ci.yml

name: nf-core CI
# This workflow runs the pipeline with the minimal test dataset to check that it completes without any syntax errors
on:
  push:
    branches:
      - dev
  pull_request:
  release:
    types: [published]

env:
  NXF_ANSI_LOG: false
  NXF_SINGULARITY_CACHEDIR: ${{ github.workspace }}/.singularity
  NXF_SINGULARITY_LIBRARYDIR: ${{ github.workspace }}/.singularity

concurrency:
  group: "${{ github.workflow }}-${{ github.event.pull_request.number || github.ref }}"
  cancel-in-progress: true

jobs:
  test:
    name: Run pipeline with test data
    # Only run on push if this is the nf-core dev branch (merged PRs)
    if: "${{ github.event_name != 'push' || (github.event_name == 'push' && github.repository == 'sanger-tol/ascc') }}"
    runs-on: ubuntu-latest
    strategy:
      matrix:
        NXF_VER:
          - "22.10.1"
          - "latest-everything"
    steps:
      - name: Get branch names
        # Pulls the names of current branches in repo
        # steps.branch-names.outputs.current_branch is used later and returns the name of the branch the PR is made FROM not to
        id: branch-names
        uses: tj-actions/branch-names@v8

      - name: Check out pipeline code
        uses: actions/checkout@v3

      - name: Install Nextflow
        uses: nf-core/setup-nextflow@v1
        with:
          version: "${{ matrix.NXF_VER }}"

      - name: Set up Singularity
        run: |
          mkdir -p $NXF_SINGULARITY_CACHEDIR
          mkdir -p $NXF_SINGULARITY_LIBRARYDIR

      - name: Setup apptainer
        uses: eWaterCycle/setup-apptainer@main

      - name: Install Python
        uses: actions/setup-python@v5
        with:
          python-version: "3.10"

      - name: Install nf-core
        run: |
          pip install nf-core

      - name: NF-Core Download - download singularity containers
        # Forcibly download repo on active branch and download SINGULARITY containers into the CACHE dir if not found
        # Must occur after singularity install or will crash trying to dl containers
        # Zip up this fresh download and run the checked out version
        run: |
          nf-core download sanger-tol/ascc --revision ${{ steps.branch-names.outputs.current_branch }} --compress none -d --force --outdir sanger-ascc --container-cache-utilisation amend --container-system singularity

      - name: Download test data
        # Download A fungal test data set that is full enough to show some real output.
        run: |
          curl https://tolit.cog.sanger.ac.uk/test-data/resources/ascc/asccTinyTest_V2.tar.gz | tar xzf -

      - name: Download the NCBI taxdump database
        run: |
          mkdir ncbi_taxdump
          curl -L https://ftp.ncbi.nih.gov/pub/taxonomy/new_taxdump/new_taxdump.tar.gz | tar -C ncbi_taxdump -xzf -

      - name: Download the FCS-gx database
        run: |
          mkdir FCS_gx
          wget -c https://ftp.ncbi.nlm.nih.gov/genomes/TOOLS/FCS/database/test-only/test-only.taxa.tsv -O FCS_gx/all.taxa.tsv
          wget -c https://ftp.ncbi.nlm.nih.gov/genomes/TOOLS/FCS/database/test-only/test-only.gxi -O FCS_gx/all.gxi
          wget -c https://ftp.ncbi.nlm.nih.gov/genomes/TOOLS/FCS/database/test-only/test-only.gxs -O FCS_gx/all.gxs
          wget -c https://ftp.ncbi.nlm.nih.gov/genomes/TOOLS/FCS/database/test-only/test-only.meta.jsonl -O FCS_gx/all.meta.jsonl
          wget -c https://ftp.ncbi.nlm.nih.gov/genomes/TOOLS/FCS/database/test-only/test-only.blast_div.tsv.gz -O FCS_gx/all.blast_div.tsv.gz

      - name: Download the BUSCO lineage database
        run: |
          mkdir busco_database
          curl -L https://tolit.cog.sanger.ac.uk/test-data/resources/busco/blobtoolkit.GCA_922984935.2.2023-08-03.lineages.tar.gz | tar -C busco_database -xzf -

      - name: Download the kraken2 database
        run: |
          mkdir kraken2
          curl -L https://raw.githubusercontent.com/nf-core/test-datasets/modules/data/genomics/sarscov2/genome/db/kraken2.tar.gz | tar -C kraken2 -xzf -

      - name: Download the subset of NT database
        run: |
          mkdir NT_database
          curl -L https://ftp.ncbi.nlm.nih.gov/blast/db/18S_fungal_sequences.tar.gz | tar -C NT_database -xzf -

      - name: Download the pacbio barcode
        run: |
          mkdir pacbio_barcode
          wget -O pacbio_barcode/SMRTbell_Barcoded_Adapter_Plate_3.0_bc2001-bc2096.fasta_.zip -c https://www.pacb.com/wp-content/uploads/SMRTbell_Barcoded_Adapter_Plate_3.0_bc2001-bc2096.fasta_.zip
          cd pacbio_barcode
          unzip SMRTbell_Barcoded_Adapter_Plate_3.0_bc2001-bc2096.fasta_.zip
          mv SMRTbell_Barcoded_Adapter_Plate_3.0_bc2001-bc2096.fasta pacbio_adaptors.fa
          rm -rf SMRTbell_Barcoded_Adapter_Plate_3.0_bc2001-bc2096.fasta_.zip __MACOSX
          cd ../

      - name: Download the subset of Diamond database
        run: |
          mkdir diamond
          wget -c https://tolit.cog.sanger.ac.uk/test-data/resources/diamond/UP000000212_1234679_tax.dmnd -O diamond/UP000000212_1234679_tax.dmnd

      - name: Download the vecscreen test data
        run: |
          mkdir vecscreen
          curl -L https://ftp.ncbi.nlm.nih.gov/blast/db/v4/16SMicrobial_v4.tar.gz | tar -C vecscreen -xzf -

      - name: Singularity - Run FULL pipeline with test data
        # TODO nf-core: You can customise CI pipeline run tests as required
        # For example: adding multiple test runs with different parameters
        # Remember that you can parallelise this by using strategy.matrix
        run: |
          nextflow run ${GITHUB_WORKSPACE} -profile test,singularity --outdir ./results --steps ALL