Fix progress bar not shown without jobId

Before we did generate a jobId but never passed it to the algorithm call.
This lead to the client never being able to fetch progress unless the jobId was provided by the user

Author: FlorentinD

changelog.md

@@ -11,6 +11,7 @@
 
 * Fixed a bug, where sessions could not be created for AuraDB instances of tier `business-critical`.
 * Fixed a bug, where sessions would fail on write-back if the Graph was empty.
+* Fixed a bug, where the progress bar would not be shown unless the `jobId` parameter was set. The progress bar can be toggled on and off via `GraphDataScience::set_show_progress`.
 
 
 ## Improvements

graphdatascience/algo/algo_proc_runner.py

@@ -14,6 +14,7 @@ class AlgoProcRunner(IllegalAttrChecker, ABC):
     @graph_type_check
     def _run_procedure(self, G: Graph, config: dict[str, Any], with_logging: bool = True) -> DataFrame:
         params = CallParameters(graph_name=G.name(), config=config)
+        params.ensure_job_id_in_config()
 
         return self._query_runner.call_procedure(endpoint=self._namespace, params=params, logging=with_logging)
 
@@ -25,6 +26,7 @@ def estimate(self, G: Graph, **config: Any) -> "Series[Any]":
 
 class StreamModeRunner(AlgoProcRunner):
     def __call__(self, G: Graph, **config: Any) -> DataFrame:
+        # insert job id if not existing
         return self._run_procedure(G, config)
 
 

graphdatascience/algo/single_mode_algo_endpoints.py

@@ -9,6 +9,7 @@ class SingleModeAlgoEndpoints(CallerBase):
     @property
     @compatible_with("triangles", min_inclusive=ServerVersion(2, 5, 0))
     def triangles(self) -> StreamModeRunner:
+        # todo insert job id
         return StreamModeRunner(self._query_runner, f"{self._namespace}.triangles", self._server_version)
 
 

graphdatascience/call_parameters.py

@@ -1,9 +1,40 @@
 from collections import OrderedDict
-from typing import Any
+from typing import Any, Optional
+from uuid import uuid4
 
 
 class CallParameters(OrderedDict[str, Any]):
     # since Python 3.6 also initializing through CallParameters(**kwargs) is order preserving
 
     def placeholder_str(self) -> str:
         return ", ".join([f"${k}" for k in self.keys()])
+
+    def get_job_id(self) -> Optional[str]:
+        config = self["config"] if "config" in self else {}
+
+        job_id = None
+        if "jobId" in config:
+            job_id = config["jobId"]
+
+        if "job_id" in config:
+            job_id = config["job_id"]
+
+        if job_id is None:
+            return None
+
+        return str(job_id)
+
+    def ensure_job_id_in_config(self) -> str:
+        """
+        Ensure that a job id is present in the `config` parameter. If not, generate a new one.
+        This enables the client to check on the progress later on.
+        """
+        config = self.get("config")
+        if not config:
+            raise ValueError("config is not set in the parameters")
+
+        job_id = self.get_job_id()
+        if job_id is None:
+            job_id = str(uuid4())
+            config["jobId"] = job_id
+        return job_id

graphdatascience/graph/base_graph_proc_runner.py

@@ -238,6 +238,7 @@ def generate(self, graph_name: str, node_count: int, average_degree: int, **conf
         params = CallParameters(
             graph_name=graph_name, node_count=node_count, average_degree=average_degree, config=config
         )
+        params.ensure_job_id_in_config()
 
         result = self._query_runner.call_procedure(
             endpoint=self._namespace,
@@ -264,6 +265,8 @@ def filter(
             relationship_filter=relationship_filter,
             config=config,
         )
+        params.ensure_job_id_in_config()
+
         result = self._query_runner.call_procedure(
             endpoint=self._namespace,
             logging=True,

graphdatascience/query_runner/neo4j_query_runner.py

@@ -5,7 +5,6 @@
 import time
 import warnings
 from typing import Any, NamedTuple, Optional, Union
-from uuid import uuid4
 
 import neo4j
 from pandas import DataFrame
@@ -222,8 +221,8 @@ def call_procedure(
         def run_cypher_query() -> DataFrame:
             return self.run_cypher(query, params, database, custom_error)
 
-        if self._resolve_show_progress(logging):
-            job_id = self._extract_or_create_job_id(params)
+        job_id = None if not params else params.get_job_id()
+        if self._resolve_show_progress(logging) and job_id:
             return self._progress_logger.run_with_progress_logging(run_cypher_query, job_id, database)
         else:
             return run_cypher_query()
@@ -310,22 +309,6 @@ def create_graph_constructor(
     def set_show_progress(self, show_progress: bool) -> None:
         self._show_progress = show_progress
 
-    @staticmethod
-    def _extract_or_create_job_id(params: CallParameters) -> str:
-        config = params["config"] if "config" in params else {}
-
-        job_id = None
-        if "jobId" in config:
-            job_id = config["jobId"]
-
-        if "job_id" in config:
-            job_id = config["job_id"]
-
-        if not job_id:
-            return str(uuid4())
-
-        return str(job_id)
-
     @staticmethod
     def handle_driver_exception(session: neo4j.Session, e: Exception) -> None:
         reg_gds_hit = re.search(

graphdatascience/tests/unit/query_runner/test_neo4j_query_runner.py

@@ -1,3 +1,5 @@
+import pytest
+
 from graphdatascience.call_parameters import CallParameters
 
 
@@ -14,3 +16,47 @@ def test_empty_params() -> None:
     params = CallParameters()
 
     assert params.placeholder_str() == ""
+
+
+def test_get_job_id() -> None:
+    # empty params
+    params = CallParameters({})
+    job_id = params.get_job_id()
+    assert job_id is None
+
+    # empty job id
+    params = CallParameters(config={job_id: None})
+    job_id = params.get_job_id()
+    assert job_id is None
+
+    # job_id given
+    params = CallParameters(config={"job_id": "foo"})
+    job_id = params.get_job_id()
+    assert job_id == "foo"
+
+    # jobId given
+    params = CallParameters(config={"jobId": "bar"})
+    job_id = params.get_job_id()
+    assert job_id == "bar"
+
+
+def test_ensure_job_id() -> None:
+    # empty params
+    params = CallParameters({})
+    with pytest.raises(ValueError, match="config is not set"):
+        params.ensure_job_id_in_config()
+
+    # empty job id
+    params = CallParameters(config={"job_id": None})
+    job_id = params.ensure_job_id_in_config()
+    assert job_id is not None and job_id != ""
+
+    # job_id given
+    params = CallParameters(config={"job_id": "foo"})
+    job_id = params.ensure_job_id_in_config()
+    assert job_id == "foo"
+
+    # jobId given
+    params = CallParameters(config={"jobId": "bar"})
+    job_id = params.ensure_job_id_in_config()
+    assert job_id == "bar"