add logs at start, end, and errors

python/packages/officeconvert/src/officeconvert/conversion.py

@@ -4,6 +4,7 @@ from __future__ import annotations
 
 from collections.abc import Callable
 from dataclasses import dataclass
+import logging
 from pathlib import Path
 import subprocess
 from typing import Iterable
@@ -35,6 +36,12 @@ PHASE_EXTRACTING_NOTES = "extracting_notes"
 PHASE_PPTX_TO_PDF = "pptx_to_pdf"
 PHASE_PDF_TO_IMAGES = "pdf_to_images"
 
+logger = logging.getLogger("uvicorn.error")
+
+
+class ConversionTimeoutError(RuntimeError):
+    """Raised when a conversion subprocess exceeds the configured timeout."""
+
 
 def convert_pptx_to_pdf(pptx_path: Path, pdf_path: Path, *, timeout_s: int = 120) -> Path:
     """Convert a PPTX file to PDF using headless LibreOffice.
@@ -75,10 +82,12 @@ def convert_pptx_to_pdf(pptx_path: Path, pdf_path: Path, *, timeout_s: int = 120
             timeout=timeout_s,
         )
     except subprocess.TimeoutExpired as exc:
-        raise RuntimeError(
+        message = (
             "LibreOffice conversion timed out after "
             f"{timeout_s} seconds while rendering {pptx_path.name}"
-        ) from exc
+        )
+        logger.error(message, exc_info=True)
+        raise ConversionTimeoutError(message) from exc
     if completed.returncode != 0:
         raise RuntimeError(
             f"LibreOffice conversion failed: {completed.stderr.strip() or completed.stdout.strip()}"
@@ -144,11 +153,13 @@ def render_pdf_to_images(
                 timeout=timeout_s,
             )
         except subprocess.TimeoutExpired as exc:
-            raise RuntimeError(
+            message = (
                 "Poppler rasterization timed out after "
                 f"{timeout_s} seconds while rendering {pdf_path.name}; "
                 "increase conversion PDF render timeout cap or lower image DPI"
-            ) from exc
+            )
+            logger.error(message, exc_info=True)
+            raise ConversionTimeoutError(message) from exc
         if completed.returncode != 0:
             raise RuntimeError(
                 f"Poppler rasterization failed: {completed.stderr.strip() or completed.stdout.strip()}"
@@ -188,11 +199,13 @@ def render_pdf_to_images(
                     if operation_timeout_s is not None
                     else f"per-page timeout {timeout_s}s"
                 )
-                raise RuntimeError(
+                message = (
                     "Poppler rasterization timed out while rendering page "
                     f"{page_index}/{total_pages} of {pdf_path.name}; "
                     f"{timeout_context}. Increase timeout settings or lower image DPI."
-                ) from exc
+                )
+                logger.error(message, exc_info=True)
+                raise ConversionTimeoutError(message) from exc
             if completed.returncode != 0:
                 raise RuntimeError(
                     "Poppler rasterization failed on page "
@@ -298,16 +311,28 @@ def convert_pptx_to_slidedeck(
     _emit_progress(progress_callback, PHASE_PPTX_TO_PDF, 1, 1)
 
     _emit_progress(progress_callback, PHASE_PDF_TO_IMAGES, 0, slide_count)
+    pdf_to_images_page_timeout = _compute_page_timeout(
+        total_timeout_s=pdf_to_images_timeout,
+        page_count=slide_count,
+        base_timeout_s=pdf_to_images_base_timeout_s,
+    )
+    logger.info(
+        "Conversion plan source=%s slides=%d dpi=%d image_format=%s "
+        "computed_timeouts_s[pptx_to_pdf_total=%d,pdf_to_images_total=%d,pdf_to_images_per_page=%d]",
+        pptx_path.name,
+        slide_count,
+        dpi,
+        image_format,
+        pptx_to_pdf_timeout,
+        pdf_to_images_timeout,
+        pdf_to_images_page_timeout,
+    )
     image_paths = render_pdf_to_images(
         pdf_path,
         image_dir,
         dpi=dpi,
         image_format=image_format,
-        timeout_s=_compute_page_timeout(
-            total_timeout_s=pdf_to_images_timeout,
-            page_count=slide_count,
-            base_timeout_s=pdf_to_images_base_timeout_s,
-        ),
+        timeout_s=pdf_to_images_page_timeout,
         total_pages=slide_count,
         operation_timeout_s=pdf_to_images_timeout,
         page_progress_callback=lambda current, max_pages: _emit_progress(