pp-doclayout: skip corrupt images instead of crashing the run

Previously, a single corrupt image (UnidentifiedImageError or OSError)
in the source bucket or dataset would terminate the entire run via an
unhandled exception in the gen() iterator. On a 5000-image run, hitting
a truncated/unreadable file at sample 2400 wasted ~1 hour of GPU time.

Now both iter_bucket_images() and iter_dataset_images() catch
UnidentifiedImageError and OSError per image, log a WARNING with the
path, and continue. A summary "Skipped N unreadable image(s)" is logged
when the iterator exhausts. Run completes with whatever images were
readable.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

pp-doclayout.py

@@ -84,7 +84,7 @@ from pathlib import Path
 from typing import Any, Dict, Iterator, List, Optional, Tuple, Union
 
 import numpy as np
-from PIL import Image
+from PIL import Image, UnidentifiedImageError
 from tqdm.auto import tqdm
 
 logging.basicConfig(level=logging.INFO)
@@ -234,13 +234,25 @@ def iter_dataset_images(
     total = len(ds)
 
     def gen() -> Iterator[SourceItem]:
+        skipped = 0
         for i in range(total):
-            row = ds[i]
+            try:
+                row = ds[i]
+                image = to_pil(row[image_column])
+            except (UnidentifiedImageError, OSError) as e:
+                skipped += 1
+                logger.warning(
+                    f"Skipping unreadable image at row {i}: "
+                    f"{type(e).__name__}: {e}"
+                )
+                continue
             yield SourceItem(
                 key=f"row-{i:08d}",
-                image=to_pil(row[image_column]),
+                image=image,
                 extras={},  # original schema is preserved by the sink via the dataset ref
             )
+        if skipped:
+            logger.info(f"Skipped {skipped} unreadable image(s) total")
 
     return gen(), total, ds
 
@@ -365,14 +377,26 @@ def iter_bucket_images(
         return path
 
     def gen() -> Iterator[SourceItem]:
+        skipped = 0
         for path in all_paths:
-            with fs.open(path, "rb") as f:
-                data = f.read()
+            try:
+                with fs.open(path, "rb") as f:
+                    data = f.read()
+                image = to_pil(data)
+            except (UnidentifiedImageError, OSError) as e:
+                skipped += 1
+                logger.warning(
+                    f"Skipping unreadable image {path}: "
+                    f"{type(e).__name__}: {e}"
+                )
+                continue
             yield SourceItem(
                 key=key_for(path),
-                image=to_pil(data),
+                image=image,
                 extras={"__source_path": key_for(path)},
             )
+        if skipped:
+            logger.info(f"Skipped {skipped} unreadable image(s) total")
 
     return gen(), total