Add support for dependent jobs in enqueue_many (#1897)

* Add support for dependent jobs in enqueue_many

* Add module to register dependencies for multiple jobs

The get_ready_jobs function will process dependencies for an array of
jobs passed in. If any jobs' dependencies are already met, those jobs
are returned so they can be enqueued.

* Add check for jobs without dependencies

* Remove extra colon in dependencies key

This seems like a bug, but if I'm mistaken please let me know.

* Add bulk deferred jobs to Redis

Need to call queue.enqueue_job to create the job hash in redis. Since all of
these jobs are deferred, they won't be added to the queue and processed
by a worker.

* Revert "Remove extra colon in dependencies key"

This reverts commit 5ebf7a3500.

* Enqueue jobs without dependencies separately

Any jobs without dependencies will be enqueued before handling

* Fix enqueue_many return value

* Rename ready_jobs function

* Fix enqueue_many return value

* Instantiate job category arrays before if statement

* Execute pipe to enqueue jobs with met dependencies

* Add tests for enqueue_many with dependencies

* Change dependency sorting function name

* Use common kwargs dict to create jobs

* Remove redundant tests for dependent jobs

* Alphebetize imports

* Test job with met dependencies using enqueue_many

* Fix typo

* Format with black

* Sort imports

rq/dependency.py

@@ -0,0 +1,27 @@
+from typing import List
+
+from redis.client import Pipeline
+from redis.exceptions import WatchError
+
+from .job import Job
+
+
+class Dependency:
+    @classmethod
+    def get_jobs_with_met_dependencies(cls, jobs: List['Job'], pipeline: Pipeline):
+        jobs_with_met_dependencies = []
+        jobs_with_unmet_dependencies = []
+        for job in jobs:
+            while True:
+                try:
+                    pipeline.watch(*[Job.key_for(dependency_id) for dependency_id in job._dependency_ids])
+                    job.register_dependency(pipeline=pipeline)
+                    if job.dependencies_are_met(pipeline=pipeline):
+                        jobs_with_met_dependencies.append(job)
+                    else:
+                        jobs_with_unmet_dependencies.append(job)
+                    pipeline.execute()
+                except WatchError:
+                    continue
+                break
+        return jobs_with_met_dependencies, jobs_with_unmet_dependencies

rq/queue.py

@@ -20,6 +20,7 @@ if TYPE_CHECKING:
 
 from .connections import resolve_connection
 from .defaults import DEFAULT_RESULT_TTL
+from .dependency import Dependency
 from .exceptions import DequeueTimeout, NoSuchJobError
 from .job import Job, JobStatus
 from .logutils import blue, green
@@ -42,6 +43,7 @@ class EnqueueData(
             "ttl",
             "failure_ttl",
             "description",
+            "depends_on",
             "job_id",
             "at_front",
             "meta",
@@ -714,6 +716,7 @@ class Queue:
         ttl: Optional[int] = None,
         failure_ttl: Optional[int] = None,
         description: Optional[str] = None,
+        depends_on: Optional[List] = None,
         job_id: Optional[str] = None,
         at_front: bool = False,
         meta: Optional[Dict] = None,
@@ -733,6 +736,7 @@ class Queue:
             ttl (Optional[int], optional): Time to live. Defaults to None.
             failure_ttl (Optional[int], optional): Failure time to live. Defaults to None.
             description (Optional[str], optional): The job description. Defaults to None.
+            depends_on (Optional[JobDependencyType], optional): The job dependencies. Defaults to None.
             job_id (Optional[str], optional): The job ID. Defaults to None.
             at_front (bool, optional): Whether to enqueue the job at the front. Defaults to False.
             meta (Optional[Dict], optional): Metadata to attach to the job. Defaults to None.
@@ -752,6 +756,7 @@ class Queue:
             ttl,
             failure_ttl,
             description,
+            depends_on,
             job_id,
             at_front,
             meta,
@@ -772,33 +777,66 @@ class Queue:
             List[Job]: A list of enqueued jobs
         """
         pipe = pipeline if pipeline is not None else self.connection.pipeline()
-        jobs = [
-            self._enqueue_job(
-                self.create_job(
-                    job_data.func,
-                    args=job_data.args,
-                    kwargs=job_data.kwargs,
-                    result_ttl=job_data.result_ttl,
-                    ttl=job_data.ttl,
-                    failure_ttl=job_data.failure_ttl,
-                    description=job_data.description,
-                    depends_on=None,
-                    job_id=job_data.job_id,
-                    meta=job_data.meta,
-                    status=JobStatus.QUEUED,
-                    timeout=job_data.timeout,
-                    retry=job_data.retry,
-                    on_success=job_data.on_success,
-                    on_failure=job_data.on_failure,
-                ),
-                pipeline=pipe,
-                at_front=job_data.at_front,
+        jobs_without_dependencies = []
+        jobs_with_unmet_dependencies = []
+        jobs_with_met_dependencies = []
+
+        def get_job_kwargs(job_data, initial_status):
+            return {
+                "func": job_data.func,
+                "args": job_data.args,
+                "kwargs": job_data.kwargs,
+                "result_ttl": job_data.result_ttl,
+                "ttl": job_data.ttl,
+                "failure_ttl": job_data.failure_ttl,
+                "description": job_data.description,
+                "depends_on": job_data.depends_on,
+                "job_id": job_data.job_id,
+                "meta": job_data.meta,
+                "status": initial_status,
+                "timeout": job_data.timeout,
+                "retry": job_data.retry,
+                "on_success": job_data.on_success,
+                "on_failure": job_data.on_failure,
+            }
+
+        # Enqueue jobs without dependencies
+        job_datas_without_dependencies = [job_data for job_data in job_datas if not job_data.depends_on]
+        if job_datas_without_dependencies:
+            jobs_without_dependencies = [
+                self._enqueue_job(
+                    self.create_job(**get_job_kwargs(job_data, JobStatus.QUEUED)),
+                    pipeline=pipe,
+                    at_front=job_data.at_front,
+                )
+                for job_data in job_datas_without_dependencies
+            ]
+            if pipeline is None:
+                pipe.execute()
+
+        job_datas_with_dependencies = [job_data for job_data in job_datas if job_data.depends_on]
+        if job_datas_with_dependencies:
+            # Save all jobs with dependencies as deferred
+            jobs_with_dependencies = [
+                self.create_job(**get_job_kwargs(job_data, JobStatus.DEFERRED))
+                for job_data in job_datas_with_dependencies
+            ]
+            for job in jobs_with_dependencies:
+                job.save(pipeline=pipe)
+            if pipeline is None:
+                pipe.execute()
+
+            # Enqueue the jobs whose dependencies have been met
+            jobs_with_met_dependencies, jobs_with_unmet_dependencies = Dependency.get_jobs_with_met_dependencies(
+                jobs_with_dependencies, pipeline=pipe
             )
-            for job_data in job_datas
-        ]
-        if pipeline is None:
-            pipe.execute()
-        return jobs
+            jobs_with_met_dependencies = [
+                self._enqueue_job(job, pipeline=pipe, at_front=job.enqueue_at_front)
+                for job in jobs_with_met_dependencies
+            ]
+            if pipeline is None:
+                pipe.execute()
+        return jobs_without_dependencies + jobs_with_unmet_dependencies + jobs_with_met_dependencies
 
     def run_job(self, job: 'Job') -> Job:
         """Run the job

tests/test_dependencies.py

@@ -103,6 +103,26 @@ class TestDependencies(RQTestCase):
 
         self.assertEqual(q.job_ids, ["fake_job_id_2", "fake_job_id_1"])
 
+    def test_multiple_jobs_with_dependencies(self):
+        """Enqueue dependent jobs only when appropriate"""
+        q = Queue(connection=self.testconn)
+        w = SimpleWorker([q], connection=q.connection)
+
+        # Multiple jobs are enqueued with correct status
+        parent_job = q.enqueue(say_hello)
+        job_no_deps = Queue.prepare_data(say_hello)
+        job_with_deps = Queue.prepare_data(say_hello, depends_on=parent_job)
+        jobs = q.enqueue_many([job_no_deps, job_with_deps])
+        self.assertEqual(jobs[0].get_status(), JobStatus.QUEUED)
+        self.assertEqual(jobs[1].get_status(), JobStatus.DEFERRED)
+        w.work(burst=True, max_jobs=1)
+        self.assertEqual(jobs[1].get_status(), JobStatus.QUEUED)
+
+        job_with_met_deps = Queue.prepare_data(say_hello, depends_on=parent_job)
+        jobs = q.enqueue_many([job_with_met_deps])
+        self.assertEqual(jobs[0].get_status(), JobStatus.QUEUED)
+        q.empty()
+
     def test_dependency_list_in_depends_on(self):
         """Enqueue with Dependency list in depends_on"""
         q = Queue(connection=self.testconn)