cleanup logging messages

Author: jgbradley1

graphrag/api/index.py

@@ -68,6 +68,7 @@ async def build_index(
 
     pipeline = PipelineFactory.create_pipeline(config, method, is_update_run)
 
+    logger.info("Pipeline started: %s", pipeline.names())
     workflow_callbacks.pipeline_start(pipeline.names())
 
     async for output in run_pipeline(
@@ -83,6 +84,7 @@ async def build_index(
             logger.info("Workflow %s completed successfully", output.workflow)
         logger.debug(str(output.result))
 
+    logger.info("Pipeline completed with %d workflows", len(outputs))
     workflow_callbacks.pipeline_end(outputs)
     return outputs
 

graphrag/callbacks/noop_workflow_callbacks.py

@@ -19,19 +19,15 @@ def __init__(self, logger_name: str = "graphrag"):
 
     def pipeline_start(self, names: list[str]) -> None:
         """Execute this callback to signal when the entire pipeline starts."""
-        self.logger.info("Pipeline started: %s", names)
 
     def pipeline_end(self, results: list[PipelineRunResult]) -> None:
         """Execute this callback to signal when the entire pipeline ends."""
-        self.logger.info("Pipeline completed with %d workflows", len(results))
 
     def workflow_start(self, name: str, instance: object) -> None:
         """Execute this callback when a workflow starts."""
-        self.logger.info("Workflow started: %s", name)
 
     def workflow_end(self, name: str, instance: object) -> None:
         """Execute this callback when a workflow ends."""
-        self.logger.info("Workflow completed: %s", name)
 
     def progress(self, progress: Progress) -> None:
         """Handle when progress occurs."""

graphrag/index/run/run_pipeline.py

@@ -116,11 +116,11 @@ async def _run_pipeline(
 
         for name, workflow_function in pipeline.run():
             last_workflow = name
-            logger.info("Starting workflow: %s", name)
+            logger.info("Workflow started: %s", name)
             context.callbacks.workflow_start(name, None)
             work_time = time.time()
             result = await workflow_function(config, context)
-            logger.info("Completed workflow: %s", name)
+            logger.info("Workflow completed: %s", name)
             context.callbacks.workflow_end(name, result)
             yield PipelineRunResult(
                 workflow=name, result=result.result, state=context.state, errors=None

graphrag/index/workflows/create_base_text_units.py

@@ -4,6 +4,7 @@
 """A module containing run_workflow method definition."""
 
 import json
+import logging
 from typing import Any, cast
 
 import pandas as pd
@@ -19,6 +20,8 @@
 from graphrag.logger.progress import Progress
 from graphrag.utils.storage import load_table_from_storage, write_table_to_storage
 
+logger = logging.getLogger(__name__)
+
 
 async def run_workflow(
     config: GraphRagConfig,

graphrag/logger/progress.py

@@ -3,11 +3,13 @@
 
 """Progress reporting types."""
 
+import logging
 from collections.abc import Callable, Iterable
 from dataclasses import dataclass
 from typing import TypeVar
 
 T = TypeVar("T")
+logger = logging.getLogger(__name__)
 
 
 @dataclass
@@ -24,7 +26,7 @@ class Progress:
     """Total number of items"""
 
     completed_items: int | None = None
-    """Number of items completed""" ""
+    """Number of items completed"""
 
 
 ProgressHandler = Callable[[Progress], None]
@@ -47,11 +49,11 @@ def __call__(self, num_ticks: int = 1) -> None:
         """Emit progress."""
         self._num_complete += num_ticks
         if self._callback is not None:
-            self._callback(
-                Progress(
-                    total_items=self._num_total, completed_items=self._num_complete
-                )
+            p = Progress(
+                total_items=self._num_total, completed_items=self._num_complete
             )
+            logger.info("Progress: %s/%s", str(p.completed_items), str(p.total_items))
+            self._callback(p)
 
     def done(self) -> None:
         """Mark the progress as done."""

graphrag/logger/standard_logging.py

@@ -136,7 +136,7 @@ def init_loggers(
         case ReportingType.console:
             # Only add console handler if not already added
             if not enable_console:
-                handler = logging.StreamHandler(sys.stdout)
+                handler = logging.StreamHandler()
                 handler.setFormatter(formatter)
                 logger.addHandler(handler)
         case ReportingType.blob: