Merge pull request #402 from broadinstitute/jb-ncbi-taxon-complete

Use fallback validation to OLS for taxon ID (SCP-6070)

Author: bistline

README.md

@@ -29,6 +29,14 @@ source env/bin/activate
 pip install -r requirements.txt
 source scripts/setup-mongo-dev.sh
 ```
+Note: if you encounter the following error:
+```
+AttributeError: module 'coverage' has no attribute 'types'
+```
+You can run the following to address this:
+```
+pip install --upgrade coverage
+```
 
 ### Docker
 

ingest/validation/ontologies/cl.min.tsv.gz

@@ -1 +1 @@
-1752088054 # validation cache key
+1764096355 # validation cache key

ingest/validation/ontologies/efo.min.tsv.gz

@@ -476,8 +476,17 @@ def retrieve_label_and_synonyms(
     :param property_type: attribute type for term (string, array, boolean)
     """
     ontology_name = re.split("[_:]", ontology_id)[0].lower()
-    if ontology_is_local(ontology_name):
+    # use OLS fallback for taxon IDs since minified version is not authoritative
+    use_fallback = ontology_name == "ncbitaxon"
+    if ontology_is_local(ontology_name) and not use_fallback:
         return minified_reader.find_ontology_entry(ontology_name, ontology_id, property_name)
+    elif ontology_is_local(ontology_name) and use_fallback:
+        try:
+            return minified_reader.find_ontology_entry(ontology_name, ontology_id, property_name)
+        except ValueError:
+            return retriever.retrieve_ontology_term_label_and_synonyms(
+                ontology_id, property_name, convention, property_type
+            )
     else:
         return retriever.retrieve_ontology_term_label_and_synonyms(
             ontology_id, property_name, convention, property_type

ingest/validation/ontologies/hancestro.min.tsv.gz

@@ -18,7 +18,7 @@ cellarium-cas[vis]==1.4.11
 
 # Dev dependencies
 pytest==7.2.1
-coverage==4.5.3
+coverage==7.12.0
 pytest-xdist==1.29.0
 pre-commit==1.18.1
 black==24.3.0