_TMPDIR"
               ln -s /tmp "$_DL_TMPDIR"
          fi

          if [[ "${_DL_TMPDIR:-}" =~ .*/d\ i\ r ]]
          then echo "mkdir $_DL_TMPDIR"
               mkdir -p "$_DL_TMPDIR"
          fi

          if [[ "${_DL_TMPDIR:-}" =~ .*/nfsmount ]]
          then echo "mkdir $_DL_TMPDIR"
               mkdir -p "$_DL_TMPDIR" "${_DL_TMPDIR}_"
               echo "/tmp/nfsmount_ localhost(rw)" | sudo bash -c 'cat - > /etc/exports'
               sudo apt-get install -y nfs-kernel-server
               sudo exportfs -a
               sudo mount -t nfs localhost:/tmp/nfsmount_ /tmp/nfsmount
          fi

      - name: Install custom Git from upstream
        if: matrix.upstream-git
        run: |
          sudo apt-get install -y gettext libcurl4-gnutls-dev
          source tools/ci/install-upstream-git.sh
          echo "$target_dir/bin-wrappers" >> "$GITHUB_PATH"

      - name: Install minimum Git
        if: matrix.minimum-git
        run: |
          sudo apt-get install -y gettext libcurl4-gnutls-dev
          tools/ci/install-minimum-git.sh
          echo "$PWD/git-src/bin-wrappers" >> "$GITHUB_PATH"

      - name: Install git-annex
        run: |
          pip install datalad-installer
          eval datalad-installer --sudo ok -E new.env ${_DL_ANNEX_INSTALL_SCENARIO}
          # Append new.env to ~/.bashrc instead of $GITHUB_OUTPUT because it
          # can include `source` and `conda activate` commands that are invalid
          # for the latter; this also necessitates using an interactive shell
          # for the workflow.
          cat new.env >> ~/.bashrc

      - name: Install Python dependencies
        run: |
          pip install --upgrade pip
          pip install codecov
          pip install -r requirements-devel.txt

      - name: Configure Git
        run: |
          git config --global user.email "test@github.land"
          git config --global user.name "GitHub Almighty"
          # We do `sudo pip install` below, and versioneer needs to run git.
          # Recent git needs to be made certain it is safe to do
          sudo git config --global --add safe.directory $PWD

      - name: Configure sudoers
        # So we could test under sudo -E with PATH pointing to installed
        # location
        run: sudo sed -i -e 's/^Defaults.*secure_path.*$//' /etc/sudoers

      # TODO: remove - should not be needed
      - name: Git-annex workaround for NFS mounts
        run: |
          if [[ "${_DL_TMPDIR:-}" =~ .*/nfsmount ]]
          then sudo git config --system annex.pidlock true
          fi

      # Now it should be safe to point TMPDIR to a "tricky" setup just for the
      # purpose of testing
      - name: Point TMPDIR to "tricky" setup
        run: |
          if [ -n "${_DL_TMPDIR:-}" ]
          then echo TMPDIR="${_DL_TMPDIR}" >> "$GITHUB_ENV"
          fi

      - name: Test installation for user
        run: sudo pip install --user .

      - name: Report WTF information using system-wide installed version
        run: |
          datalad wtf
          python -m pip freeze

      - name: Run tests
        run: |
          PYTEST_OPTS=( -v )
          # If we requested to run only not slow (typically <10sec) tests, fail
          # if a test takes 3x more than that - it needs to get @slow or
          # @turtle annotation
          if echo "$PYTEST_SELECTION_OP($PYTEST_SELECTION)" | grep -q "^not.*slow"
          then
            PYTEST_OPTS=( "${PYTEST_OPTS[@]}" --doctest-modules --durations=0 --durations-min=5 --fail-slow 60 )
            export DATALAD_TESTS_SETUP_TESTREPOS=1
          fi

          mkdir -p __testhome__
          cd __testhome__
          # Note: adding --log-cli-level=INFO would result in
          # DATALAD_LOG_TARGET=/dev/null being not in effect, dumping too many
          # logs.
          set -x
          http_proxy=
          PATH=$PWD/../tools/coverage-bin:$PATH
          $PYTEST_WRAPPER python \
            -m pytest "${PYTEST_OPTS[@]}" \
            -c ../tox.ini \
            -n 2 \
            -m "${PYTEST_SELECTION:+$PYTEST_SELECTION_OP($PYTEST_SELECTION) and }not(turtle)" \
            --doctest-modules \
            --cov=datalad \
            --cov-report=xml \
            --pyargs ${TESTS_TO_PERFORM:-datalad}

      # Makes it only more difficult to comprehend the failing output.  Enable
      # only when necessary for a particular debugging.
      - name: Dump debug info
        #if: "failure()"  # Enabled
        if: "failure() && false"  # Disabled
        run: |
          if [ ! -z "$DATALAD_TESTS_NONETWORK" ]
          then sudo route add -net 0.0.0.0 netmask 0.0.0.0 dev lo
          fi

          DATALAD_LOG_LEVEL=DEBUG \
            $PYTEST_WRAPPER `which pytest` \
            -s -v --doctest-modules --cov datalad --log-cli-level=DEBUG

          if [ ! -z "$DATALAD_TESTS_NONETWORK" ]
          then sudo route del -net 0.0.0.0 netmask 0.0.0.0 dev lo
          fi

      # cron jobs test more and then PRs will be falling behind since they
      # would not trigger some codepaths.  So submit coverage only from
      # non-cron jobs, but report for all

      - name: Report coverage
        run: |
          python -m coverage debug sys
          python -m coverage report
        working-directory: __testhome__

      - name: Upload coverage to Codecov
        if: github.event_name != 'schedule'
        uses: codecov/codecov-action@v5
        with:
          directory: __testhome__
          fail_ci_if_error: false
          token: ${{ secrets.CODECOV_TOKEN }}

# vim:set et sts=2: