Merge pull request #249 from bento-platform/develop

Version 2.2.2

chord_metadata_service/chord/tests/constants.py

@@ -19,6 +19,7 @@
     "TEST_SEARCH_QUERY_6",
     "TEST_SEARCH_QUERY_7",
     "TEST_SEARCH_QUERY_8",
+    "TEST_SEARCH_QUERY_9",
     "TEST_FHIR_SEARCH_QUERY",
 ]
 
@@ -247,4 +248,5 @@ def valid_phenotypic_feature(biosample=None, phenopacket=None):
                        "URINARY BLADDER"]
 TEST_SEARCH_QUERY_7 = ["#eq", ["#resolve", "experiment_results", "[item]", "file_format"], "VCF"]
 TEST_SEARCH_QUERY_8 = ["#ico", ["#resolve", "experiment_type"], "chromatin"]
+TEST_SEARCH_QUERY_9 = ["#eq", ["#resolve", "subject", "id"], "patient:1"]
 TEST_FHIR_SEARCH_QUERY = {"query": {"match": {"gender": "FEMALE"}}}

chord_metadata_service/chord/tests/test_api_search.py

@@ -35,6 +35,7 @@
     TEST_SEARCH_QUERY_6,
     TEST_SEARCH_QUERY_7,
     TEST_SEARCH_QUERY_8,
+    TEST_SEARCH_QUERY_9,
     TEST_FHIR_SEARCH_QUERY,
 )
 from ..models import Project, Dataset, TableOwnership, Table
@@ -392,6 +393,17 @@ def test_private_search_11_experiment(self):
 
     # TODO table search fr experiments
 
+    def test_private_table_search_12(self):
+        # Valid query to search for subject id
+
+        r = self.client.post(reverse("private-table-search", args=[str(self.table.identifier)]), data=json.dumps({
+            "query": TEST_SEARCH_QUERY_9
+        }), content_type="application/json")
+        self.assertEqual(r.status_code, status.HTTP_200_OK)
+        c = r.json()
+        self.assertEqual(len(c["results"]), 1)
+        self.assertIn("patient:1", [phenopacket["subject"]["id"] for phenopacket in c["results"]])
+
     @patch('chord_metadata_service.chord.views_search.es')
     def test_fhir_search(self, mocked_es):
         mocked_es.search.return_value = SEARCH_SUCCESS

chord_metadata_service/chord/views_search.py

@@ -8,6 +8,7 @@
 from datetime import datetime
 from django.db import connection
 from django.conf import settings
+from django.views.decorators.cache import cache_page
 from psycopg2 import sql
 from rest_framework.decorators import api_view, permission_classes
 from rest_framework.permissions import AllowAny
@@ -235,6 +236,8 @@ def count_individual(ind):
 }
 
 
+# Cache page for the requested url
+@cache_page(60 * 60 * 2)
 @api_view(["GET"])
 @permission_classes([OverrideOrSuperUserOnly])
 def chord_table_summary(_request, table_id):
@@ -328,6 +331,8 @@ def search(request, internal_data=False):
     }, start))
 
 
+# Cache page for the requested url
+@cache_page(60 * 60 * 2)
 @api_view(["POST"])
 @permission_classes([AllowAny])
 def chord_search(request):
@@ -336,6 +341,8 @@ def chord_search(request):
 
 # Mounted on /private/, so will get protected anyway; this allows for access from federation service
 # TODO: Ugly and misleading permissions
+# Cache page for the requested url
+@cache_page(60 * 60 * 2)
 @api_view(["POST"])
 @permission_classes([AllowAny])
 def chord_private_search(request):
@@ -473,6 +480,8 @@ def chord_table_search(request, table_id, internal=False):
     return Response(len(query_results) > 0)
 
 
+# Cache page for the requested url
+@cache_page(60 * 60 * 2)
 @api_view(["POST"])
 @permission_classes([AllowAny])
 def chord_public_table_search(request, table_id):
@@ -482,6 +491,8 @@ def chord_public_table_search(request, table_id):
 
 # Mounted on /private/, so will get protected anyway; this allows for access from federation service
 # TODO: Ugly and misleading permissions
+# Cache page for the requested url
+@cache_page(60 * 60 * 2)
 @api_view(["POST"])
 @permission_classes([AllowAny])
 def chord_private_table_search(request, table_id):

chord_metadata_service/package.cfg

@@ -1,4 +1,4 @@
 [package]
 name = katsu
-version = 2.2.1
+version = 2.2.2
 authors = Ksenia Zaytseva, David Lougheed, Simon Chénard, Romain Grégoire

chord_metadata_service/phenopackets/search_schemas.py

@@ -49,15 +49,15 @@ def _tag_with_database_attrs(schema: dict, db_attrs: dict):
     "properties": {
         "id": {
             "search": {
-                **search_optional_eq(0, queryable="internal"),
+                **search_optional_eq(0),
                 "database": {
                     "field": models.Individual._meta.pk.column
                 }
             }
         },
         "alternate_ids": {
             "items": {
-                "search": search_optional_str(0, queryable="internal", multiple=True)
+                "search": search_optional_str(1, queryable="internal", multiple=True)
             },
             "search": {
                 "database": {
@@ -68,14 +68,14 @@ def _tag_with_database_attrs(schema: dict, db_attrs: dict):
         "date_of_birth": {
             # TODO: Internal?
             # TODO: Allow lt / gt
-            "search": search_optional_eq(1, queryable="internal")
+            "search": search_optional_eq(2, queryable="internal")
         },
         # TODO: Age
         "sex": {
-            "search": search_optional_eq(2)
+            "search": search_optional_eq(3)
         },
         "karyotypic_sex": {
-            "search": search_optional_eq(3)
+            "search": search_optional_eq(4)
         },
         "taxonomy": ONTOLOGY_SEARCH_SCHEMA,
     },
@@ -230,7 +230,7 @@ def _tag_with_database_attrs(schema: dict, db_attrs: dict):
     "properties": {
         "id": {
             "search": merge_schema_dictionaries(
-                search_optional_eq(0, queryable="internal"),
+                search_optional_eq(0),
                 {"database": {"field": models.Biosample._meta.pk.column}}
             )
         },
@@ -334,7 +334,12 @@ def _tag_with_database_attrs(schema: dict, db_attrs: dict):
 PHENOPACKET_SEARCH_SCHEMA = tag_schema_with_search_properties(schemas.PHENOPACKET_SCHEMA, {
     "properties": {
         "id": {
-            "search": {"database": {"field": models.Phenopacket._meta.pk.column}}
+            "search": {
+                **search_optional_eq(0),
+                "database": {
+                    "field": models.Phenopacket._meta.pk.column
+                }
+            }
         },
         "subject": merge_schema_dictionaries(
             INDIVIDUAL_SEARCH_SCHEMA,

chord_metadata_service/phenopackets/tests/constants.py

@@ -53,6 +53,9 @@
     "id": "patient:1",
     "date_of_birth": "1967-01-01",
     "sex": "MALE",
+    "age": {
+        "age": "P45Y"
+    },
     "extra_properties": {
         "education": "Bachelor's Degree"
     }
@@ -61,7 +64,15 @@
 VALID_INDIVIDUAL_2 = {
     "id": "patient:2",
     "date_of_birth": "1978-01-01",
-    "sex": "FEMALE"
+    "sex": "FEMALE",
+    "age": {
+        "start": {
+            "age": "P40Y"
+        },
+        "end": {
+            "age": "P45Y"
+        }
+    }
 }
 
 VALID_HTS_FILE = {

chord_metadata_service/restapi/api_views.py

@@ -71,8 +71,8 @@ def overview(_request):
     individuals_taxonomy = Counter()
     individuals_age = Counter()
     individuals_ethnicity = Counter()
-    individuals_extra_prop = {}
-    extra_prop_counter_dict = {}
+    # individuals_extra_prop = {}
+    # extra_prop_counter_dict = {}
 
     def count_individual(ind):
 
@@ -84,14 +84,15 @@ def count_individual(ind):
             individuals_ethnicity.update((ind.ethnicity,))
 
         # Generic Counter on all available extra properties
-        if ind.extra_properties:
-            for key in ind.extra_properties:
-                # Declare new Counter() if it's not delcared
-                if key not in extra_prop_counter_dict:
-                    extra_prop_counter_dict[key] = Counter()
-
-                extra_prop_counter_dict[key].update((ind.extra_properties[key],))
-                individuals_extra_prop[key] = dict(extra_prop_counter_dict[key])
+        # Comment out this count for now since it explodes the response
+        # if ind.extra_properties:
+        #     for key in ind.extra_properties:
+        #         # Declare new Counter() if it's not delcared
+        #         if key not in extra_prop_counter_dict:
+        #             extra_prop_counter_dict[key] = Counter()
+        #
+        #         extra_prop_counter_dict[key].update((ind.extra_properties[key],))
+        #         individuals_extra_prop[key] = dict(extra_prop_counter_dict[key])
 
         if ind.age is not None:
             individuals_age.update((parse_individual_age(ind.age),))
@@ -164,7 +165,7 @@ def count_object_fields(obj, container: dict):
                 "taxonomy": dict(individuals_taxonomy),
                 "age": dict(individuals_age),
                 "ethnicity": dict(individuals_ethnicity),
-                "extra_properties": dict(individuals_extra_prop),
+                # "extra_properties": dict(individuals_extra_prop),
             },
             "phenotypic_features": {
                 # count is a number of unique phenotypic feature types (not all pfs in the database)

chord_metadata_service/restapi/tests/test_api.py

@@ -58,6 +58,7 @@ def test_overview(self):
         # phenopackets
         self.assertEqual(response_obj['phenopackets'], 2)
         self.assertEqual(response_obj['data_type_specific']['individuals']['count'], 2)
+        self.assertIsInstance(response_obj['data_type_specific']['individuals']['age'], dict)
         self.assertEqual(response_obj['data_type_specific']['biosamples']['count'], 2)
         self.assertEqual(response_obj['data_type_specific']['phenotypic_features']['count'], 1)
         self.assertEqual(response_obj['data_type_specific']['diseases']['count'], 1)

chord_metadata_service/restapi/utils.py

@@ -50,13 +50,26 @@ def parse_duration(string):
 
 def parse_individual_age(age_obj):
     """ Parses two possible age representations and returns average age or age as integer. """
+    # AGE OPTIONS
+    # "age": {
+    #     "age": "P96Y"
+    # }
+    # AND
+    # "age": {
+    #     "start": {
+    #         "age": "P45Y"
+    #     },
+    #     "end": {
+    #         "age": "P49Y"
+    #     }
+    # }
     if 'start' in age_obj:
         start_age = parse_duration(age_obj['start']['age'])
         end_age = parse_duration(age_obj['end']['age'])
         # for the duration calculate the average age
         age = (start_age + end_age) // 2
-    elif isinstance(age_obj, str):
-        age = parse_duration(age_obj)
+    elif 'age' in age_obj:
+        age = parse_duration(age_obj['age'])
     else:
         raise ValueError(f"Error: {age_obj} format not supported")
     return age