Merge pull request #126 from f-kretschmer/docker

Move to docker

.Rprofile

@@ -1 +1,5 @@
+options(renv.config.ppm.enabled = TRUE)
+options(renv.config.ppm.default = TRUE)
+options(repos = c(CRAN = "https://packagemanager.posit.co/cran/__linux__/bookworm/2024-02-06"))
+options(renv.config.repos.override = getOption("repos"))
 source("renv/activate.R")

.github/workflows/build_container.yaml

@@ -0,0 +1,27 @@
+name: Build and push docker container
+
+on:
+  push:
+    paths:
+      # - 'scripts/**' # whole scripts probably too much
+      - 'scripts/Python/requirements.txt'
+      - 'renv.lock'
+      - 'renv/**'
+      - 'Dockerfile'
+      - '.Rprofile'
+
+jobs:
+  dockerloginbuildpush:
+    runs-on: ubuntu-latest
+    steps:
+      - name: Login to GitHub Container Registry
+        uses: docker/login-action@v3
+        with:
+          registry: ghcr.io
+          username: ${{ github.actor }}
+          password: ${{ secrets.GITHUB_TOKEN }}
+      - uses: actions/checkout@v3
+      - name: Create image from repository
+        run: docker build -t ghcr.io/${{ github.actor }}/repo_rt_preprocessing .
+      - name: Push image to GitHub Container Registry
+        run: docker push ghcr.io/${{ github.actor }}/repo_rt_preprocessing:latest

.github/workflows/preprocess.yaml

@@ -11,10 +11,14 @@ on:
 jobs:
   preprocess:
     name: Preprocess raw data
-    # NOTE: on windows as computing of descriptors has a bug on linux right now
-    runs-on: windows-2019
-    env:
-      GITHUB_PAT: ${{ secrets.GITHUB_TOKEN }} # needed for pulling R packages from github
+    runs-on: ubuntu-latest
+    container:
+      image: ghcr.io/${{ github.actor }}/repo_rt_preprocessing:latest
+      env:
+        RENV_PATHS_LIBRARY: '/renv/library'
+    defaults:
+      run:
+        shell: bash
 
     steps:
     - name: Checkout fork repository
@@ -33,49 +37,14 @@ jobs:
     - name: Get new/changed datasets
       id: filesfolders
       shell: bash {0}
-      run: echo "files=$(for f in ${{ steps.files.outputs.all }}; do basename $(dirname $f); done | grep -E '^[0-9]+$' | sort | uniq | tr '\n' ' ')" >> $GITHUB_OUTPUT
+      run: echo "files=$(for f in ${{ steps.files.outputs.added_modified_renamed }}; do basename $(dirname $f); done | grep -E '^[0-9]+$' | sort | uniq | tr '\n' ' ')" >> $GITHUB_OUTPUT
 
     - name: List all added files
-      shell: bash {0}
       run: |
         for f in ${{ steps.filesfolders.outputs.files }}; do
           ls -lh raw_data/$f
         done
 
-    - name: Pyton dependencies
-      run: pip install -r scripts/Python/requirements.txt
-
-    - name: Setup java
-      uses: actions/setup-java@v3
-      with:
-        distribution: 'temurin'
-        java-version: '17'
-
-    - name: Set RENV_PATHS_ROOT
-      shell: bash
-      run: |
-        echo "RENV_PATHS_ROOT=${{ runner.temp }}/renv" >> $GITHUB_ENV
-
-    - name: Setup R
-      uses: r-lib/actions/setup-r@v2
-      with:
-        use-public-rspm: true
-        r-version: 4.2.2
-
-    - name: Restore Renv package cache
-      uses: actions/cache@v3
-      with:
-        path: ${{ env.RENV_PATHS_ROOT }}
-        key: ${{ runner.os }}-renv-${{ hashFiles('**/renv.lock') }}
-        restore-keys: |
-          ${{ runner.os }}-renv-
-
-    - name: Activate renv
-      shell: Rscript {0}
-      run: |
-        options(renv.config.mran.enabled = FALSE)
-        renv::restore()
-
     - name: Standardize compounds
       run: Rscript scripts/R_ci/compounds_standardize.R ${{ steps.filesfolders.outputs.files }}
 
@@ -117,6 +86,7 @@ jobs:
       run: |
         git config --global user.email '[email protected]'
         git config --global user.name 'Github Actions'
+        git config --global safe.directory '*'
         # Use LFS storage of main repository: no push access to fork LFS storage
         # TODO: change once repository is moved
         git config lfs.url 'https://github.com/f-kretschmer/pr_test_repo.git/info/lfs'

.github/workflows/preprocess_manual.yaml

@@ -35,66 +35,31 @@ on:
 jobs:
   preprocess:
     name: Preprocess raw data
-    # NOTE: on windows as computing of descriptors has a bug on linux right now
-    runs-on: windows-2019
-    env:
-      GITHUB_PAT: ${{ secrets.GITHUB_TOKEN }} # needed for pulling R packages from github
+    runs-on: ubuntu-latest
+    container:
+      image: ghcr.io/${{ github.actor }}/repo_rt_preprocessing:latest
+      env:
+        RENV_PATHS_LIBRARY: '/renv/library'
+    defaults:
+      run:
+        shell: bash
 
     steps:
     - name: Checkout repository
       uses: actions/checkout@v3
       with:
          lfs: true
 
-
     - name: List all files for selected datasets
-      shell: bash {0}
       run: |
         for f in ${{ inputs.datasets }}; do
           ls -lh raw_data/$f || true
           ls -lh processed_data/$f || true
         done
       continue-on-error: true
 
-    - name: Pyton dependencies
-      run: pip install -r scripts/Python/requirements.txt
-      # needed for every task
-
-    - name: Setup java
-      uses: actions/setup-java@v3
-      with:
-        distribution: 'temurin'
-        java-version: '17'
-      # needed for every task
-
-    - name: Set RENV_PATHS_ROOT
-      shell: bash
-      run: |
-        echo "RENV_PATHS_ROOT=${{ runner.temp }}/renv" >> $GITHUB_ENV
-      # needed for every task
-
-    - name: Setup R
-      uses: r-lib/actions/setup-r@v2
-      with:
-        use-public-rspm: true
-        r-version: 4.2.2
-      # needed for every task
-
-    - name: Restore Renv package cache
-      uses: actions/cache@v3
-      with:
-        path: ${{ env.RENV_PATHS_ROOT }}
-        key: ${{ runner.os }}-renv-${{ hashFiles('**/renv.lock') }}
-        restore-keys: |
-          ${{ runner.os }}-renv-
-      # needed for every task
-
-    - name: Activate renv
-      shell: Rscript {0}
-      run: |
-        options(renv.config.mran.enabled = FALSE)
-        renv::restore()
-      # needed for every task
+    - name: check renv
+      run: Rscript -e "renv::status()"
 
     - name: Standardize compounds
       run: Rscript scripts/R_ci/compounds_standardize.R ${{ inputs.datasets }}
@@ -145,8 +110,8 @@ jobs:
       run: |
         git config --global user.email '[email protected]'
         git config --global user.name 'Github Actions'
-        # Use LFS storage of main repository: no push access to fork LFS storage
-        # TODO: change once repository is moved
+        # because of dockerized environment, git will otherwise complain about "dubious ownership of directory"
+        git config --global safe.directory '*'
         git add processed_data raw_data
         git commit -m "Preprocessing ${{ inputs.datasets }}" -m "Tasks:
           - standardize compounds: ${{ inputs.do_standardize }}