ES Docs: allow to ptovide a custom id property

app/blueprints/es_proxy/controllers/es_proxy_controller.py

@@ -63,7 +63,8 @@ def get_es_doc(index_name, doc_id):
     """
     try:
         source = request.args.get('source')
-        json_response = es_proxy_service.get_es_doc(index_name, doc_id, source)
+        custom_id_property = request.args.get('custom_id_property')
+        json_response = es_proxy_service.get_es_doc(index_name, doc_id, source, custom_id_property)
         http_response = jsonify(json_response)
         http_cache_utils.add_cache_headers_to_response(http_response)
         return http_response

app/blueprints/es_proxy/controllers/marshmallow_schemas.py

@@ -23,3 +23,4 @@ class ESProxyDoc(CommonSchema):
     index_name = fields.String(required=True)
     doc_id = fields.String(required=True)
     source = fields.String()
+    custom_id_property = fields.String()

app/blueprints/es_proxy/services/es_proxy_service.py

@@ -62,16 +62,36 @@ def get_es_data(index_name, raw_es_query, raw_context, raw_contextual_sort_data,
     return response
 
 
-def get_es_doc(index_name, doc_id, source=None):
+def get_es_doc(index_name, doc_id, source=None, custom_id_property=None):
     """
     :param index_name: name of the index to which the doc belongs
     :param doc_id: id of the document to search for
     :param source: list of properties to get
+    :param custom_id_property: custom id property to use instead of _id
     :return: the json response from elasticsearch of the document
     """
 
     try:
-        response = es_data.get_es_doc(index_name, doc_id, source)
+        item_id = doc_id
+        use_custom_id_property = custom_id_property is not None
+        if use_custom_id_property:
+            es_query = {
+                "_source": source,
+                "query": {
+                    "terms": {
+                        custom_id_property: [
+                            doc_id
+                        ]
+                    }
+                }
+            }
+            search_response = es_data.get_es_response(index_name, es_query)
+            hits = search_response.get('hits', {}).get('hits',[])
+            if len(hits) == 0:
+                raise ESDataNotFoundError(f'No item found in index {index_name} with {custom_id_property} == {doc_id}')
+            item_id = hits[0]['_id']
+
+        response = es_data.get_es_doc(index_name, item_id, source)
         return response
     except es_data.ESDataNotFoundError as error:
         raise ESDataNotFoundError(str(error))

app/es_data/es_data.py

@@ -113,7 +113,7 @@ def get_es_doc(index_name, doc_id, source=None):
     """
     properties_list = [] if source is None else source.split(',')
     using_source = source is not None
-    cache_key = f'document-{index_name}-{doc_id}-{",".join(properties_list)}-{RUN_CONFIG.get("cache_key_suffix")}'
+    cache_key = f'document-{index_name}-{doc_id}-source:{",".join(properties_list)}-{RUN_CONFIG.get("cache_key_suffix")}'
     app_logging.debug(f'cache_key: {cache_key}')
 
     equivalent_query = {

app/swagger/swagger.yaml

@@ -99,6 +99,12 @@ paths:
           required: false
           type: 'string'
           default: 'molecule_chembl_id,pref_name'
+        - name: "custom_id_property"
+          in: "query"
+          description: "By default the doc_id will match with the _id field,
+          but you can pass another field here if you prefer"
+          type: 'string'
+          default: 'target_chembl_id'
       responses:
         '200':
           description: "success"

functional_tests/run_functional_tests.py

@@ -7,7 +7,8 @@ import argparse
 
 from specific_tests import fun_test_simple_query, fun_test_query_with_context, \
     fun_test_query_with_search_by_ids_context, fun_test_property_config, \
-    fun_test_group_config, fun_test_facets_group_config, fun_test_get_document, fun_test_id_properties, \
+    fun_test_group_config, fun_test_facets_group_config, fun_test_get_document, fun_test_get_document_by_custom_id_prop, \
+    fun_test_id_properties, \
     fun_test_get_context_data, fun_test_search_parsing, fun_test_url_shortening, fun_test_element_usage, \
     fun_test_go_slim_target_classification, fun_test_in_vivo_assay_classification, \
     fun_test_organism_taxonomy_target_classification, fun_test_protein_target_classification, \
@@ -34,6 +35,7 @@ def run():
     for test_module in [fun_test_simple_query, fun_test_query_with_context,
                         fun_test_query_with_search_by_ids_context, fun_test_property_config,
                         fun_test_group_config, fun_test_facets_group_config, fun_test_get_document,
+                        fun_test_get_document_by_custom_id_prop
                         fun_test_id_properties,
                         fun_test_get_context_data, fun_test_search_parsing, fun_test_url_shortening,
                         fun_test_element_usage,

functional_tests/specific_tests/fun_test_get_document_by_custom_id_prop.py

+# pylint: disable=import-error,unused-argument
+"""
+Module that tests a group config
+"""
+from specific_tests import utils
+
+
+def run_test(server_base_url, delayed_jobs_server_base_path):
+    """
+    Tests getting the configuration of a group
+    :param server_base_url: base url of the running server. E.g. http://127.0.0.1:5000
+    :param delayed_jobs_server_base_path: base path for the delayed_jobs
+    """
+
+    print('-------------------------------------------')
+    print('Testing getting a document by id')
+    print('-------------------------------------------')
+
+    url = f'{server_base_url}/es_data/get_es_document/chembl_eubopen_target/CHEMBL4524126'
+    utils.assert_get_request_succeeds(url)
+
+    params = {
+        'source': 'pref_name',
+        'custom_id_property': 'target_chembl_id'
+    }
+
+    utils.assert_get_request_succeeds(url, params)
\ No newline at end of file