Merge branch 'cau/input-format-abstraction' of github.com:DS4SD/docling into cau/input-format-abstraction

docling/backend/html_backend.py

@@ -1,5 +1,5 @@
 import logging
-from io import BytesIO, TextIOWrapper
+from io import BytesIO
 from pathlib import Path
 from typing import Set, Union
 
@@ -81,9 +81,10 @@ class HTMLDocumentBackend(DeclarativeDocumentBackend):
                 try:
                     self.analyse_element(element, idx, doc)
                 except Exception as exc_child:
+
                     _log.error(" -> error treating child: ", exc_child)
                     _log.error(" => element: ", element, "\n")
-                    pass
+                    raise exc_child
 
         except Exception as exc:
             pass
@@ -212,10 +213,18 @@ class HTMLDocumentBackend(DeclarativeDocumentBackend):
     def handle_list(self, element, idx, doc):
         """Handles list tags (ul, ol) and their list items."""
 
-        # create a list group
+        if element.name == "ul":
-        self.parents[self.level + 1] = doc.add_group(
+            # create a list group
-            parent=self.parents[self.level], name="list", label=GroupLabel.LIST
+            self.parents[self.level + 1] = doc.add_group(
-        )
+                parent=self.parents[self.level], name="list", label=GroupLabel.LIST
+            )
+        elif element.name == "ol":
+            # create a list group
+            self.parents[self.level + 1] = doc.add_group(
+                parent=self.parents[self.level],
+                name="ordered list",
+                label=GroupLabel.ORDERED_LIST,
+            )
         self.level += 1
 
         self.walk(element, doc)
@@ -226,13 +235,26 @@ class HTMLDocumentBackend(DeclarativeDocumentBackend):
     def handle_listitem(self, element, idx, doc):
         """Handles listitem tags (li)."""
         nested_lists = element.find(["ul", "ol"])
+
+        parent_list_label = self.parents[self.level].label
+        index_in_list = len(self.parents[self.level].children) + 1
+
         if nested_lists:
             name = element.name
             text = self.get_direct_text(element)
 
+            marker = ""
+            enumerated = False
+            if parent_list_label == GroupLabel.ORDERED_LIST:
+                marker = str(index_in_list)
+                enumerated = True
+
             # create a list-item
-            self.parents[self.level + 1] = doc.add_text(
+            self.parents[self.level + 1] = doc.add_list_item(
-                label=DocItemLabel.LIST_ITEM, text=text, parent=self.parents[self.level]
+                text=text,
+                enumerated=enumerated,
+                marker=marker,
+                parent=self.parents[self.level],
             )
             self.level += 1
 
@@ -244,8 +266,16 @@ class HTMLDocumentBackend(DeclarativeDocumentBackend):
         elif isinstance(element.text, str):
             text = element.text.strip()
 
-            doc.add_text(
+            marker = ""
-                label=DocItemLabel.LIST_ITEM, text=text, parent=self.parents[self.level]
+            enumerated = False
+            if parent_list_label == GroupLabel.ORDERED_LIST:
+                marker = f"{str(index_in_list)}."
+                enumerated = True
+            doc.add_list_item(
+                text=text,
+                enumerated=enumerated,
+                marker=marker,
+                parent=self.parents[self.level],
             )
         else:
             _log.warn("list-item has no text: ", element)

docling/backend/mspowerpoint_backend.py

@@ -43,7 +43,7 @@ class MsPowerpointDocumentBackend(DeclarativeDocumentBackend, PaginatedDocumentB
         try:
             self.pptx_obj = Presentation(self.path_or_stream)
             self.valid = True
-        except Exception:
+        except Exception as e:
             raise RuntimeError(
                 f"MsPowerpointDocumentBackend could not load document with hash {document_hash}"
             ) from e
@@ -134,6 +134,8 @@ class MsPowerpointDocumentBackend(DeclarativeDocumentBackend, PaginatedDocumentB
             prov = self.generate_prov(shape, slide_ind, shape.text.strip())
 
             if is_a_list:
+                # TODO: determine if this is an unordered list or an ordered list.
+                #  Set GroupLabel.ORDERED_LIST when it fits.
                 new_list = doc.add_group(
                     label=GroupLabel.LIST, name=f"list", parent=parent_slide
                 )
@@ -157,9 +159,10 @@ class MsPowerpointDocumentBackend(DeclarativeDocumentBackend, PaginatedDocumentB
                         e_is_a_list_item = True
                     else:
                         e_is_a_list_item = False
+
                     if e_is_a_list_item:
-                        doc.add_text(
+                        # TODO: Set marker and enumerated arguments if this is an enumeration element.
-                            label=DocItemLabel.LIST_ITEM,
+                        doc.add_list_item(
                             parent=new_list,
                             text=e.text.strip(),
                             prov=prov,

docling/backend/msword_backend.py

@@ -312,9 +312,8 @@ class MsWordDocumentBackend(DeclarativeDocumentBackend):
                 label=GroupLabel.LIST, name="list", parent=self.parents[level - 1]
             )
 
-            doc.add_text(
+            # TODO: Set marker and enumerated arguments if this is an enumeration element.
-                label=DocItemLabel.LIST_ITEM, parent=self.parents[level], text=text
+            doc.add_list_item(parent=self.parents[level], text=text)
-            )
 
         elif (
             self.prev_numid() == numid and self.prev_indent() < ilevel
@@ -323,12 +322,14 @@ class MsWordDocumentBackend(DeclarativeDocumentBackend):
                 self.level_at_new_list + self.prev_indent() + 1,
                 self.level_at_new_list + ilevel + 1,
             ):
+                # TODO: determine if this is an unordered list or an ordered list.
+                #  Set GroupLabel.ORDERED_LIST when it fits.
                 self.parents[i] = doc.add_group(
                     label=GroupLabel.LIST, name="list", parent=self.parents[i - 1]
                 )
 
-            doc.add_text(
+            # TODO: Set marker and enumerated arguments if this is an enumeration element.
-                label=DocItemLabel.LIST_ITEM,
+            doc.add_list_item(
                 parent=self.parents[self.level_at_new_list + ilevel],
                 text=text,
             )
@@ -338,16 +339,15 @@ class MsWordDocumentBackend(DeclarativeDocumentBackend):
                 if k > self.level_at_new_list + ilevel:
                     self.parents[k] = None
 
-            doc.add_text(
+            # TODO: Set marker and enumerated arguments if this is an enumeration element.
-                label=DocItemLabel.LIST_ITEM,
+            doc.add_list_item(
                 parent=self.parents[self.level_at_new_list + ilevel],
                 text=text,
             )
 
         elif self.prev_numid() == numid or self.prev_indent() == ilevel:
-            doc.add_text(
+            # TODO: Set marker and enumerated arguments if this is an enumeration element.
-                label=DocItemLabel.LIST_ITEM, parent=self.parents[level - 1], text=text
+            doc.add_list_item(parent=self.parents[level - 1], text=text)
-            )
         return
 
     def handle_tables(self, element, docx_obj, doc):

docling/datamodel/document.py

@@ -3,7 +3,7 @@ import re
 from enum import Enum
 from io import BytesIO
 from pathlib import Path, PurePath
-from typing import Dict, Iterable, List, Optional, Tuple, Type, Union
+from typing import Dict, Iterable, List, Optional, Tuple, Union
 
 import filetype
 from docling_core.types import BaseText
@@ -24,10 +24,6 @@ from pydantic import BaseModel
 from typing_extensions import deprecated
 
 from docling.backend.abstract_backend import AbstractDocumentBackend
-from docling.backend.docling_parse_backend import DoclingParseDocumentBackend
-from docling.backend.html_backend import HTMLDocumentBackend
-from docling.backend.mspowerpoint_backend import MsPowerpointDocumentBackend
-from docling.backend.msword_backend import MsWordDocumentBackend
 from docling.datamodel.base_models import (
     AssembledUnit,
     ConversionStatus,
@@ -372,14 +368,20 @@ class ConversionResult(BaseModel):
         strict_text: bool = False,
         image_placeholder: str = "<!-- image -->",
     ) -> str:
-        return self.legacy_output.export_to_markdown(
+        if self.legacy_output is None:
-            delim=delim,
+            raise RuntimeError(
-            main_text_start=main_text_start,
+                "No legacy output was produced, can not export as markdown. "
-            main_text_stop=main_text_stop,
+                "Please use output.export_to_markdown() instead."
-            main_text_labels=main_text_labels,
+            )
-            strict_text=strict_text,
+        else:
-            image_placeholder=image_placeholder,
+            return self.legacy_output.export_to_markdown(
-        )
+                delim=delim,
+                main_text_start=main_text_start,
+                main_text_stop=main_text_stop,
+                main_text_labels=main_text_labels,
+                strict_text=strict_text,
+                image_placeholder=image_placeholder,
+            )
 
     @deprecated("Use output.export_to_text() instead.")
     def render_as_text(
@@ -394,13 +396,19 @@ class ConversionResult(BaseModel):
             "caption",
         ],
     ) -> str:
-        return self.legacy_output.export_to_markdown(
+        if self.legacy_output is None:
-            delim=delim,
+            raise RuntimeError(
-            main_text_start=main_text_start,
+                "No legacy output was produced, can not export as text. "
-            main_text_stop=main_text_stop,
+                "Please use output.export_to_markdown() instead."
-            main_text_labels=main_text_labels,
+            )
-            strict_text=True,
+        else:
-        )
+            return self.legacy_output.export_to_markdown(
+                delim=delim,
+                main_text_start=main_text_start,
+                main_text_stop=main_text_stop,
+                main_text_labels=main_text_labels,
+                strict_text=True,
+            )
 
     @deprecated("Use output.export_to_document_tokens() instead.")
     def render_as_doctags(
@@ -426,21 +434,27 @@ class ConversionResult(BaseModel):
         add_table_cell_label: bool = True,
         add_table_cell_text: bool = True,
     ) -> str:
-        return self.legacy_output.export_to_document_tokens(
+        if self.legacy_output is None:
-            delim=delim,
+            raise RuntimeError(
-            main_text_start=main_text_start,
+                "No legacy output was produced, can not export as doctags. "
-            main_text_stop=main_text_stop,
+                "Please use output.export_to_markdown() instead."
-            main_text_labels=main_text_labels,
+            )
-            xsize=xsize,
+        else:
-            ysize=ysize,
+            return self.legacy_output.export_to_document_tokens(
-            add_location=add_location,
+                delim=delim,
-            add_content=add_content,
+                main_text_start=main_text_start,
-            add_page_index=add_page_index,
+                main_text_stop=main_text_stop,
-            # table specific flags
+                main_text_labels=main_text_labels,
-            add_table_cell_location=add_table_cell_location,
+                xsize=xsize,
-            add_table_cell_label=add_table_cell_label,
+                ysize=ysize,
-            add_table_cell_text=add_table_cell_text,
+                add_location=add_location,
-        )
+                add_content=add_content,
+                add_page_index=add_page_index,
+                # table specific flags
+                add_table_cell_location=add_table_cell_location,
+                add_table_cell_label=add_table_cell_label,
+                add_table_cell_text=add_table_cell_text,
+            )
 
     def render_element_images(
         self, element_types: Tuple[PageElement] = (FigureElement,)

docling/models/tesseract_ocr_model.py

@@ -1,7 +1,6 @@
 import logging
 from typing import Iterable
 
-import numpy
 from docling_core.types.experimental import BoundingBox, CoordOrigin
 
 from docling.datamodel.base_models import OcrCell, Page

docling/pipeline/simple_model_pipeline.py

@@ -1,7 +1,4 @@
 import logging
-from typing import Iterable
-
-from docling_core.types.experimental import NodeItem
 
 from docling.backend.abstract_backend import (
     AbstractDocumentBackend,
@@ -9,7 +6,7 @@ from docling.backend.abstract_backend import (
 )
 from docling.datamodel.base_models import ConversionStatus
 from docling.datamodel.document import ConversionResult, InputDocument
-from docling.datamodel.pipeline_options import PdfPipelineOptions, PipelineOptions
+from docling.datamodel.pipeline_options import PipelineOptions
 from docling.pipeline.base_model_pipeline import AbstractModelPipeline
 
 _log = logging.getLogger(__name__)

examples/run_with_formats.py

@@ -2,13 +2,10 @@ import json
 import logging
 from pathlib import Path
 
-from docling.backend.docling_parse_backend import DoclingParseDocumentBackend
-from docling.backend.msword_backend import MsWordDocumentBackend
 from docling.backend.pypdfium2_backend import PyPdfiumDocumentBackend
 from docling.datamodel.base_models import InputFormat
 from docling.document_converter import (
     DocumentConverter,
-    FormatOption,
     PdfFormatOption,
     WordFormatOption,
 )
@@ -38,6 +35,7 @@ doc_converter = DocumentConverter(  # all of the below is optional, has internal
         # InputFormat.IMAGE,
         InputFormat.DOCX,
         InputFormat.HTML,
+        InputFormat.PPTX,
     ],  # whitelist formats, other files are ignored.
     format_options={
         InputFormat.PDF: PdfFormatOption(
@@ -50,6 +48,30 @@ doc_converter = DocumentConverter(  # all of the below is optional, has internal
     },
 )
 
+doc_converter = DocumentConverter(  # all of the below is optional, has internal defaults.
+    pdf=None,
+    docx=WordFormatOption(
+        pipeline_cls=SimpleModelPipeline  # , backend=MsWordDocumentBackend
+    ),
+    formats=[
+        InputFormat.PDF,
+        # InputFormat.IMAGE,
+        InputFormat.DOCX,
+        InputFormat.HTML,
+        InputFormat.PPTX,
+    ],  # whitelist formats, other files are ignored.
+    format_options={
+        InputFormat.PDF: PdfFormatOption(
+            pipeline_cls=StandardPdfModelPipeline, backend=PyPdfiumDocumentBackend
+        ),  # PdfFormatOption(backend=PyPdfiumDocumentBackend),
+        InputFormat.DOCX: WordFormatOption(
+            pipeline_cls=SimpleModelPipeline  # , backend=MsWordDocumentBackend
+        ),
+        # InputFormat.IMAGE: PdfFormatOption(),
+    },
+)
+
+
 conv_results = doc_converter.convert_all(input_paths)
 
 for res in conv_results:

tests/verify_utils.py

@@ -3,7 +3,6 @@ import warnings
 from pathlib import Path
 from typing import List
 
-from docling_core.types import BaseText
 from docling_core.types import Document as DsDocument
 from docling_core.types.experimental import DoclingDocument
 from pydantic import TypeAdapter