feat!: Docling v2 (#117)

---------

Signed-off-by: Christoph Auer <cau@zurich.ibm.com>
Signed-off-by: Maxim Lysak <mly@zurich.ibm.com>
Signed-off-by: Michele Dolfi <dol@zurich.ibm.com>
Signed-off-by: Panos Vagenas <35837085+vagenas@users.noreply.github.com>
Co-authored-by: Maxim Lysak <mly@zurich.ibm.com>
Co-authored-by: Michele Dolfi <dol@zurich.ibm.com>
Co-authored-by: Panos Vagenas <35837085+vagenas@users.noreply.github.com>

docs/examples/export_multimodal.py

@@ -5,10 +5,11 @@ from pathlib import Path
 
 import pandas as pd
 
-from docling.datamodel.base_models import AssembleOptions, ConversionStatus
-from docling.datamodel.document import DocumentConversionInput
-from docling.document_converter import DocumentConverter
+from docling.datamodel.base_models import InputFormat
+from docling.datamodel.pipeline_options import PdfPipelineOptions
+from docling.document_converter import DocumentConverter, PdfFormatOption
 from docling.utils.export import generate_multimodal_pages
+from docling.utils.utils import create_hash
 
 _log = logging.getLogger(__name__)
 
@@ -18,71 +19,66 @@ IMAGE_RESOLUTION_SCALE = 2.0
 def main():
     logging.basicConfig(level=logging.INFO)
 
-    input_doc_paths = [
-        Path("./tests/data/2206.01062.pdf"),
-    ]
-    output_dir = Path("./scratch")
-
-    input_files = DocumentConversionInput.from_paths(input_doc_paths)
+    input_doc_path = Path("./tests/data/2206.01062.pdf")
+    output_dir = Path("scratch")
 
     # Important: For operating with page images, we must keep them, otherwise the DocumentConverter
     # will destroy them for cleaning up memory.
     # This is done by setting AssembleOptions.images_scale, which also defines the scale of images.
     # scale=1 correspond of a standard 72 DPI image
-    assemble_options = AssembleOptions()
-    assemble_options.images_scale = IMAGE_RESOLUTION_SCALE
+    pipeline_options = PdfPipelineOptions()
+    pipeline_options.images_scale = IMAGE_RESOLUTION_SCALE
+    pipeline_options.generate_page_images = True
 
-    doc_converter = DocumentConverter(assemble_options=assemble_options)
+    doc_converter = DocumentConverter(
+        format_options={
+            InputFormat.PDF: PdfFormatOption(pipeline_options=pipeline_options)
+        }
+    )
 
     start_time = time.time()
 
-    converted_docs = doc_converter.convert(input_files)
+    conv_res = doc_converter.convert(input_doc_path)
 
-    success_count = 0
-    failure_count = 0
     output_dir.mkdir(parents=True, exist_ok=True)
-    for doc in converted_docs:
-        if doc.status != ConversionStatus.SUCCESS:
-            _log.info(f"Document {doc.input.file} failed to convert.")
-            failure_count += 1
-            continue
 
-        rows = []
-        for (
-            content_text,
-            content_md,
-            content_dt,
-            page_cells,
-            page_segments,
-            page,
-        ) in generate_multimodal_pages(doc):
+    rows = []
+    for (
+        content_text,
+        content_md,
+        content_dt,
+        page_cells,
+        page_segments,
+        page,
+    ) in generate_multimodal_pages(conv_res):
 
-            dpi = page._default_image_scale * 72
+        dpi = page._default_image_scale * 72
 
-            rows.append(
-                {
-                    "document": doc.input.file.name,
-                    "hash": doc.input.document_hash,
-                    "page_hash": page.page_hash,
-                    "image": {
-                        "width": page.image.width,
-                        "height": page.image.height,
-                        "bytes": page.image.tobytes(),
-                    },
-                    "cells": page_cells,
-                    "contents": content_text,
-                    "contents_md": content_md,
-                    "contents_dt": content_dt,
-                    "segments": page_segments,
-                    "extra": {
-                        "page_num": page.page_no + 1,
-                        "width_in_points": page.size.width,
-                        "height_in_points": page.size.height,
-                        "dpi": dpi,
-                    },
-                }
-            )
-        success_count += 1
+        rows.append(
+            {
+                "document": conv_res.input.file.name,
+                "hash": conv_res.input.document_hash,
+                "page_hash": create_hash(
+                    conv_res.input.document_hash + ":" + str(page.page_no - 1)
+                ),
+                "image": {
+                    "width": page.image.width,
+                    "height": page.image.height,
+                    "bytes": page.image.tobytes(),
+                },
+                "cells": page_cells,
+                "contents": content_text,
+                "contents_md": content_md,
+                "contents_dt": content_dt,
+                "segments": page_segments,
+                "extra": {
+                    "page_num": page.page_no + 1,
+                    "width_in_points": page.size.width,
+                    "height_in_points": page.size.height,
+                    "dpi": dpi,
+                },
+            }
+        )
 
     # Generate one parquet from all documents
     df = pd.json_normalize(rows)
@@ -92,12 +88,9 @@ def main():
 
     end_time = time.time() - start_time
 
-    _log.info(f"All documents were converted in {end_time:.2f} seconds.")
-
-    if failure_count > 0:
-        raise RuntimeError(
-            f"The example failed converting {failure_count} on {len(input_doc_paths)}."
-        )
+    _log.info(
+        f"Document converted and multimodal pages generated in {end_time:.2f} seconds."
+    )
 
     # This block demonstrates how the file can be opened with the HF datasets library
     # from datasets import Dataset