Merge pull request #2 from datalab-to/dev

Dev

chandra/model/__init__.py

@@ -48,6 +48,7 @@ class InferenceManager:
                     page_box=[0, 0, input_item.image.width, input_item.image.height],
                     token_count=result.token_count,
                     images=extract_images(result.raw, chunks, input_item.image),
+                    error=result.error,
                 )
             )
         return output

chandra/model/schema.py

@@ -27,3 +27,4 @@ class BatchOutputItem:
     page_box: List[int]
     token_count: int
     images: dict
+    error: bool

chandra/output.py

@@ -30,7 +30,11 @@ def extract_images(html: str, chunks: dict, image: Image.Image):
             if not img:
                 continue
             bbox = chunk["bbox"]
-            block_image = image.crop(bbox)
+            try:
+                block_image = image.crop(bbox)
+            except ValueError:
+                # Happens when bbox coordinates are invalid
+                continue
             img_name = get_image_name(html, div_idx)
             images[img_name] = block_image
     return images

chandra/scripts/templates/screenshot.html

@@ -137,6 +137,7 @@
             padding: 30px;
             line-height: 1.6;
             color: #333;
+            font-size: 24px;
         }
 
         .markdown-content h1, .markdown-content h2, .markdown-content h3 {