Merge pull request #607 from adamnsch/gds-construct-correct-db

Use provided `database` for fetching metadata at `GraphDataScience` construction

Author: adamnsch

changelog.md

@@ -19,6 +19,7 @@
 
 * Fixed an issue where source and target IDs of relationships in heterogeneous OGBL graphs were not parsed correctly.
 * Fixed an issue where configuration parameters such as `aggregation` were ignored by `gds.graph.toUndirected`.
+* Fixed an issue where the `database` given for the `GraphDataScience` construction was not used for metadata retrieval, causing an exception to be raised if the default "neo4j" database was missing.
 
 
 ## Improvements

doc/modules/ROOT/pages/getting-started.adoc

@@ -38,6 +38,9 @@ assert gds.version()
 "2.1.9"
 ----
 
+Please note that the `GraphDataScience` object needs to communicate with a Neo4j database upon construction, and uses the default "neo4j" database by default.
+If there is no such database, you will need to <<specifying-targeted-database, provide a valid database using the `database` keyword parameter>>.
+
 
 === AuraDS
 
@@ -75,6 +78,7 @@ using_enterprise = gds.is_licensed()
 ----
 
 
+[[specifying-targeted-database]]
 === Specifying targeted database
 
 If we don't want to use the default database of our DBMS we can provide the `GraphDataScience` constructor with the keyword parameter `database`:
@@ -102,10 +106,10 @@ If Apache Arrow is available on the https://neo4j.com/docs/graph-data-science/cu
 [source,python,role=no-test]
 ----
 gds = GraphDataScience(
-  NEO4J_URI, 
-  auth=(NEO4J_USER, NEO4J_PASSWORD), 
-  arrow=True, 
-  arrow_disable_server_verification=False, 
+  NEO4J_URI,
+  auth=(NEO4J_USER, NEO4J_PASSWORD),
+  arrow=True,
+  arrow_disable_server_verification=False,
   arrow_tls_root_certs=CERT
 )
 ----

graphdatascience/query_runner/neo4j_query_runner.py

@@ -46,18 +46,20 @@ def create(
             driver = neo4j.GraphDatabase.driver(endpoint, auth=auth, **config)
 
             query_runner = Neo4jQueryRunner(
-                driver, auto_close=True, bookmarks=bookmarks, config=config, server_version=server_version
+                driver,
+                auto_close=True,
+                bookmarks=bookmarks,
+                config=config,
+                server_version=server_version,
+                database=database,
             )
 
         elif isinstance(endpoint, neo4j.Driver):
-            query_runner = Neo4jQueryRunner(endpoint, auto_close=False, bookmarks=bookmarks)
+            query_runner = Neo4jQueryRunner(endpoint, auto_close=False, bookmarks=bookmarks, database=database)
 
         else:
             raise ValueError(f"Invalid endpoint type: {type(endpoint)}")
 
-        if database:
-            query_runner.set_database(database)
-
         return query_runner
 
     @staticmethod
@@ -97,7 +99,7 @@ def run_cypher(
         if database is None:
             database = self._database
 
-        self._verify_connectivity()
+        self._verify_connectivity(database=database)
 
         with self._driver.session(database=database, bookmarks=self.bookmarks()) as session:
             try:
@@ -303,11 +305,14 @@ def handle_driver_exception(session: neo4j.Session, e: Exception) -> None:
 
         raise SyntaxError(generate_suggestive_error_message(requested_endpoint, all_endpoints)) from e
 
-    def _verify_connectivity(self) -> None:
+    def _verify_connectivity(self, database: Optional[str] = None) -> None:
         WAIT_TIME = 1
         MAX_RETRYS = 10 * 60
         WARN_INTERVAL = 10
 
+        if database is None:
+            database = self._database
+
         exception = None
         retrys = 0
         while retrys < MAX_RETRYS:
@@ -318,7 +323,16 @@ def _verify_connectivity(self) -> None:
                         category=neo4j.ExperimentalWarning,
                         message=r"^The configuration may change in the future.$",
                     )
-                self._driver.verify_connectivity()
+                else:
+                    warnings.filterwarnings(
+                        "ignore",
+                        category=neo4j.ExperimentalWarning,
+                        message=(
+                            r"^All configuration key-word arguments to verify_connectivity\(\) are experimental. "
+                            "They might be changed or removed in any future version without prior notice.$"
+                        ),
+                    )
+                self._driver.verify_connectivity(database=database)
                 break
             except neo4j.exceptions.DriverError as e:
                 exception = e

graphdatascience/tests/integration/test_database_ops.py

@@ -13,6 +13,23 @@
 GRAPH_NAME = "g"
 
 
+@pytest.mark.skip_on_aura
+def test_init_without_neo4j_db(runner: Neo4jQueryRunner) -> None:
+    default_database = runner.database()
+
+    MY_DB_NAME = "bananas"
+    runner.run_cypher("CREATE DATABASE $dbName WAIT", {"dbName": MY_DB_NAME})
+
+    runner.run_cypher("DROP DATABASE $dbName WAIT", {"dbName": default_database})
+
+    try:
+        gds = GraphDataScience(URI, AUTH, database=MY_DB_NAME)
+        gds.close()
+    finally:
+        runner.run_cypher("CREATE DATABASE $dbName WAIT", {"dbName": default_database}, database=MY_DB_NAME)
+        runner.run_cypher("DROP DATABASE $dbName WAIT", {"dbName": MY_DB_NAME})
+
+
 @pytest.mark.skip_on_aura
 def test_switching_db(runner: Neo4jQueryRunner) -> None:
     default_database = runner.database()

tox.ini

@@ -68,6 +68,7 @@ deps =
 	-r {toxinidir}/requirements/dev/notebook-ci.txt
 commands =
 	python ./scripts/run_notebooks.py
+
 [testenv:jupyter-notebook-session-ci]
 passenv =
 	AURA_API_CLIENT_ID