Fix parsing analysis workflow id (#55)

Get the first subworkflow id from the
analysis output path, which corresponds
to the analysis workflow id.

Author: samanehsan

pipeline_tools/get_analysis_metadata.py

@@ -16,9 +16,8 @@ def get_workflow_id(analysis_output_path):
         workflow_id (str): string giving Cromwell UUID of the workflow.
     """
     url = analysis_output_path
-    hash_end = url.rfind("/call-")
-    hash_start = url.rfind('/', 0, hash_end) + 1
-    workflow_id = url[hash_start:hash_end]
+    calls = url.split('/call-')
+    workflow_id = calls[1].split('/')[-1]
     with open('workflow_id.txt', 'w') as f:
         f.write(workflow_id)
     return workflow_id

pipeline_tools/tests/test_get_analysis_metadata.py

@@ -19,7 +19,6 @@ def setUp(self):
         self.cromwell_url = '{}/{}/metadata?expandSubWorkflows=true'.format(base_url, self.workflow_id)
         self.caas_url = '{}/{}/metadata?expandSubWorkflows=true'.format(caas_base_url, self.workflow_id)
 
-    @unittest.skip('Skipping until get_worflow_id correctly parses the analysis subworkflow id')
     def test_get_workflow_id(self):
         analysis_output_path = 'gs://broad-dsde-mint-dev-cromwell-execution/cromwell-executions/AdapterSmartSeq2SingleCell/workflow_id/call-analysis/SmartSeq2SingleCell/analysis_subworkflow_id/call-qc/RunHisat2Pipeline/qc_workflow_id/call-Hisat2/12345_qc.hisat2.met.txt'
         result = get_analysis_metadata.get_workflow_id(analysis_output_path)