refactor: remove jobs-table dependencies from Python SDK

src/galileo/experiment.py

@@ -3,17 +3,20 @@
 import builtins
 import datetime
 import re
+import warnings
 from collections.abc import Iterator
+from time import sleep
 from typing import TYPE_CHECKING, Any
 
+from tqdm.auto import tqdm
+
 from galileo.config import GalileoPythonConfig
 from galileo.datasets import Dataset as LegacyDataset
 from galileo.exceptions import NotFoundError
 from galileo.experiment_tags import upsert_experiment_tag
 from galileo.experiments import Experiments as ExperimentsService
 from galileo.experiments import _default_prompt_settings
 from galileo.export import ExportClient
-from galileo.job_progress import get_run_scorer_jobs, job_progress
 from galileo.prompts import PromptTemplate, get_prompt
 from galileo.resources.api.experiment import (
     delete_experiment_projects_project_id_experiments_experiment_id_delete,
@@ -1122,54 +1125,68 @@ def get_status(self) -> ExperimentStatusInfo:
 
         return ExperimentStatusInfo(self._experiment_response)
 
-    def monitor_progress(self, job_id: str | None = None) -> str:
+    def monitor_progress(self, poll_interval_seconds: float = 2.0, *, job_id: str | None = None) -> None:
         """
-        Monitor the progress of the experiment job with a progress bar.
+        Monitor the progress of the experiment with a progress bar.
 
-        Args:
-            job_id: Optional job ID to monitor. If not provided, will attempt to find
-                   the primary job for this experiment.
+        Polls the experiment status via the API until the experiment completes,
+        displaying a tqdm progress bar reflecting `log_generation` progress.
+
+        Parameters
+        ----------
+        poll_interval_seconds : float, optional
+            Seconds to wait between status polls. Defaults to 2.0.
+        job_id : str or None, optional
+            Deprecated. This parameter is ignored; it existed in a prior version
+            that polled the jobs table, which has been retired.
 
         Returns
         -------
-            str: The unique identifier of the completed job.
+        None
 
         Raises
         ------
-            ValueError: If the experiment lacks required id or project_id attributes,
-                       or if no job_id is provided and no job can be found.
+        ValueError
+            If the experiment lacks required id or project_id attributes.
 
         Examples
         --------
-            experiment = Experiment.get(name="ml-evaluation", project_name="My AI Project")
-            result = experiment.run()
+            experiment = Experiment(
+                name="ml-evaluation",
+                dataset_name="ml-dataset",
+                project_name="My AI Project"
+            ).create()
 
-            # Monitor the job progress
-            completed_job_id = experiment.monitor_progress()
+            experiment.monitor_progress()
         """
+        if job_id is not None:
+            warnings.warn(
+                "The 'job_id' parameter of monitor_progress() is deprecated and will be removed in a future release. "
+                "Progress is now tracked directly via experiment status; the job_id value is ignored.",
+                DeprecationWarning,
+                stacklevel=2,
+            )
+
         if self.id is None:
             raise ValueError("Experiment ID is not set. Cannot monitor progress for a local-only experiment.")
         if self.project_id is None:
             raise ValueError("Project ID is not set. Cannot monitor progress without project_id.")
 
-        if job_id is None:
-            # Try to get job from stored state or query for it
-            if self._job_id:
-                job_id = self._job_id
-            else:
-                # Get the first scorer job
-                scorer_jobs = get_run_scorer_jobs(project_id=self.project_id, run_id=self.id)
-                if not scorer_jobs:
-                    raise ValueError("No job found for this experiment. Run the experiment first.")
-                job_id = str(scorer_jobs[0].id)
-
-        _logger.info(f"Experiment.monitor_progress: experiment_id='{self.id}' job_id='{job_id}' - started")
+        _logger.info(f"Experiment.monitor_progress: experiment_id='{self.id}' - started")
 
-        # Monitor job progress with progress bar
-        completed_job_id = job_progress(job_id=job_id, project_id=self.project_id, run_id=self.id)
+        status = self.get_status()
+        progress_bar = tqdm(total=100, unit="%", desc="Experiment progress")
+        try:
+            while not status.is_complete:
+                new_progress = status.overall_progress
+                progress_bar.update(new_progress - progress_bar.n)
+                sleep(poll_interval_seconds)
+                status = self.get_status()
+            progress_bar.update(100 - progress_bar.n)
+        finally:
+            progress_bar.close()
 
         _logger.info(f"Experiment.monitor_progress: experiment_id='{self.id}' - completed")
-        return str(completed_job_id)
 
     # Query and export methods - similar to LogStream
 

tests/test_experiment_progress.py

@@ -0,0 +1,102 @@
+from unittest.mock import MagicMock, patch
+from uuid import uuid4
+
+import pytest
+
+from galileo.experiment import Experiment
+from galileo.shared.base import SyncState
+from galileo.shared.experiment_result import ExperimentStatusInfo
+
+FIXED_PROJECT_ID = str(uuid4())
+FIXED_EXPERIMENT_ID = str(uuid4())
+
+
+def _make_status(progress_percent: float) -> ExperimentStatusInfo:
+    """Build an ExperimentStatusInfo with a given log_generation progress (0-100)."""
+    phase = MagicMock()
+    phase.progress_percent = progress_percent / 100.0  # API uses 0.0-1.0
+    response = MagicMock()
+    response.status.log_generation = phase
+    return ExperimentStatusInfo(response)
+
+
+def _make_experiment() -> Experiment:
+    exp = Experiment._create_empty()
+    exp.id = FIXED_EXPERIMENT_ID
+    exp.project_id = FIXED_PROJECT_ID
+    exp.name = "test-experiment"
+    exp._set_state(SyncState.SYNCED)
+    return exp
+
+
+class TestMonitorProgress:
+    @patch("galileo.experiment.Experiment.get_status")
+    @patch("galileo.experiment.sleep", return_value=None)
+    def test_completes_when_status_reaches_100(self, mock_sleep, mock_get_status):
+        # Given: an experiment that progresses through 0%, 50%, then 100%
+        mock_get_status.side_effect = [_make_status(0.0), _make_status(50.0), _make_status(100.0)]
+        exp = _make_experiment()
+
+        # When: monitoring progress until completion
+        exp.monitor_progress(poll_interval_seconds=0.0)
+
+        # Then: get_status is polled until 100% is reached
+        assert mock_get_status.call_count == 3
+
+    @patch("galileo.experiment.Experiment.get_status")
+    @patch("galileo.experiment.sleep", return_value=None)
+    def test_already_complete_on_first_poll(self, mock_sleep, mock_get_status):
+        # Given: an experiment that is already at 100% on the first poll
+        mock_get_status.return_value = _make_status(100.0)
+        exp = _make_experiment()
+
+        # When: monitoring progress
+        exp.monitor_progress(poll_interval_seconds=0.0)
+
+        # Then: get_status is called once and sleep is never called
+        assert mock_get_status.call_count == 1
+        mock_sleep.assert_not_called()
+
+    @patch("galileo.experiment.Experiment.get_status")
+    @patch("galileo.experiment.sleep", return_value=None)
+    def test_uses_poll_interval_seconds(self, mock_sleep, mock_get_status):
+        # Given: an experiment that completes on the second poll
+        mock_get_status.side_effect = [_make_status(0.0), _make_status(100.0)]
+        exp = _make_experiment()
+
+        # When: monitoring with a custom poll interval
+        exp.monitor_progress(poll_interval_seconds=5.0)
+
+        # Then: sleep is called once with the specified interval
+        mock_sleep.assert_called_once_with(5.0)
+
+    def test_raises_without_experiment_id(self):
+        # Given: an experiment without an id
+        exp = Experiment._create_empty()
+        exp.id = None
+        exp.project_id = FIXED_PROJECT_ID
+
+        # When/Then: monitoring raises ValueError about the missing experiment id
+        with pytest.raises(ValueError, match="Experiment ID is not set"):
+            exp.monitor_progress()
+
+    def test_raises_without_project_id(self):
+        # Given: an experiment without a project_id
+        exp = Experiment._create_empty()
+        exp.id = FIXED_EXPERIMENT_ID
+        exp.project_id = None
+
+        # When/Then: monitoring raises ValueError about the missing project id
+        with pytest.raises(ValueError, match="Project ID is not set"):
+            exp.monitor_progress()
+
+    @patch("galileo.experiment.Experiment.get_status")
+    @patch("galileo.experiment.sleep", return_value=None)
+    def test_deprecated_job_id_warns(self, mock_sleep, mock_get_status):
+        # Given: an experiment that is already complete, and a caller passing the deprecated job_id
+        mock_get_status.return_value = _make_status(100.0)
+        exp = _make_experiment()
+
+        # When/Then: monitor_progress emits a DeprecationWarning when job_id is supplied
+        with pytest.warns(DeprecationWarning, match="job_id"):
+            exp.monitor_progress(job_id="some-old-job-id")