docling 2.30.0__py3-none-any.whl → 2.31.0__py3-none-any.whl

docling/backend/asciidoc_backend.py

@@ -34,7 +34,7 @@ class AsciiDocBackend(DeclarativeDocumentBackend):
                 text_stream = self.path_or_stream.getvalue().decode("utf-8")
                 self.lines = text_stream.split("\n")
             if isinstance(self.path_or_stream, Path):
-                with open(self.path_or_stream, "r", encoding="utf-8") as f:
+                with open(self.path_or_stream, encoding="utf-8") as f:
                     self.lines = f.readlines()
             self.valid = True
 
@@ -75,14 +75,12 @@ class AsciiDocBackend(DeclarativeDocumentBackend):
 
         return doc
 
-    def _parse(self, doc: DoclingDocument):
+    def _parse(self, doc: DoclingDocument):  # noqa: C901
         """
         Main function that orchestrates the parsing by yielding components:
         title, section headers, text, lists, and tables.
         """
 
-        content = ""
-
         in_list = False
         in_table = False
 
@@ -95,7 +93,7 @@ class AsciiDocBackend(DeclarativeDocumentBackend):
         # indents: dict[int, Union[DocItem, GroupItem, None]] = {}
         indents: dict[int, Union[GroupItem, None]] = {}
 
-        for i in range(0, 10):
+        for i in range(10):
             parents[i] = None
             indents[i] = None
 
@@ -125,7 +123,6 @@ class AsciiDocBackend(DeclarativeDocumentBackend):
 
             # Lists
             elif self._is_list_item(line):
-
                 _log.debug(f"line: {line}")
                 item = self._parse_list_item(line)
                 _log.debug(f"parsed list-item: {item}")
@@ -147,7 +144,6 @@ class AsciiDocBackend(DeclarativeDocumentBackend):
                     indents[level + 1] = item["indent"]
 
                 elif in_list and item["indent"] < indents[level]:
-
                     # print(item["indent"], " => ", indents[level])
                     while item["indent"] < indents[level]:
                         # print(item["indent"], " => ", indents[level])
@@ -176,7 +172,6 @@ class AsciiDocBackend(DeclarativeDocumentBackend):
             elif in_table and (
                 (not self._is_table_line(line)) or line.strip() == "|==="
             ):  # end of table
-
                 caption = None
                 if len(caption_data) > 0:
                     caption = doc.add_text(
@@ -195,7 +190,6 @@ class AsciiDocBackend(DeclarativeDocumentBackend):
 
             # Picture
             elif self._is_picture(line):
-
                 caption = None
                 if len(caption_data) > 0:
                     caption = doc.add_text(
@@ -250,7 +244,6 @@ class AsciiDocBackend(DeclarativeDocumentBackend):
                 text_data = []
 
             elif len(line.strip()) > 0:  # allow multiline texts
-
                 item = self._parse_text(line)
                 text_data.append(item["text"])
 
@@ -273,14 +266,14 @@ class AsciiDocBackend(DeclarativeDocumentBackend):
 
     def _get_current_level(self, parents):
         for k, v in parents.items():
-            if v == None and k > 0:
+            if v is None and k > 0:
                 return k - 1
 
         return 0
 
     def _get_current_parent(self, parents):
         for k, v in parents.items():
-            if v == None and k > 0:
+            if v is None and k > 0:
                 return parents[k - 1]
 
         return None
@@ -328,7 +321,7 @@ class AsciiDocBackend(DeclarativeDocumentBackend):
                     "marker": marker,
                     "text": text.strip(),
                     "numbered": False,
-                    "indent": 0 if indent == None else len(indent),
+                    "indent": 0 if indent is None else len(indent),
                 }
             else:
                 return {
@@ -336,7 +329,7 @@ class AsciiDocBackend(DeclarativeDocumentBackend):
                     "marker": marker,
                     "text": text.strip(),
                     "numbered": True,
-                    "indent": 0 if indent == None else len(indent),
+                    "indent": 0 if indent is None else len(indent),
                 }
         else:
             # Fallback if no match
@@ -357,7 +350,6 @@ class AsciiDocBackend(DeclarativeDocumentBackend):
         return [cell.strip() for cell in line.split("|") if cell.strip()]
 
     def _populate_table_as_grid(self, table_data):
-
         num_rows = len(table_data)
 
         # Adjust the table data into a grid format

docling/backend/csv_backend.py

@@ -58,7 +58,7 @@ class CsvDocumentBackend(DeclarativeDocumentBackend):
         head = self.content.readline()
         dialect = csv.Sniffer().sniff(head, ",;\t|:")
         _log.info(f'Parsing CSV with delimiter: "{dialect.delimiter}"')
-        if not dialect.delimiter in {",", ";", "\t", "|", ":"}:
+        if dialect.delimiter not in {",", ";", "\t", "|", ":"}:
             raise RuntimeError(
                 f"Cannot convert csv with unknown delimiter {dialect.delimiter}."
             )

docling/backend/docling_parse_backend.py

@@ -1,8 +1,9 @@
 import logging
 import random
+from collections.abc import Iterable
 from io import BytesIO
 from pathlib import Path
-from typing import Iterable, List, Optional, Union
+from typing import List, Optional, Union
 
 import pypdfium2 as pdfium
 from docling_core.types.doc import BoundingBox, CoordOrigin, Size
@@ -156,7 +157,6 @@ class DoclingParsePageBackend(PdfPageBackend):
     def get_page_image(
         self, scale: float = 1, cropbox: Optional[BoundingBox] = None
     ) -> Image.Image:
-
         page_size = self.get_size()
 
         if not cropbox:

docling/backend/docling_parse_v2_backend.py

@@ -1,8 +1,9 @@
 import logging
 import random
+from collections.abc import Iterable
 from io import BytesIO
 from pathlib import Path
-from typing import TYPE_CHECKING, Iterable, List, Optional, Union
+from typing import TYPE_CHECKING, List, Optional, Union
 
 import pypdfium2 as pdfium
 from docling_core.types.doc import BoundingBox, CoordOrigin
@@ -172,7 +173,6 @@ class DoclingParseV2PageBackend(PdfPageBackend):
     def get_page_image(
         self, scale: float = 1, cropbox: Optional[BoundingBox] = None
     ) -> Image.Image:
-
         page_size = self.get_size()
 
         if not cropbox:

docling/backend/docling_parse_v4_backend.py

@@ -1,14 +1,14 @@
 import logging
-import random
+from collections.abc import Iterable
 from io import BytesIO
 from pathlib import Path
-from typing import TYPE_CHECKING, Iterable, List, Optional, Union
+from typing import TYPE_CHECKING, Optional, Union
 
 import pypdfium2 as pdfium
 from docling_core.types.doc import BoundingBox, CoordOrigin
 from docling_core.types.doc.page import SegmentedPdfPage, TextCell
 from docling_parse.pdf_parser import DoclingPdfParser, PdfDocument
-from PIL import Image, ImageDraw
+from PIL import Image
 from pypdfium2 import PdfPage
 
 from docling.backend.pdf_backend import PdfDocumentBackend, PdfPageBackend
@@ -93,7 +93,6 @@ class DoclingParseV4PageBackend(PdfPageBackend):
     def get_page_image(
         self, scale: float = 1, cropbox: Optional[BoundingBox] = None
     ) -> Image.Image:
-
         page_size = self.get_size()
 
         if not cropbox:

docling/backend/docx/latex/latex_dict.py

@@ -1,12 +1,8 @@
-# -*- coding: utf-8 -*-
-
 """
 Adapted from https://github.com/xiilei/dwml/blob/master/dwml/latex_dict.py
 On 23/01/2025
 """
 
-from __future__ import unicode_literals
-
 CHARS = ("{", "}", "_", "^", "#", "&", "$", "%", "~")
 
 BLANK = ""
@@ -79,7 +75,6 @@ CHR_BO = {
 }
 
 T = {
-    "\u2192": "\\rightarrow ",
     # Greek letters
     "\U0001d6fc": "\\alpha ",
     "\U0001d6fd": "\\beta ",

docling/backend/docx/latex/omml.py

@@ -76,8 +76,7 @@ def get_val(key, default=None, store=CHR):
         return default
 
 
-class Tag2Method(object):
-
+class Tag2Method:
     def call_method(self, elm, stag=None):
         getmethod = self.tag2meth.get
         if stag is None:
@@ -130,7 +129,6 @@ class Tag2Method(object):
 
 
 class Pr(Tag2Method):
-
     text = ""
 
     __val_tags = ("chr", "pos", "begChr", "endChr", "type")
@@ -159,7 +157,7 @@ class Pr(Tag2Method):
     def do_common(self, elm):
         stag = elm.tag.replace(OMML_NS, "")
         if stag in self.__val_tags:
-            t = elm.get("{0}val".format(OMML_NS))
+            t = elm.get(f"{OMML_NS}val")
             self.__innerdict[stag] = t
         return None
 
@@ -248,7 +246,6 @@ class oMath2Latex(Tag2Method):
         """
         the Pre-Sub-Superscript object -- Not support yet
         """
-        pass
 
     def do_sub(self, elm):
         text = self.process_children(elm)
@@ -331,7 +328,7 @@ class oMath2Latex(Tag2Method):
         t_dict = self.process_children_dict(elm, include=("e", "lim"))
         latex_s = LIM_FUNC.get(t_dict["e"])
         if not latex_s:
-            raise NotSupport("Not support lim %s" % t_dict["e"])
+            raise RuntimeError("Not support lim {}".format(t_dict["e"]))
         else:
             return latex_s.format(lim=t_dict.get("lim"))
 
@@ -413,7 +410,7 @@ class oMath2Latex(Tag2Method):
         """
         _str = []
         _base_str = []
-        found_text = elm.findtext("./{0}t".format(OMML_NS))
+        found_text = elm.findtext(f"./{OMML_NS}t")
         if found_text:
             for s in found_text:
                 out_latex_str = self.process_unicode(s)

docling/backend/html_backend.py

@@ -26,6 +26,8 @@ _log = logging.getLogger(__name__)
 
 # tags that generate NodeItem elements
 TAGS_FOR_NODE_ITEMS: Final = [
+    "address",
+    "details",
     "h1",
     "h2",
     "h3",
@@ -38,6 +40,7 @@ TAGS_FOR_NODE_ITEMS: Final = [
     "ul",
     "ol",
     "li",
+    "summary",
     "table",
     "figure",
     "img",
@@ -55,7 +58,7 @@ class HTMLDocumentBackend(DeclarativeDocumentBackend):
         self.max_levels = 10
         self.level = 0
         self.parents: dict[int, Optional[Union[DocItem, GroupItem]]] = {}
-        for i in range(0, self.max_levels):
+        for i in range(self.max_levels):
             self.parents[i] = None
 
         try:
@@ -126,7 +129,6 @@ class HTMLDocumentBackend(DeclarativeDocumentBackend):
         return doc
 
     def walk(self, tag: Tag, doc: DoclingDocument) -> None:
-
         # Iterate over elements in the body of the document
         text: str = ""
         for element in tag.children:
@@ -135,7 +137,7 @@ class HTMLDocumentBackend(DeclarativeDocumentBackend):
                     self.analyze_tag(cast(Tag, element), doc)
                 except Exception as exc_child:
                     _log.error(
-                        f"Error processing child from tag {tag.name}: {repr(exc_child)}"
+                        f"Error processing child from tag {tag.name}: {exc_child!r}"
                     )
                     raise exc_child
             elif isinstance(element, NavigableString) and not isinstance(
@@ -147,7 +149,7 @@ class HTMLDocumentBackend(DeclarativeDocumentBackend):
                     item for item in element.next_siblings if isinstance(item, Tag)
                 ]
                 if element.next_sibling is None or any(
-                    [item.name in TAGS_FOR_NODE_ITEMS for item in siblings]
+                    item.name in TAGS_FOR_NODE_ITEMS for item in siblings
                 ):
                     text = text.strip()
                     if text and tag.name in ["div"]:
@@ -164,7 +166,7 @@ class HTMLDocumentBackend(DeclarativeDocumentBackend):
     def analyze_tag(self, tag: Tag, doc: DoclingDocument) -> None:
         if tag.name in ["h1", "h2", "h3", "h4", "h5", "h6"]:
             self.handle_header(tag, doc)
-        elif tag.name in ["p"]:
+        elif tag.name in ["p", "address", "summary"]:
             self.handle_paragraph(tag, doc)
         elif tag.name in ["pre", "code"]:
             self.handle_code(tag, doc)
@@ -178,6 +180,8 @@ class HTMLDocumentBackend(DeclarativeDocumentBackend):
             self.handle_figure(tag, doc)
         elif tag.name == "img":
             self.handle_image(tag, doc)
+        elif tag.name == "details":
+            self.handle_details(tag, doc)
         else:
             self.walk(tag, doc)
 
@@ -202,6 +206,21 @@ class HTMLDocumentBackend(DeclarativeDocumentBackend):
 
         return ["".join(result) + " "]
 
+    def handle_details(self, element: Tag, doc: DoclingDocument) -> None:
+        """Handle details tag (details) and its content."""
+
+        self.parents[self.level + 1] = doc.add_group(
+            name="details",
+            label=GroupLabel.SECTION,
+            parent=self.parents[self.level],
+            content_layer=self.content_layer,
+        )
+
+        self.level += 1
+        self.walk(element, doc)
+        self.parents[self.level + 1] = None
+        self.level -= 1
+
     def handle_header(self, element: Tag, doc: DoclingDocument) -> None:
         """Handles header tags (h1, h2, etc.)."""
         hlevel = int(element.name.replace("h", ""))
@@ -222,7 +241,6 @@ class HTMLDocumentBackend(DeclarativeDocumentBackend):
             )
         else:
             if hlevel > self.level:
-
                 # add invisible group
                 for i in range(self.level + 1, hlevel):
                     self.parents[i] = doc.add_group(
@@ -234,7 +252,6 @@ class HTMLDocumentBackend(DeclarativeDocumentBackend):
                 self.level = hlevel
 
             elif hlevel < self.level:
-
                 # remove the tail
                 for key in self.parents.keys():
                     if key > hlevel:
@@ -261,7 +278,7 @@ class HTMLDocumentBackend(DeclarativeDocumentBackend):
             )
 
     def handle_paragraph(self, element: Tag, doc: DoclingDocument) -> None:
-        """Handles paragraph tags (p)."""
+        """Handles paragraph tags (p) or equivalent ones."""
         if element.text is None:
             return
         text = element.text.strip()
@@ -360,7 +377,7 @@ class HTMLDocumentBackend(DeclarativeDocumentBackend):
             marker = ""
             enumerated = False
             if parent_label == GroupLabel.ORDERED_LIST:
-                marker = f"{str(index_in_list)}."
+                marker = f"{index_in_list!s}."
                 enumerated = True
             doc.add_list_item(
                 text=text,

docling/backend/md_backend.py

@@ -83,7 +83,7 @@ class MarkdownDocumentBackend(DeclarativeDocumentBackend):
                 # otherwise they represent emphasis (bold or italic)
                 self.markdown = self._shorten_underscore_sequences(text_stream)
             if isinstance(self.path_or_stream, Path):
-                with open(self.path_or_stream, "r", encoding="utf-8") as f:
+                with open(self.path_or_stream, encoding="utf-8") as f:
                     md_content = f.read()
                     # remove invalid sequences
                     # very long sequences of underscores will lead to unnecessary long processing times.
@@ -168,7 +168,7 @@ class MarkdownDocumentBackend(DeclarativeDocumentBackend):
             )
         self.inline_texts = []
 
-    def _iterate_elements(
+    def _iterate_elements(  # noqa: C901
         self,
         element: marko.element.Element,
         depth: int,
@@ -176,7 +176,6 @@ class MarkdownDocumentBackend(DeclarativeDocumentBackend):
         visited: Set[marko.element.Element],
         parent_item: Optional[NodeItem] = None,
     ):
-
         if element in visited:
             return
 
@@ -236,7 +235,7 @@ class MarkdownDocumentBackend(DeclarativeDocumentBackend):
             if has_non_empty_list_items:
                 label = GroupLabel.ORDERED_LIST if element.ordered else GroupLabel.LIST
                 parent_item = doc.add_group(
-                    label=label, name=f"list", parent=parent_item
+                    label=label, name="list", parent=parent_item
                 )
 
         elif (
@@ -320,7 +319,7 @@ class MarkdownDocumentBackend(DeclarativeDocumentBackend):
             self._html_blocks += 1
             self._process_inline_text(parent_item, doc)
             self._close_table(doc)
-            _log.debug("HTML Block: {}".format(element))
+            _log.debug(f"HTML Block: {element}")
             if (
                 len(element.body) > 0
             ):  # If Marko doesn't return any content for HTML block, skip it
@@ -332,7 +331,7 @@ class MarkdownDocumentBackend(DeclarativeDocumentBackend):
         else:
             if not isinstance(element, str):
                 self._close_table(doc)
-                _log.debug("Some other element: {}".format(element))
+                _log.debug(f"Some other element: {element}")
 
         processed_block_types = (
             marko.block.Heading,
@@ -398,7 +397,6 @@ class MarkdownDocumentBackend(DeclarativeDocumentBackend):
 
             # if HTML blocks were detected, export to HTML and delegate to HTML backend
             if self._html_blocks > 0:
-
                 # export to HTML
                 html_backend_cls = HTMLDocumentBackend
                 html_str = doc.export_to_html()

docling/backend/msexcel_backend.py

@@ -184,7 +184,6 @@ class MsExcelDocumentBackend(DeclarativeDocumentBackend, PaginatedDocumentBacken
         """
 
         if self.workbook is not None:
-
             # Iterate over all sheets
             for sheet_name in self.workbook.sheetnames:
                 _log.info(f"Processing sheet: {sheet_name}")
@@ -253,7 +252,6 @@ class MsExcelDocumentBackend(DeclarativeDocumentBackend, PaginatedDocumentBacken
                 )
 
                 for excel_cell in excel_table.data:
-
                     cell = TableCell(
                         text=excel_cell.text,
                         row_span=excel_cell.row_span,
@@ -303,7 +301,6 @@ class MsExcelDocumentBackend(DeclarativeDocumentBackend, PaginatedDocumentBacken
         # Iterate over all cells in the sheet
         for ri, row in enumerate(sheet.iter_rows(values_only=False)):
             for rj, cell in enumerate(row):
-
                 # Skip empty or already visited cells
                 if cell.value is None or (ri, rj) in visited:
                     continue
@@ -342,7 +339,6 @@ class MsExcelDocumentBackend(DeclarativeDocumentBackend, PaginatedDocumentBacken
         visited_cells: set[tuple[int, int]] = set()
         for ri in range(start_row, max_row + 1):
             for rj in range(start_col, max_col + 1):
-
                 cell = sheet.cell(row=ri + 1, column=rj + 1)  # 1-based indexing
 
                 # Check if the cell belongs to a merged range
@@ -350,14 +346,12 @@ class MsExcelDocumentBackend(DeclarativeDocumentBackend, PaginatedDocumentBacken
                 col_span = 1
 
                 for merged_range in sheet.merged_cells.ranges:
-
                     if (
                         merged_range.min_row <= ri + 1
                         and ri + 1 <= merged_range.max_row
                         and merged_range.min_col <= rj + 1
                         and rj + 1 <= merged_range.max_col
                     ):
-
                         row_span = merged_range.max_row - merged_range.min_row + 1
                         col_span = merged_range.max_col - merged_range.min_col + 1
                         break
@@ -499,7 +493,7 @@ class MsExcelDocumentBackend(DeclarativeDocumentBackend, PaginatedDocumentBacken
                             ),
                         ),
                     )
-                except:
+                except Exception:
                     _log.error("could not extract the image from excel sheets")
 
         return doc

docling/backend/mspowerpoint_backend.py

@@ -120,13 +120,12 @@ class MsPowerpointDocumentBackend(DeclarativeDocumentBackend, PaginatedDocumentB
 
         return prov
 
-    def handle_text_elements(self, shape, parent_slide, slide_ind, doc, slide_size):
+    def handle_text_elements(self, shape, parent_slide, slide_ind, doc, slide_size):  # noqa: C901
         is_a_list = False
         is_list_group_created = False
         enum_list_item_value = 0
         new_list = None
         bullet_type = "None"
-        list_text = ""
         list_label = GroupLabel.LIST
         doc_label = DocItemLabel.LIST_ITEM
         prov = self.generate_prov(shape, slide_ind, shape.text.strip(), slide_size)
@@ -243,7 +242,7 @@ class MsPowerpointDocumentBackend(DeclarativeDocumentBackend, PaginatedDocumentB
                     enum_marker = str(enum_list_item_value) + "."
                 if not is_list_group_created:
                     new_list = doc.add_group(
-                        label=list_label, name=f"list", parent=parent_slide
+                        label=list_label, name="list", parent=parent_slide
                     )
                     is_list_group_created = True
                 doc.add_list_item(
@@ -368,11 +367,9 @@ class MsPowerpointDocumentBackend(DeclarativeDocumentBackend, PaginatedDocumentB
         slide_width = pptx_obj.slide_width
         slide_height = pptx_obj.slide_height
 
-        text_content = []  # type: ignore
-
         max_levels = 10
         parents = {}  # type: ignore
-        for i in range(0, max_levels):
+        for i in range(max_levels):
             parents[i] = None
 
         # Loop through each slide
@@ -383,7 +380,7 @@ class MsPowerpointDocumentBackend(DeclarativeDocumentBackend, PaginatedDocumentB
             )
 
             slide_size = Size(width=slide_width, height=slide_height)
-            parent_page = doc.add_page(page_no=slide_ind + 1, size=slide_size)
+            doc.add_page(page_no=slide_ind + 1, size=slide_size)
 
             def handle_shapes(shape, parent_slide, slide_ind, doc, slide_size):
                 handle_groups(shape, parent_slide, slide_ind, doc, slide_size)

docling/backend/msword_backend.py

@@ -158,7 +158,7 @@ class MsWordDocumentBackend(DeclarativeDocumentBackend):
     def _get_level(self) -> int:
         """Return the first None index."""
         for k, v in self.parents.items():
-            if k >= 0 and v == None:
+            if k >= 0 and v is None:
                 return k
         return 0
 
@@ -418,7 +418,7 @@ class MsWordDocumentBackend(DeclarativeDocumentBackend):
             else prev_parent
         )
 
-    def _handle_text_elements(
+    def _handle_text_elements(  # noqa: C901
         self,
         element: BaseOxmlElement,
         docx_obj: DocxDocument,
@@ -812,7 +812,7 @@ class MsWordDocumentBackend(DeclarativeDocumentBackend):
                     f" col {col_idx} grid_span {cell.grid_span} grid_cols_before {row.grid_cols_before}"
                 )
                 if cell is None or cell._tc in cell_set:
-                    _log.debug(f"  skipped since repeated content")
+                    _log.debug("  skipped since repeated content")
                     col_idx += cell.grid_span
                     continue
                 else:
@@ -879,7 +879,7 @@ class MsWordDocumentBackend(DeclarativeDocumentBackend):
                     image=ImageRef.from_pil(image=pil_image, dpi=72),
                     caption=None,
                 )
-            except (UnidentifiedImageError, OSError) as e:
+            except (UnidentifiedImageError, OSError):
                 _log.warning("Warning: image cannot be loaded by Pillow")
                 doc.add_picture(
                     parent=self.parents[level - 1],

docling/backend/pdf_backend.py

@@ -1,7 +1,8 @@
 from abc import ABC, abstractmethod
+from collections.abc import Iterable
 from io import BytesIO
 from pathlib import Path
-from typing import Iterable, Optional, Set, Union
+from typing import Optional, Set, Union
 
 from docling_core.types.doc import BoundingBox, Size
 from docling_core.types.doc.page import SegmentedPdfPage, TextCell

docling/backend/pypdfium2_backend.py

@@ -1,8 +1,9 @@
 import logging
 import random
+from collections.abc import Iterable
 from io import BytesIO
 from pathlib import Path
-from typing import TYPE_CHECKING, Iterable, List, Optional, Union
+from typing import TYPE_CHECKING, List, Optional, Union
 
 import pypdfium2 as pdfium
 import pypdfium2.raw as pdfium_c
@@ -29,7 +30,7 @@ class PyPdfiumPageBackend(PdfPageBackend):
         self.valid = True  # No better way to tell from pypdfium.
         try:
             self._ppage: pdfium.PdfPage = pdfium_doc[page_no]
-        except PdfiumError as e:
+        except PdfiumError:
             _log.info(
                 f"An exception occurred when loading page {page_no} of document {document_hash}.",
                 exc_info=True,
@@ -225,7 +226,6 @@ class PyPdfiumPageBackend(PdfPageBackend):
     def get_page_image(
         self, scale: float = 1, cropbox: Optional[BoundingBox] = None
     ) -> Image.Image:
-
         page_size = self.get_size()
 
         if not cropbox: