#!/usr/bin/env nextflow
/*
main.nf
*
* --------------------------------------------------------------------------
* Licensed under MIT (https://git.biohpc.swmed.edu/BICF/Astrocyte/cellranger_mkfastq/blob/develop/LICENSE)
* --------------------------------------------------------------------------
*
*/

//  ########  ####  ######  ######## 
//  ##     ##  ##  ##    ## ##       
//  ##     ##  ##  ##       ##       
//  ########   ##  ##       ######   
//  ##     ##  ##  ##       ##       
//  ##     ##  ##  ##    ## ##       
//  ########  ####  ######  ##       

// Define input variables
params.name = "run"
params.bcl = "${baseDir}/../test_data/simple1/*.tar.gz"
params.designFile = "${baseDir}/../test_data/single1/cellranger-tiny-bcl-simple-1_2_0.csv"
params.mask = ""
params.astrocyte = false
params.outDir = "${baseDir}/output"

// Define list of files
tarList = Channel
  .fromPath( params.bcl )
bclCount = Channel
  .fromPath( params.bcl )
  .count()

// Define regular variables
pipelineVersion = "2.1.5"
name = params.name
designLocation = Channel
  .fromPath(params.designFile)
  .ifEmpty { exit 1, "design file not found: ${params.designFile}" }
mask = params.mask
outDir = params.outDir

// Define script files
check_designScript = Channel.fromPath("$baseDir/scripts/check_design.py")
untarBCLScript = Channel.fromPath("$baseDir/scripts/untarBCL.sh")
countDesignScript = Channel.fromPath("$baseDir/scripts/countDesign.sh")
fastqcScript = Channel.fromPath("$baseDir/scripts/fastqc.sh")
versionsScript = Channel.fromPath("$baseDir/scripts/generate_versions.py")
referencesScript = Channel.fromPath("$baseDir/scripts/generate_references.py")
versions_pythonScript = Channel.fromPath("$baseDir/scripts/versions_python.sh")
//versions_pigzScript = Channel.fromPath("$baseDir/scripts/versions_pigz.sh")
versions_cellrangerScript = Channel.fromPath("$baseDir/scripts/versions_cellranger.sh")
versions_bcl2fastqScript = Channel.fromPath("$baseDir/scripts/versions_bcl2fastq.sh")
versions_fastqcScript = Channel.fromPath("$baseDir/scripts/versions_fastqc.sh")

// Define report files
multiqcConf = Channel.fromPath("${baseDir}/configs/multiqc_config.yaml")
references = Channel.fromPath("${baseDir}/../docs/references.md")


/*
 * trackStart: track start of pipeline
 */
params.ci = false
params.dev = false
/*
process trackStart {
  script:
  """
  hostname
  ulimit -a
  export https_proxy=\${http_proxy}
 
  curl -H 'Content-Type: application/json' -X PUT -d '{ \
      "sessionId": "${workflow.sessionId}", \
      "pipeline": "cellranger_mkfastq", \
      "pipelineVersion": "${pipelineVersion}", \
      "start": "${workflow.start}", \
      "astrocyte": ${params.astrocyte}, \
      "status": "started", \
      "nextflowVersion": "${workflow.nextflow.version}",
      "ci": ${params.ci},
      "dev": ${params.dev}}' \
  "https://xku43pcwnf.execute-api.us-east-1.amazonaws.com/ProdDeploy/pipeline-tracking"
  """
}
*/

process checkDesignFile {

  tag "${name}"

  input:
    file check_designScript
    file versions_pythonScript
    file designLocation

  output:
    file("design.checked.csv") into designPaths
    file("design.checked.csv") into designCount
    file("version_pipeline.txt") into version_pipeline
    file("version_nextflow.txt") into version_nextflow
    file("version_python.txt") into version_python

  script:
    """
    hostname
    ulimit -a
    noSpaceDesign=\$(echo "${designLocation}" | tr -d ' ')
    if [[ "\${noSpaceDesign}" != "${designLocation}" ]]; then
      mv "${designLocation}" "\${noSpaceDesign}"
    fi
    python3 check_design.py -d \${noSpaceDesign}
    bash versions_python.sh > version_python.txt
    echo "${workflow.nextflow.version}" > version_nextflow.txt
    echo "${pipelineVersion}" > version_pipeline.txt
    """

}
/* nextflow workflow manifest version calls that aren't compatible with Asrcocyte
    echo "${workflow.manifest.version}" > version_pipeline.txt
*/

process untarBCL {  

  tag "${tar.simpleName}"

  input:
    file untarBCLScript
    //file versions_pigzScript
    each file(tar) from tarList

  output:
    file("*[!version_pigz.txt]") into bclPaths mode flatten
    //file("version_pigz.txt") into version_pigz

  script:
    """
    hostname
    ulimit -u 16384
    ulimit -a
    bash untarBCL.sh -t ${tar}
    #bash versions_pigz.sh > version_pigz.txt
    """

}


process mkfastq {

  tag "${bcl.simpleName}"
  publishDir "${outDir}/${task.process}", mode: 'copy', pattern: "{*/outs/**/*.fastq.gz}"
  queue '128GB,256GB,256GBv1,384GB'

  input:
    file versions_cellrangerScript
    file versions_bcl2fastqScript
    each file(bcl) from bclPaths.collect()
    file design from designPaths

  output:
    file("fq/${bcl.simpleName}/*.fastq.gz") into fastqPaths
    val "${bcl.simpleName}" into bclName
    file("**/outs/**/*.fastq.gz") into cellrangerCount mode flatten
    file("**/outs/fastq_path/Stats/*") into bqcPaths
    file("version_cellranger.txt") into version_cellranger
    file("version_bcl2fastq.txt") into version_bcl2fastq

  script:
    """
    hostname
    ulimit -u 16384
    ulimit -a
    cellranger mkfastq --id=mkfastq_${bcl.simpleName} --run=${bcl} --csv=${design} --ignore-dual-index ${mask}
    mkdir fq
    mkdir "fq/${bcl.simpleName}"
    find . -name "*.fastq.gz" -exec cp {} fq/${bcl.simpleName}/ \\;
    bash versions_cellranger.sh > version_cellranger.txt
    bash versions_bcl2fastq.sh > version_bcl2fastq.txt
    """

}


if (bclCount.value == 1) {

  process countDesign {

    tag "${name}"
    publishDir "${outDir}/${task.process}/${name}", mode: 'copy'

    input:
      file countDesignScript
      file fastqs from cellrangerCount.collect()
      file design from designCount

    output:
      file("Cellranger_Count_Design.csv") into CountDesign

    script:
      """
      hostname
      ulimit -a
      bash countDesign.sh
      """

  }

}


process fastqc {

  tag "${bcl}"

  input:
    file fastqcScript    
    file versions_fastqcScript
    file(fastq) from fastqPaths
    each bcl from bclName

  output:
    file("*fastqc.zip") into fqcPaths mode flatten
    file("version_fastqc.txt") into version_fastqc

  script:
    """
    hostname
    ulimit -u 16384
    ulimit -a
    find *.fastq.gz -exec mv {} ${bcl}.{} \\;
    bash fastqc.sh
    bash versions_fastqc.sh > version_fastqc.txt
    """

}


process versions {

  tag "${name}"

  input:
  file versionsScript
  file referencesScript
  file version_pipeline
  file version_nextflow
  file version_python
  //file version_pigz
  file version_cellranger
  file version_bcl2fastq
  file version_fastqc
  file references

  output:
    file("*.yaml") into yamlPaths

  script:
    """
    hostname
    ulimit -a
    python3 generate_versions.py -f version_*.txt -o versions
    python3 generate_references.py -r ${references} -o references
    """

}


process multiqc {

  tag "${name}"
  publishDir "${outDir}/${task.process}/${name}", mode: 'copy', pattern: "{multiqc*}"

  input:
    file multiqcConf
    file bqc name "bqc/?/*" from bqcPaths.collect()
    file fqc name "fqc/?/*" from fqcPaths.collect()
    file yamlPaths

  output:
    file("multiqc_report.html") into mqcPaths

  script:
    """
    hostname
    ulimit -a
    #export LC_ALL=C.UTF-8
    #export LANG=C.UTF-8
    multiqc -c ${multiqcConf} .
    """

}