Merge pull request #252 from hammerlab/header-spec-to-extant-columns

header_spec -> extant_columns

cycledash/genotypes.py

@@ -20,10 +20,10 @@ def spec(vcf_id):
             ...},
     SAMPLE: {attrA:  ...}}
     """
-    query = "SELECT vcf_header, header_spec FROM vcfs WHERE id = %s"
+    query = "SELECT vcf_header, extant_columns FROM vcfs WHERE id = %s"
     with db.engine.connect() as connection:
         res = dict(connection.execute(query, (vcf_id,)).fetchall()[0])
-        spec = _vcf_header_spec(res['vcf_header'], res['header_spec'])
+        spec = _vcf_header_spec(res['vcf_header'], res['extant_columns'])
     return spec
 
 

schema.sql

@@ -9,7 +9,7 @@ CREATE TABLE vcfs (
        notes TEXT, -- Any notes, params, etc the user might add. Ideally in JSON format.
        uri TEXT UNIQUE, -- URI of source file, if any
        vcf_header TEXT, -- Plaintext header of the VCF
-       header_spec TEXT, -- list of non-null columns in the VCF
+       extant_columns TEXT, -- JSON list of non-null columns in the VCF
 
        precision NUMERIC,
        recall NUMERIC,

workers/genotype_extractor.py

@@ -37,7 +37,7 @@ def extractor(run):
     # (as this is a time-consuming operation) in the vcfs table for later use.
     extant_cols = json.dumps(extant_columns(metadata, connection, vcf_id))
     vcfs = metadata.tables.get('vcfs')
-    vcfs.update().where(vcfs.c.id == vcf_id).values(header_spec=extant_cols).execute()
+    vcfs.update().where(vcfs.c.id == vcf_id).values(extant_columns=extant_cols).execute()
 
     connection.close()
     return True