Docling outputting Json

mainJson.py

@@ -0,0 +1,59 @@
+import json
+import torch
+from pathlib import Path
+from docling.datamodel.base_models import InputFormat
+from docling.datamodel.pipeline_options import PdfPipelineOptions
+from docling.document_converter import DocumentConverter, PdfFormatOption
+from docling_core.types.doc import PictureItem
+
+# 1. Hardware Monkeypatch for Apple Silicon M3 CPU routing
+torch.backends.mps.is_available = lambda: False
+torch.backends.mps.is_built = lambda: False
+torch.set_default_device("cpu")
+
+def convert_to_json_and_images(pdf_name: str, output_json_name: str, image_folder_name: str = "extracted_images"):
+    source = f"./{pdf_name}"
+    output_dir = Path(image_folder_name)
+    output_dir.mkdir(exist_ok=True)
+
+    # 2. Configure pipeline to capture images
+    pipeline_options = PdfPipelineOptions()
+    pipeline_options.images_scale = 2.0             # 2.0 = roughly 144 DPI for crisp extraction
+    pipeline_options.generate_picture_images = True # Required to isolate figure items
+
+    # 3. Initialize converter with hardware-isolated options
+    converter = DocumentConverter(
+        format_options={
+            InputFormat.PDF: PdfFormatOption(pipeline_options=pipeline_options)
+        }
+    )
+
+    print(f"🚀 Processing {pdf_name} and building document schema...")
+    result = converter.convert(source)
+
+    # 4. Loop through the document items and save physical picture blocks
+    picture_counter = 0
+    for element, _level in result.document.iterate_items():
+        if isinstance(element, PictureItem):
+            picture_counter += 1
+            image_path = output_dir / f"figure_{picture_counter}.png"
+            
+            # Save the raw image block cleanly using Pillow
+            element.get_image(result.document).save(image_path, "PNG")
+
+    print(f"✨ Compilation finished! Serializing to JSON schema...")
+    
+    # 5. Export the Document object directly into a standard Python dictionary matrix
+    doc_dict = result.document.export_to_dict()
+    
+    # Write cleanly structured, indented JSON payload to disk
+    with open(output_json_name, "w", encoding="utf-8") as f:
+        json.dump(doc_dict, f, ensure_ascii=False, indent=2)
+    
+    print(f"✅ Success! Saved {output_json_name} and extracted {picture_counter} figures to ./{image_folder_name}/")
+
+# Run it for your target wiring rules document
+convert_to_json_and_images(
+    "1-AS_NZS_3000_2018_Wiring_Rules_Standards_Australia.pdf", 
+    "1-AS_NZS_3000_2018_Wiring_Rules_Standards_Australia.json"
+)