Merge branch 'track.pipeline' into 'develop'

Track.pipeline

See merge request !29

.gitlab-ci.yml

@@ -126,14 +126,14 @@ integration_se:
   script:
   - hostname
   - ulimit -a
-  - nextflow -bg run ./workflow/rna-seq.nf --deriva ./test_data/auth/credential.json --bdbag ./test_data/auth/cookies.txt --repRID 16-1ZX4
+  - nextflow -bg run ./workflow/rna-seq.nf --deriva ./test_data/auth/credential.json --bdbag ./test_data/auth/cookies.txt --repRID 16-1ZX4 --ci true
 
 integration_pe:
   stage: integration
   script:
   - hostname
   - ulimit -a
-  - nextflow -bg run ./workflow/rna-seq.nf --deriva ./test_data/auth/credential.json --bdbag ./test_data/auth/cookies.txt --repRID Q-Y5JA -with-dag dag.png
+  - nextflow -bg run ./workflow/rna-seq.nf --deriva ./test_data/auth/credential.json --bdbag ./test_data/auth/cookies.txt --repRID Q-Y5JA -with-dag dag.png --ci true
   artifacts:
     name: "$CI_JOB_NAME"
     when: always

workflow/conf/biohpc.config

@@ -3,6 +3,9 @@ process {
   queue = 'super'
   clusterOptions = '--hold'
 
+  withLabel: trackStart {
+    executor = 'local'
+  }
   withName: getBag {
     executor = 'local'
   }

workflow/rna-seq.nf

@@ -44,6 +44,29 @@ script_parseMeta = Channel.fromPath("${baseDir}/scripts/parseMeta.py")
 script_calculateTPM = Channel.fromPath("${baseDir}/scripts/calculateTPM.R")
 script_inferMeta = Channel.fromPath("${baseDir}/scripts/inferMeta.sh")
 
+/*
+ * trackStart: track start of pipeline
+ */
+params.ci = false
+process trackStart {
+  script:
+  """
+  hostname
+  ulimit -a
+  export https_proxy=\${http_proxy}
+  
+  curl -H 'Content-Type: application/json' -X PUT -d '{ \
+      "sessionId": "${workflow.sessionId}", \
+      "pipeline": "gudmap.rbk_rnaseq", \
+      "start": "${workflow.start}", \
+      "astrocyte": false, \
+      "status": "started", \
+      "nextflowVersion": "${workflow.nextflow.version}",
+      "ci": ${params.ci}}' \
+  "https://xku43pcwnf.execute-api.us-east-1.amazonaws.com/ProdDeploy/pipeline-tracking"
+  """
+ }
+
 /*
  * splitData: split bdbag files by replicate so fetch can occure in parallel, and rename files to replicate rid
  */
@@ -828,4 +851,4 @@ process dataQC {
       echo "echo \"LOG: running tin.py on \${i}\" >> ${repRID}.rseqc.err; tin.py -i ${repRID}.sorted.deduped.\${i}.bam  -r ./bed/genome.bed 1>>${repRID}.rseqc.log 2>>${repRID}.rseqc.err; cat ${repRID}.sorted.deduped.\${i}.tin.xls | tr -s \"\\w\" \"\\t\" | grep -P \\\"\\\\t\${i}\\\\t\\\";";
     done | parallel -j `nproc` -k 1>> ${repRID}.sorted.deduped.tin.xls 2>>${repRID}.rseqc.err
     """
-}
\ No newline at end of file
+}