queuerPy 0.1.0__py3-none-any.whl

queuerPy/queuer_job.py

@@ -0,0 +1,600 @@
+"""
+Job-related methods for the Python queuer implementation.
+Mirrors Go's queuerJob.go functionality.
+"""
+
+import asyncio
+import logging
+from datetime import datetime
+from typing import Dict, List, Optional, Any, Tuple, Union, Callable, TYPE_CHECKING
+from uuid import UUID
+
+from core.broadcaster import BroadcasterQueue
+from core.runner import Runner, SmallRunner
+from core.retryer import Retryer
+from core.scheduler import Scheduler
+from helper.error import QueuerError
+from helper.task import get_task_name_from_interface
+from model.job import Job, JobStatus, new_job as create_job
+from model.options import Options
+from model.worker import Worker
+from queuer_global import QueuerGlobalMixin
+
+if TYPE_CHECKING:
+    from model.batch_job import BatchJob
+
+# Set up logger
+logger = logging.getLogger(__name__)
+
+
+class QueuerJobMixin(QueuerGlobalMixin):
+    """
+    Mixin class containing job-related methods for the Queuer.
+    This mirrors the job methods from Go's queuerJob.go.
+    """
+
+    def __init__(self):
+        super().__init__()
+
+    def add_job(self, task: Union[Callable[..., Any], str], *parameters: Any) -> Job:
+        """
+        Add a job to the queue with the given task and parameters.
+
+        :param task: Either a function or a string with the task name
+        :param parameters: Parameters to pass to the task
+        :returns: The created job
+        :raises Exception: If something goes wrong
+        """
+        options: Optional[Options] = self._merge_options(None)
+        job: Job = self._add_job(task, options, *parameters)
+
+        logger.info(f"Job added: {job.rid}")
+
+        return job
+
+    def add_job_with_options(
+        self,
+        options: Optional[Options],
+        task: Union[Callable[..., Any], str],
+        *parameters: Any,
+    ) -> Job:
+        """Add a new job with specific options."""
+        # Merge default options with provided options
+        options_merged: Optional[Options] = self._merge_options(options)
+
+        try:
+            # Create new job
+            new_job: Job = create_job(task, options_merged, *parameters)
+            job: Job = self.db_job.insert_job(new_job)
+
+            logger.info(f"Job with options added: {job.rid}")
+            return job
+        except Exception as e:
+            logger.error(f"Error adding job with options: {str(e)}")
+            raise Exception(f"Adding job: {str(e)}")
+
+    def add_jobs(self, batch_jobs: List["BatchJob"]):
+        """
+        Add a batch of jobs to the queue.
+
+        :param batch_jobs: List of BatchJob objects to add
+        :raises Exception: If something goes wrong during the process
+        """
+        jobs: List[Job] = []
+        for batch_job in batch_jobs:
+            options: Optional[Options] = self._merge_options(batch_job.options)
+            task_name: str = get_task_name_from_interface(batch_job.task)
+            job: Job = create_job(task_name, options, *batch_job.parameters)
+            jobs.append(job)
+
+        self.db_job.batch_insert_jobs(jobs)
+        logger.info(f"Jobs added: {len(jobs)}")
+
+    def wait_for_job_added(self, timeout_seconds: float = 30.0) -> Optional[Job]:
+        """
+        Wait for any job to start and return the job.
+        Uses the shared Runner event loop.
+
+        :param timeout_seconds: Maximum time to wait for a job to be added
+        :returns: The job that was added, or None if cancelled or timeout
+        """
+        runner = Runner(
+            task=self._wait_for_job_added_inner,
+            args=(timeout_seconds,),
+            kwargs={},
+        )
+        runner.go()
+        return runner.get_results(timeout=timeout_seconds)
+
+    async def _wait_for_job_added_inner(
+        self, timeout_seconds: float = 30.0
+    ) -> Optional[Job]:
+        """
+        Wait for any job to start and return the job.
+        Uses the broadcaster approach.
+
+        :param timeout_seconds: Maximum time to wait for a job to be added
+        :returns: The job that was added, or None if cancelled or timeout
+        :raises Exception: If broadcaster is not available
+        """
+        channel: BroadcasterQueue[Job] = await self.job_insert_broadcaster.subscribe()
+        try:
+            job = await asyncio.wait_for(channel.get(), timeout=timeout_seconds)
+            return job
+        except asyncio.TimeoutError:
+            logger.debug(f"Timed out waiting for job added ({timeout_seconds}s)")
+            return None
+        except Exception as e:
+            logger.error(f"Error waiting for job added: {e}")
+            return None
+        finally:
+            # Unsubscribe from broadcaster
+            await self.job_insert_broadcaster.unsubscribe(channel)
+
+    def wait_for_job_finished(
+        self, job_rid: UUID, timeout_seconds: float = 30.0
+    ) -> Optional[Job]:
+        """
+        Wait for a job to finish and return the job.
+        Uses SmallRunner for shared state access.
+
+        :param job_rid: The RID of the job to wait for
+        :param timeout_seconds: Maximum time to wait for job completion
+        :returns: The completed job, or None if not found or timeout
+        :raises Exception: If broadcaster is not available
+        """
+        # Add buffer to runner timeout to allow async method final check to complete
+        runner_timeout = timeout_seconds + 1.0
+        runner = SmallRunner(self._wait_with_listener, job_rid, timeout_seconds)
+        runner.go()
+        return runner.get_results(timeout=runner_timeout)
+
+    async def _wait_with_listener(
+        self, job_rid: UUID, timeout_seconds: float
+    ) -> Optional[Job]:
+        """
+        Wait for job completion using pure database notifications.
+
+        :param job_rid: The RID of the job to wait for
+        :param timeout_seconds: Maximum time to wait for job completion
+        :returns: The completed job, or None if not found or timeout
+        :raises Exception: If broadcaster is not available
+        """
+        try:
+            job_ended: Job = self.get_job_ended(job_rid)
+            if job_ended:
+                return job_ended
+        except Exception:
+            pass
+
+        channel: BroadcasterQueue[Job] = await self.job_delete_broadcaster.subscribe()
+        try:
+            while True:
+                job = await asyncio.wait_for(channel.get(), timeout=timeout_seconds)
+                if job.rid == job_rid:
+                    return job
+        except asyncio.TimeoutError:
+            logger.debug(f"Timed out waiting for job ended ({timeout_seconds}s)")
+            try:
+                job_ended: Job = self.get_job_ended(job_rid)
+                if job_ended:
+                    return job_ended
+            except Exception as e:
+                raise QueuerError("get job ended", e)
+        except Exception as e:
+            raise QueuerError("waiting for job", e)
+        finally:
+            await self.job_insert_broadcaster.unsubscribe(channel)
+
+    def cancel_job(self, job_rid: UUID) -> Job:
+        """
+         Cancel a job with the given job RID.
+
+        :param job_rid: The RID of the job to cancel
+        :returns: The cancelled job
+        :raises Exception: If the job is not found or already cancelled
+        """
+        job = self.db_job.select_job(job_rid)
+        if not job:
+            raise Exception(f"Job not found: {job_rid}")
+
+        self._cancel_job(job)
+        return job
+
+    def cancel_all_jobs_by_worker(self, worker_rid: UUID, entries: int = 100):
+        """
+        Cancel all jobs assigned to a specific worker by its RID.
+
+        :param worker_rid: The RID of the worker
+        :param entries: Maximum number of jobs to cancel
+        :raises Exception: If something goes wrong during the process
+        """
+        jobs = self.db_job.select_all_jobs_by_worker_rid(worker_rid, 0, entries)
+        for job in jobs:
+            self._cancel_job(job)
+
+    def readd_job_from_archive(self, job_rid: UUID) -> Job:
+        """
+        Readd a job from the archive back to the queue.
+
+        :param job_rid: The RID of the job to readd
+        :returns: The new job created from the archived job
+        :raises Exception: If the job is not found in archive
+        """
+        job = self.db_job.select_job_from_archive(job_rid)
+        if not job:
+            raise Exception(f"Job not found in archive: {job_rid}")
+
+        # Readd the job to the queue
+        new_job = self.add_job_with_options(job.options, job.task_name, *job.parameters)
+        logger.info(f"Job readded: {new_job.rid}")
+        return new_job
+
+    def get_job(self, job_rid: UUID) -> Job:
+        """
+        Retrieve a job by its RID.
+
+        :param job_rid: The RID of the job
+        :returns: The job
+        :raises Exception: If the job is not found
+        """
+        job = self.db_job.select_job(job_rid)
+        if not job:
+            raise Exception(f"Job not found: {job_rid}")
+        return job
+
+    def get_jobs(self, last_id: int = 0, entries: int = 100) -> List[Job]:
+        """
+        Retrieve all jobs in the queue.
+
+        :param last_id: Last job ID for pagination
+        :param entries: Number of entries to retrieve
+        :returns: List of jobs
+        """
+        return self.db_job.select_all_jobs(last_id, entries)
+
+    def get_job_ended(self, job_rid: UUID) -> Job:
+        """
+        Retrieve a job that has ended (succeeded, cancelled or failed) by its RID.
+
+        :param job_rid: The RID of the job
+        :returns: The ended job
+        :raises Exception: If the job is not found
+        """
+        job = self.db_job.select_job_from_archive(job_rid)
+        if not job:
+            raise Exception(f"Ended job not found: {job_rid}")
+        return job
+
+    def get_jobs_ended(self, last_id: int = 0, entries: int = 100) -> List[Job]:
+        """
+        Retrieve all jobs that have ended (succeeded, cancelled or failed).
+
+        :param last_id: Last job ID for pagination
+        :param entries: Number of entries to retrieve
+        :returns: List of ended jobs
+        """
+        return self.db_job.select_all_jobs_from_archive(last_id, entries)
+
+    def get_jobs_by_worker_rid(
+        self, worker_rid: UUID, last_id: int = 0, entries: int = 100
+    ) -> List[Job]:
+        """
+        Retrieve jobs assigned to a specific worker by its RID.
+
+        :param worker_rid: The RID of the worker
+        :param last_id: Last job ID for pagination
+        :param entries: Number of entries to retrieve
+        :returns: List of jobs assigned to the worker
+        """
+        return self.db_job.select_all_jobs_by_worker_rid(worker_rid, last_id, entries)
+
+    # Internal/Private methods
+
+    def _merge_options(self, options: Optional[Options]) -> Optional[Options]:
+        """
+        Merge the worker options with optional job options.
+
+        :param options: Job-specific options (can be None)
+        :returns: Merged options
+        """
+        worker_options = self.worker.options if self.worker else None
+        if options is not None and options.on_error is None:
+            options.on_error = worker_options
+        elif options is None and worker_options is not None:
+            options = Options(on_error=worker_options)
+
+        return options
+
+    def _add_job(
+        self,
+        task: Union[Callable[..., Any], str],
+        options: Optional["Options"],
+        *parameters: Any,
+    ) -> Job:
+        """
+        Add a job to the queue with all necessary parameters.
+
+        :param task: Either a function or a string with the task name
+        :param options: Job-specific options (can be None)
+        :param parameters: Parameters to pass to the task
+        :returns: The created job
+        :raises Exception: If something goes wrong
+        """
+        try:
+            new_job = create_job(task, options, *parameters)
+            job = self.db_job.insert_job(new_job)
+
+            return job
+
+        except Exception as e:
+            raise Exception(f"Creating or inserting job: {str(e)}")
+
+    def _run_job_initial(self):
+        """
+        Called to run the next job in the queue.
+        Updates job status to running with worker.
+        """
+        if not hasattr(self, "worker") or not getattr(self, "worker"):
+            logger.debug("No worker available, skipping job run")
+            return
+
+        # Check if database connection is available
+        if not self.db_job or not self.db_job.db or not self.db_job.db.instance:
+            logger.debug("Database connection unavailable, skipping job run")
+            return
+
+        worker: Worker = getattr(self, "worker")
+
+        logger.debug(f"Worker available_tasks: {worker.available_tasks}")
+        logger.debug(f"Worker max_concurrency: {worker.max_concurrency}")
+
+        jobs = self.db_job.update_jobs_initial(worker)
+        if not jobs:
+            logger.debug("No jobs found to run")
+            return
+
+        logger.info(f"Found {len(jobs)} jobs to run")
+
+        for job in jobs:
+            logger.info(f"Processing job: {job.rid}")
+            if (
+                job.options
+                and job.options.schedule
+                and job.options.schedule.start
+                and job.options.schedule.start > datetime.now()
+            ):
+                try:
+                    scheduler = Scheduler(
+                        self._run_job_sync,
+                        job.options.schedule.start,
+                        job,
+                    )
+                    scheduler.go()
+                except Exception as e:
+                    logger.error(f"Failed to schedule job with Scheduler: {e}")
+
+                logger.info(f"Scheduled job {job.rid} for {job.options.schedule.start}")
+            else:
+                logger.info(f"Running job immediately: {job.rid}")
+                runner = SmallRunner(self._run_job, job)
+                runner.go()
+
+    async def _wait_for_job(
+        self, job: Job
+    ) -> tuple[List[Any], bool, Optional[Exception]]:
+        """
+        Execute the job and return the results or an error.
+
+        :param job: The job to execute
+        :returns: A tuple containing the results, a cancellation flag, and an error if any
+        """
+        task_obj = self.tasks.get(job.task_name)
+        if not task_obj:
+            return [], False, Exception(f"Task not found: {job.task_name}")
+
+        # Extract the actual callable function from the Task object
+        task = task_obj.task if hasattr(task_obj, "task") else task_obj
+        if not callable(task):
+            return [], False, Exception(f"Task is not callable: {job.task_name}")
+
+        try:
+            parameters = getattr(job, "parameters", [])
+            runner = Runner(task, *parameters)
+
+            logger.info(f"Created runner for job {job.rid}")
+
+            if not hasattr(self, "active_runners"):
+                self.active_runners: Dict[UUID, Runner] = {}
+            self.active_runners[job.rid] = runner
+
+            try:
+                runner.go()
+                results = runner.get_results(timeout=300)
+                logger.debug(f"Runner task {job.task_name} completed: {results}")
+                return results, False, None
+
+            except asyncio.CancelledError:
+                if runner:
+                    runner.cancel()
+                return [], True, None
+            except Exception as e:
+                logger.error(f"Process runner task {job.task_name} failed: {e}")
+                return [], False, e
+            finally:
+                self.active_runners.pop(job.rid, None)
+
+        except Exception as e:
+            logger.error(f"Error setting up runner for task {job.task_name}: {e}")
+            return [], False, e
+
+    async def _retry_job(self, job: Job, job_error: Exception):
+        """
+        Retry the job with the given job error.
+
+        :param job: The job to retry
+        :param job_error: The error that occurred during the job execution
+        """
+        if (
+            not job.options
+            or not job.options.on_error
+            or job.options.on_error.max_retries <= 0
+        ):
+            await self._fail_job(job, job_error)
+            return
+
+        async def retry_function() -> Tuple[List[Job], bool]:
+            logger.debug(f"Trying/retrying job: {job.rid}")
+            results, cancelled, e = await self._wait_for_job(job)
+            if e:
+                raise QueuerError("retrying job", e)
+            return results, cancelled
+
+        # Create retryer and attempt retry
+        retryer = Retryer(retry_function, job.options.on_error)
+        error = await retryer.retry()
+        if error:
+            await self._fail_job(
+                job, QueuerError(f"Retrying job failed: {error}", job_error)
+            )
+        else:
+            results, _, _ = await self._wait_for_job(job)
+            await self._succeed_job(job, results)
+
+    async def _run_job(self, job: Job):
+        """
+        Run the job.
+
+        :param job: The job to run
+        """
+        logger.info(f"Running job: {job.rid}")
+
+        results, cancelled, error = await self._wait_for_job(job)
+        if cancelled:
+            return
+        elif error:
+            await self._retry_job(job, error)
+        else:
+            await self._succeed_job(job, results)
+
+    def _run_job_sync(self, job: Job):
+        """
+        Synchronous wrapper for _run_job that can be called by the Scheduler.
+        Mirrors the Go pattern where scheduler calls the function directly.
+
+        :param job: The job to run
+        """
+        try:
+            # Create a Runner process for the async job
+            runner = Runner(self._run_job, job)
+            runner.go()
+            logger.info(f"Scheduled job {job.rid} started in process {runner.pid}")
+
+        except Exception as e:
+            logger.error(f"Error executing scheduled job {job.rid}: {e}")
+
+    def _cancel_job(self, job: Job):
+        """
+        Cancel a job.
+
+        :param job: The job to cancel
+        """
+        if job.status in [JobStatus.RUNNING, JobStatus.SCHEDULED, JobStatus.QUEUED]:
+            # Cancel the running process if it exists
+            if hasattr(self, "active_runners") and job.rid in self.active_runners:
+                runner = self.active_runners[job.rid]
+                try:
+                    runner.cancel()
+                    logger.info(f"Cancelled running process for job: {job.rid}")
+                except Exception as e:
+                    logger.warning(
+                        f"Failed to cancel running process for job: {job.rid}: {e}"
+                    )
+
+            job.status = JobStatus.CANCELLED
+            try:
+                self.db_job.update_job_final(job)
+                logger.info(f"Job cancelled: {job.rid}")
+            except Exception as e:
+                logger.error(f"Error updating job status to cancelled: {e}")
+
+    async def _succeed_job(self, job: Job, results: List[Any]):
+        """
+        Update the job status to succeeded and run the next job if available.
+
+        :param job: The job that succeeded
+        :param results: The results from the job
+        """
+        job.status = JobStatus.SUCCEEDED
+        job.results = results
+        await self._end_job(job)
+
+    async def _fail_job(self, job: Job, job_error: Exception):
+        """
+        Update the job status to failed.
+
+        :param job: The job that failed
+        :param job_error: The error that occurred
+        """
+        job.status = JobStatus.FAILED
+        job.error = str(job_error)
+        await self._end_job(job)
+
+    async def _end_job(self, job: Job):
+        """
+        End a job and potentially schedule it again if it's a recurring job.
+
+        :param job: The job to end
+        """
+        if not self.db_job or not self.db_job.db or not self.db_job.db.instance:
+            logger.error(f"Database connection unavailable for job {job.rid}")
+            return
+
+        if job.worker_id != self.worker.id:
+            return
+
+        try:
+            ended_job = self.db_job.update_job_final(job)
+            logger.debug(f"Job ended: {ended_job.status} - {ended_job.rid}")
+
+            if (
+                ended_job.options
+                and ended_job.scheduled_at
+                and ended_job.options.schedule
+                and ended_job.options.schedule.interval
+                and ended_job.schedule_count < ended_job.options.schedule.max_count
+            ):
+                if ended_job.options.schedule.next_interval:
+                    # Use next interval function if available
+                    next_interval_func = self.next_interval_funcs.get(
+                        ended_job.options.schedule.next_interval
+                    )
+                    if not next_interval_func:
+                        logger.error(
+                            f"NextIntervalFunc not found: {ended_job.options.schedule.next_interval} for job {ended_job.rid}"
+                        )
+                        return
+                    new_scheduled_at = next_interval_func(
+                        ended_job.scheduled_at, ended_job.schedule_count
+                    )
+                else:
+                    # Use regular interval
+                    new_scheduled_at = ended_job.scheduled_at + (
+                        ended_job.schedule_count * ended_job.options.schedule.interval
+                    )
+
+                ended_job.scheduled_at = new_scheduled_at
+                ended_job.status = JobStatus.SCHEDULED
+
+                new_job = self.db_job.insert_job(ended_job)
+                logger.info(f"Job added for next iteration to the queue: {new_job.rid}")
+
+        except Exception as e:
+            logger.error(f"Error updating finished job: {e}")
+
+        # Try to run the next job in the queue, so if one job is finished another can directly start
+        try:
+            if self.db_job and self.db_job.db and self.db_job.db.instance:
+                self._run_job_initial()
+        except Exception as e:
+            logger.error(f"Error running next job: {e}")

queuerPy/queuer_listener.py

@@ -0,0 +1,73 @@
+"""
+Listener-related methods for the Python queuer implementation.
+Mirrors Go's queuerListener.go functionality.
+"""
+
+import logging
+from typing import Callable
+
+from queuer_global import QueuerGlobalMixin
+from model.job import Job
+
+logger = logging.getLogger(__name__)
+
+
+class QueuerListenerMixin(QueuerGlobalMixin):
+    """
+    Mixin class containing listener-related methods for the Queuer.
+    Mirrors Go's queuerListener.go functionality.
+    """
+
+    def __init__(self):
+        super().__init__()
+
+    def listen_for_job_update(self, notify_function: Callable[[Job], None]) -> None:
+        """
+        Listen for job update events and notify the provided function when a job is updated.
+
+        :param notify_function: Function to call when a job update event occurs
+        :raises: RuntimeError: If queuer is not running
+        """
+        if not self.running:
+            raise RuntimeError("Cannot listen with not running Queuer")
+
+        try:
+            # Start listening in the background
+            self.job_update_listener.listen(notify_function=notify_function)
+        except Exception as e:
+            logger.error(f"Error listening for job updates: {str(e)}")
+            raise RuntimeError(f"Failed to listen for job updates: {str(e)}")
+
+    def listen_for_job_delete(self, notify_function: Callable[[Job], None]) -> None:
+        """
+        Listen for job delete events and notify the provided function when a job is deleted.
+
+        :param notify_function: Function to call when a job delete event occurs
+        :raises: RuntimeError: If queuer is not running
+        """
+        if not self.running:
+            raise RuntimeError("Cannot listen with not running Queuer")
+
+        try:
+            # Start listening in the background
+            self.job_delete_listener.listen(notify_function=notify_function)
+        except Exception as e:
+            logger.error(f"Error listening for job deletes: {str(e)}")
+            raise RuntimeError(f"Failed to listen for job deletes: {str(e)}")
+
+    def listen_for_job_insert(self, notify_function: Callable[[Job], None]) -> None:
+        """
+        Listen for job insert events and notify the provided function when a job is inserted.
+
+        :param notify_function: Function to call when a job insert event occurs
+        :raises: RuntimeError: If queuer is not running
+        """
+        if not self.running:
+            raise RuntimeError("Cannot listen with not running Queuer")
+
+        try:
+            # Start listening in the background
+            self.job_insert_listener.listen(notify_function=notify_function)
+        except Exception as e:
+            logger.error(f"Error listening for job inserts: {str(e)}")
+            raise RuntimeError(f"Failed to listen for job inserts: {str(e)}")