.gitlab-ci.yml

# Include shared CI
include:
    - project: "epi2melabs/ci-templates"
      file: "wf-containers.yaml"

variables:
    NF_BEFORE_SCRIPT: mkdir -p ${CI_PROJECT_NAME}/data/ && wget -O ${CI_PROJECT_NAME}/data/wf-16s-demo.tar.gz https://ont-exd-int-s3-euwst1-epi2me-labs.s3.amazonaws.com/wf-16s/wf-16s-demo.tar.gz && tar -xzvf ${CI_PROJECT_NAME}/data/wf-16s-demo.tar.gz -C ${CI_PROJECT_NAME}/data/
    NF_WORKFLOW_OPTS: "--fastq ${CI_PROJECT_NAME}/data/wf-16s-demo/test_data/
    --classifier minimap2
    --minimap2_by_reference
    --database_set ncbi_16s_18s"
    PYTEST_CONTAINER_NAME: "wf-metagenomics"
    NF_IGNORE_PROCESSES: "rebatchFastq"
    GIT_SUBMODULE_STRATEGY: recursive
    CI_FLAVOUR: "new"
    CWG_AWS_ENV_NAME: "stack"

aws-run:
    variables:
        NF_WORKFLOW_OPTS: "--fastq test_data/case01 --store_dir s3://$${XAWS_BUCKET}/${CI_PROJECT_NAME}/store"
        NF_IGNORE_PROCESSES: "rebatchFastq"
    artifacts:
        when: always
        paths:
            - ${CI_PROJECT_NAME}
            - .nextflow.log
        exclude: [] # give me everything pal
    allow_failure: false


docker-run:

    # Remove this directive in downstream templates
    tags:
        - large_ram

    # Define a 1D job matrix to inject a variable named MATRIX_NAME into
    #   the CI environment, we can use the value of MATRIX_NAME to determine
    #   which options to apply as part of the rules block below
    # NOTE There is a slightly cleaner way to define this matrix to include
    #   the variables, but it is broken when using long strings! See CW-756
    parallel:
        matrix:
            - MATRIX_NAME: [
                "kraken2", "kraken2-real-time", "minimap2", "minimap2-sample-sheet",
                "kraken2-bam", "minimap2-bam"]

    rules:
     - if: ($CI_COMMIT_BRANCH == null || $CI_COMMIT_BRANCH == "dev-template")
       when: never
     - if: $MATRIX_NAME == "kraken2"
       variables:
          NF_PROCESS_FILES: "wf-metagenomics/subworkflows/kraken_pipeline.nf"
          NF_WORKFLOW_OPTS: "--fastq test_data/case01 --classifier kraken2 --include_read_assignments"
          NF_IGNORE_PROCESSES: ""
     # In wf-metagenomics, the wf runs indefinitely if there is no condition to stop it.
     # With the read limit we can stop the wf if the limit is reached.
     # It creates a STOP.fastq.gz that will be a new input in the wf and make it stop.
     # Use the creation of this file to test if watch_path is able to detect new input files.
     - if: $MATRIX_NAME == "kraken2-real-time"
       variables:
          NF_PROCESS_FILES: "wf-metagenomics/subworkflows/real_time_pipeline.nf"
          NF_WORKFLOW_OPTS: "--fastq test_data/case01 --real_time --read_limit 3000 --include_read_assignments --batch_size 1000 --classifier kraken2"
          NF_IGNORE_PROCESSES: ""
     - if: $MATRIX_NAME == "minimap2"
       variables:
          NF_PROCESS_FILES: "wf-metagenomics/subworkflows/minimap_pipeline.nf"
          NF_WORKFLOW_OPTS: "--fastq test_data/case01 --minimap2_by_reference --keep_bam"
          NF_IGNORE_PROCESSES: "extractMinimap2Reads"
     - if: $MATRIX_NAME == "minimap2-sample-sheet"
       variables:
          NF_PROCESS_FILES: "wf-metagenomics/subworkflows/minimap_pipeline.nf"
          NF_WORKFLOW_OPTS: "--fastq test_data/case02 --sample_sheet test_data/case02/sample_sheet.csv --taxonomic_rank G --n_taxa_barplot 5 --abundance_threshold 1"
          NF_IGNORE_PROCESSES: "extractMinimap2Reads,getAlignmentStats"
    # BAM INGRESS
     # Compare counts with case01_no_duplicateIDs, must be the same
     - if: $MATRIX_NAME == "kraken2-bam"
       variables:
          NF_PROCESS_FILES: "wf-metagenomics/subworkflows/kraken_pipeline.nf"
          NF_WORKFLOW_OPTS: "--bam test_data/case05_bam --include_read_assignments --abundance_threshold 1 --classifier kraken2"
          NF_IGNORE_PROCESSES: ""
     ## Regular test minimap2 - mapping stats
     - if: $MATRIX_NAME == "minimap2-bam"
       variables:
          NF_PROCESS_FILES: "wf-metagenomics/subworkflows/minimap_pipeline.nf"
          NF_WORKFLOW_OPTS: "--bam test_data/case05_bam --minimap2_by_reference --database_set ncbi_16s_18s --classifier minimap2"
          NF_IGNORE_PROCESSES: "extractMinimap2Reads"