combiner setup fixes (#963)

cpg_workflows/stages/large_cohort.py

@@ -3,7 +3,7 @@
 from typing import TYPE_CHECKING, Any, Tuple
 
 from cpg_utils import Path
-from cpg_utils.config import config_retrieve, get_config, image_path
+from cpg_utils.config import config_retrieve, genome_build, get_config, image_path
 from cpg_utils.hail_batch import get_batch, query_command
 from cpg_workflows.targets import Cohort, SequencingGroup
 from cpg_workflows.utils import slugify
@@ -63,26 +63,24 @@ def queue_jobs(self, cohort: Cohort, inputs: StageInput) -> StageOutput | None:
             f"{self.tmp_prefix}/{workflow_config['cohort']}-{workflow_config['sequencing_type']}-{combiner_config['vds_version']}",
         )
 
-        vds_paths: list[str] | None = None
-        sg_ids_in_vds: list[str] | None = None
-        new_sg_gvcfs: list[str] | None = None
+        # create these as empty lists instead of None, they have the same truthiness
+        vds_paths: list[str] = []
+        sg_ids_in_vds: list[str] = []
 
         if combiner_config.get('vds_analysis_ids', None) is not None:
-            vds_paths = []
-            sg_ids_in_vds = []
             for vds_id in combiner_config['vds_analysis_ids']:
                 tmp_query_res, tmp_sg_ids_in_vds = self.get_vds_ids_output(vds_id)
                 vds_paths.append(tmp_query_res)
                 sg_ids_in_vds = sg_ids_in_vds + tmp_sg_ids_in_vds
 
         # Get SG IDs from the cohort object itself, rather than call Metamist.
         # Get VDS IDs first and filter out from this list
-        sg_ids: list[SequencingGroup] = cohort.get_sequencing_groups(only_active=True)
-        new_sg_gvcfs = [str(sg.gvcf) for sg in sg_ids if sg_ids_in_vds and sg.id not in sg_ids_in_vds]
-        if len(new_sg_gvcfs) == 0:
-            new_sg_gvcfs = None
-            if not vds_paths or len(vds_paths) == 1:
-                return self.make_outputs(cohort, self.expected_outputs(cohort))
+        cohort_sgs: list[SequencingGroup] = cohort.get_sequencing_groups(only_active=True)
+
+        new_sg_gvcfs: list[str] = [str(sg.gvcf) for sg in cohort_sgs if sg.id not in sg_ids_in_vds]
+
+        if len(new_sg_gvcfs) == 0 and len(vds_paths) <= 1:
+            return self.make_outputs(cohort, self.expected_outputs(cohort))
 
         j.image(image_path('cpg_workflows'))
         j.memory(combiner_config['memory'])
@@ -94,8 +92,8 @@ def queue_jobs(self, cohort: Cohort, inputs: StageInput) -> StageOutput | None:
             output_vds_path=str(output_vds_path),
             sequencing_type=workflow_config['sequencing_type'],
             tmp_prefix=tmp_prefix,
-            genome_build=workflow_config['workflow'].get('genome_build', 'GRCh38'),
-            gvcf_paths=new_sg_gvcfs,
+            genome_build=genome_build(),
+            gvcf_paths=new_sg_gvcfs,  # this is a list or None, and in new_combiner None is made into []
             vds_paths=vds_paths,
         )