docling 2.23.0__py3-none-any.whl → 2.24.0__py3-none-any.whl

docling/backend/xml/uspto_backend.py

@@ -14,7 +14,7 @@ from abc import ABC, abstractmethod
 from enum import Enum, unique
 from io import BytesIO
 from pathlib import Path
-from typing import Any, Final, Optional, Union
+from typing import Final, Optional, Union
 
 from bs4 import BeautifulSoup, Tag
 from docling_core.types.doc import (
@@ -1406,6 +1406,10 @@ class XmlTable:
     http://oasis-open.org/specs/soextblx.dtd
     """
 
+    class ColInfo(TypedDict):
+        ncols: int
+        colinfo: list[dict]
+
     class MinColInfoType(TypedDict):
         offset: list[int]
         colwidth: list[int]
@@ -1425,7 +1429,7 @@ class XmlTable:
         self.empty_text = ""
         self._soup = BeautifulSoup(input, features="xml")
 
-    def _create_tg_range(self, tgs: list[dict[str, Any]]) -> dict[int, ColInfoType]:
+    def _create_tg_range(self, tgs: list[ColInfo]) -> dict[int, ColInfoType]:
         """Create a unified range along the table groups.
 
         Args:
@@ -1532,19 +1536,26 @@ class XmlTable:
         Returns:
             A docling table object.
         """
-        tgs_align = []
-        tg_secs = table.find_all("tgroup")
+        tgs_align: list[XmlTable.ColInfo] = []
+        tg_secs = table("tgroup")
         if tg_secs:
             for tg_sec in tg_secs:
-                ncols = tg_sec.get("cols", None)
-                if ncols:
-                    ncols = int(ncols)
-                tg_align = {"ncols": ncols, "colinfo": []}
-                cs_secs = tg_sec.find_all("colspec")
+                if not isinstance(tg_sec, Tag):
+                    continue
+                col_val = tg_sec.get("cols")
+                ncols = (
+                    int(col_val)
+                    if isinstance(col_val, str) and col_val.isnumeric()
+                    else 1
+                )
+                tg_align: XmlTable.ColInfo = {"ncols": ncols, "colinfo": []}
+                cs_secs = tg_sec("colspec")
                 if cs_secs:
                     for cs_sec in cs_secs:
-                        colname = cs_sec.get("colname", None)
-                        colwidth = cs_sec.get("colwidth", None)
+                        if not isinstance(cs_sec, Tag):
+                            continue
+                        colname = cs_sec.get("colname")
+                        colwidth = cs_sec.get("colwidth")
                         tg_align["colinfo"].append(
                             {"colname": colname, "colwidth": colwidth}
                         )
@@ -1565,16 +1576,23 @@ class XmlTable:
         table_data: list[TableCell] = []
         i_row_global = 0
         is_row_empty: bool = True
-        tg_secs = table.find_all("tgroup")
+        tg_secs = table("tgroup")
         if tg_secs:
             for itg, tg_sec in enumerate(tg_secs):
+                if not isinstance(tg_sec, Tag):
+                    continue
                 tg_range = tgs_range[itg]
-                row_secs = tg_sec.find_all(["row", "tr"])
+                row_secs = tg_sec(["row", "tr"])
 
                 if row_secs:
                     for row_sec in row_secs:
-                        entry_secs = row_sec.find_all(["entry", "td"])
-                        is_header: bool = row_sec.parent.name in ["thead"]
+                        if not isinstance(row_sec, Tag):
+                            continue
+                        entry_secs = row_sec(["entry", "td"])
+                        is_header: bool = (
+                            row_sec.parent is not None
+                            and row_sec.parent.name == "thead"
+                        )
 
                         ncols = 0
                         local_row: list[TableCell] = []
@@ -1582,23 +1600,26 @@ class XmlTable:
                         if entry_secs:
                             wrong_nbr_cols = False
                             for ientry, entry_sec in enumerate(entry_secs):
+                                if not isinstance(entry_sec, Tag):
+                                    continue
                                 text = entry_sec.get_text().strip()
 
                                 # start-end
-                                namest = entry_sec.attrs.get("namest", None)
-                                nameend = entry_sec.attrs.get("nameend", None)
-                                if isinstance(namest, str) and namest.isnumeric():
-                                    namest = int(namest)
-                                else:
-                                    namest = ientry + 1
+                                namest = entry_sec.get("namest")
+                                nameend = entry_sec.get("nameend")
+                                start = (
+                                    int(namest)
+                                    if isinstance(namest, str) and namest.isnumeric()
+                                    else ientry + 1
+                                )
                                 if isinstance(nameend, str) and nameend.isnumeric():
-                                    nameend = int(nameend)
+                                    end = int(nameend)
                                     shift = 0
                                 else:
-                                    nameend = ientry + 2
+                                    end = ientry + 2
                                     shift = 1
 
-                                if nameend > len(tg_range["cell_offst"]):
+                                if end > len(tg_range["cell_offst"]):
                                     wrong_nbr_cols = True
                                     self.nbr_messages += 1
                                     if self.nbr_messages <= self.max_nbr_messages:
@@ -1608,8 +1629,8 @@ class XmlTable:
                                     break
 
                                 range_ = [
-                                    tg_range["cell_offst"][namest - 1],
-                                    tg_range["cell_offst"][nameend - 1] - shift,
+                                    tg_range["cell_offst"][start - 1],
+                                    tg_range["cell_offst"][end - 1] - shift,
                                 ]
 
                                 # add row and replicate cell if needed
@@ -1668,7 +1689,7 @@ class XmlTable:
             A docling table data.
         """
         section = self._soup.find("table")
-        if section is not None:
+        if isinstance(section, Tag):
             table = self._parse_table(section)
             if table.num_rows == 0 or table.num_cols == 0:
                 _log.warning("The parsed USPTO table is empty")

docling/models/page_assemble_model.py

@@ -52,6 +52,14 @@ class PageAssembleModel(BasePageModel):
 
         sanitized_text = "".join(lines)
 
+        # Text normalization
+        sanitized_text = sanitized_text.replace("⁄", "/")
+        sanitized_text = sanitized_text.replace("’", "'")
+        sanitized_text = sanitized_text.replace("‘", "'")
+        sanitized_text = sanitized_text.replace("“", '"')
+        sanitized_text = sanitized_text.replace("”", '"')
+        sanitized_text = sanitized_text.replace("•", "·")
+
         return sanitized_text.strip()  # Strip any leading or trailing whitespace
 
     def __call__(

docling/models/readingorder_model.py

@@ -0,0 +1,389 @@
+import copy
+import random
+from pathlib import Path
+from typing import Dict, List
+
+from docling_core.types.doc import (
+    BoundingBox,
+    CoordOrigin,
+    DocItem,
+    DocItemLabel,
+    DoclingDocument,
+    DocumentOrigin,
+    GroupLabel,
+    NodeItem,
+    ProvenanceItem,
+    RefItem,
+    TableData,
+)
+from docling_core.types.doc.document import ContentLayer
+from docling_core.types.legacy_doc.base import Ref
+from docling_core.types.legacy_doc.document import BaseText
+from docling_ibm_models.reading_order.reading_order_rb import (
+    PageElement as ReadingOrderPageElement,
+)
+from docling_ibm_models.reading_order.reading_order_rb import ReadingOrderPredictor
+from PIL import ImageDraw
+from pydantic import BaseModel, ConfigDict
+
+from docling.datamodel.base_models import (
+    BasePageElement,
+    Cluster,
+    ContainerElement,
+    FigureElement,
+    Table,
+    TextElement,
+)
+from docling.datamodel.document import ConversionResult
+from docling.datamodel.settings import settings
+from docling.utils.profiling import ProfilingScope, TimeRecorder
+
+
+class ReadingOrderOptions(BaseModel):
+    model_config = ConfigDict(protected_namespaces=())
+
+    model_names: str = ""  # e.g. "language;term;reference"
+
+
+class ReadingOrderModel:
+    def __init__(self, options: ReadingOrderOptions):
+        self.options = options
+        self.ro_model = ReadingOrderPredictor()
+
+    def _assembled_to_readingorder_elements(
+        self, conv_res: ConversionResult
+    ) -> List[ReadingOrderPageElement]:
+
+        elements: List[ReadingOrderPageElement] = []
+        page_no_to_pages = {p.page_no: p for p in conv_res.pages}
+
+        for element in conv_res.assembled.elements:
+
+            page_height = page_no_to_pages[element.page_no].size.height  # type: ignore
+            bbox = element.cluster.bbox.to_bottom_left_origin(page_height)
+            text = element.text or ""
+
+            elements.append(
+                ReadingOrderPageElement(
+                    cid=len(elements),
+                    ref=RefItem(cref=f"#/{element.page_no}/{element.cluster.id}"),
+                    text=text,
+                    page_no=element.page_no,
+                    page_size=page_no_to_pages[element.page_no].size,
+                    label=element.label,
+                    l=bbox.l,
+                    r=bbox.r,
+                    b=bbox.b,
+                    t=bbox.t,
+                    coord_origin=bbox.coord_origin,
+                )
+            )
+
+        return elements
+
+    def _add_child_elements(
+        self, element: BasePageElement, doc_item: NodeItem, doc: DoclingDocument
+    ):
+
+        child: Cluster
+        for child in element.cluster.children:
+            c_label = child.label
+            c_bbox = child.bbox.to_bottom_left_origin(
+                doc.pages[element.page_no + 1].size.height
+            )
+            c_text = " ".join(
+                [
+                    cell.text.replace("\x02", "-").strip()
+                    for cell in child.cells
+                    if len(cell.text.strip()) > 0
+                ]
+            )
+
+            c_prov = ProvenanceItem(
+                page_no=element.page_no + 1, charspan=(0, len(c_text)), bbox=c_bbox
+            )
+            if c_label == DocItemLabel.LIST_ITEM:
+                # TODO: Infer if this is a numbered or a bullet list item
+                doc.add_list_item(parent=doc_item, text=c_text, prov=c_prov)
+            elif c_label == DocItemLabel.SECTION_HEADER:
+                doc.add_heading(parent=doc_item, text=c_text, prov=c_prov)
+            else:
+                doc.add_text(parent=doc_item, label=c_label, text=c_text, prov=c_prov)
+
+    def _readingorder_elements_to_docling_doc(
+        self,
+        conv_res: ConversionResult,
+        ro_elements: List[ReadingOrderPageElement],
+        el_to_captions_mapping: Dict[int, List[int]],
+        el_to_footnotes_mapping: Dict[int, List[int]],
+        el_merges_mapping: Dict[int, List[int]],
+    ) -> DoclingDocument:
+
+        id_to_elem = {
+            RefItem(cref=f"#/{elem.page_no}/{elem.cluster.id}").cref: elem
+            for elem in conv_res.assembled.elements
+        }
+        cid_to_rels = {rel.cid: rel for rel in ro_elements}
+
+        origin = DocumentOrigin(
+            mimetype="application/pdf",
+            filename=conv_res.input.file.name,
+            binary_hash=conv_res.input.document_hash,
+        )
+        doc_name = Path(origin.filename).stem
+        out_doc: DoclingDocument = DoclingDocument(name=doc_name, origin=origin)
+
+        for page in conv_res.pages:
+            page_no = page.page_no + 1
+            size = page.size
+
+            assert size is not None
+
+            out_doc.add_page(page_no=page_no, size=size)
+
+        current_list = None
+        skippable_cids = {
+            cid
+            for mapping in (
+                el_to_captions_mapping,
+                el_to_footnotes_mapping,
+                el_merges_mapping,
+            )
+            for lst in mapping.values()
+            for cid in lst
+        }
+
+        page_no_to_pages = {p.page_no: p for p in conv_res.pages}
+
+        for rel in ro_elements:
+            if rel.cid in skippable_cids:
+                continue
+            element = id_to_elem[rel.ref.cref]
+
+            page_height = page_no_to_pages[element.page_no].size.height  # type: ignore
+
+            if isinstance(element, TextElement):
+                if element.label == DocItemLabel.CODE:
+                    cap_text = element.text
+                    prov = ProvenanceItem(
+                        page_no=element.page_no + 1,
+                        charspan=(0, len(cap_text)),
+                        bbox=element.cluster.bbox.to_bottom_left_origin(page_height),
+                    )
+                    code_item = out_doc.add_code(text=cap_text, prov=prov)
+
+                    if rel.cid in el_to_captions_mapping.keys():
+                        for caption_cid in el_to_captions_mapping[rel.cid]:
+                            caption_elem = id_to_elem[cid_to_rels[caption_cid].ref.cref]
+                            new_cap_item = self._add_caption_or_footnote(
+                                caption_elem, out_doc, code_item, page_height
+                            )
+
+                            code_item.captions.append(new_cap_item.get_ref())
+
+                    if rel.cid in el_to_footnotes_mapping.keys():
+                        for footnote_cid in el_to_footnotes_mapping[rel.cid]:
+                            footnote_elem = id_to_elem[
+                                cid_to_rels[footnote_cid].ref.cref
+                            ]
+                            new_footnote_item = self._add_caption_or_footnote(
+                                footnote_elem, out_doc, code_item, page_height
+                            )
+
+                            code_item.footnotes.append(new_footnote_item.get_ref())
+                else:
+
+                    new_item, current_list = self._handle_text_element(
+                        element, out_doc, current_list, page_height
+                    )
+
+                    if rel.cid in el_merges_mapping.keys():
+                        for merged_cid in el_merges_mapping[rel.cid]:
+                            merged_elem = id_to_elem[cid_to_rels[merged_cid].ref.cref]
+
+                            self._merge_elements(
+                                element, merged_elem, new_item, page_height
+                            )
+
+            elif isinstance(element, Table):
+
+                tbl_data = TableData(
+                    num_rows=element.num_rows,
+                    num_cols=element.num_cols,
+                    table_cells=element.table_cells,
+                )
+
+                prov = ProvenanceItem(
+                    page_no=element.page_no + 1,
+                    charspan=(0, 0),
+                    bbox=element.cluster.bbox.to_bottom_left_origin(page_height),
+                )
+
+                tbl = out_doc.add_table(
+                    data=tbl_data, prov=prov, label=element.cluster.label
+                )
+
+                if rel.cid in el_to_captions_mapping.keys():
+                    for caption_cid in el_to_captions_mapping[rel.cid]:
+                        caption_elem = id_to_elem[cid_to_rels[caption_cid].ref.cref]
+                        new_cap_item = self._add_caption_or_footnote(
+                            caption_elem, out_doc, tbl, page_height
+                        )
+
+                        tbl.captions.append(new_cap_item.get_ref())
+
+                if rel.cid in el_to_footnotes_mapping.keys():
+                    for footnote_cid in el_to_footnotes_mapping[rel.cid]:
+                        footnote_elem = id_to_elem[cid_to_rels[footnote_cid].ref.cref]
+                        new_footnote_item = self._add_caption_or_footnote(
+                            footnote_elem, out_doc, tbl, page_height
+                        )
+
+                        tbl.footnotes.append(new_footnote_item.get_ref())
+
+                # TODO: Consider adding children of Table.
+
+            elif isinstance(element, FigureElement):
+                cap_text = ""
+                prov = ProvenanceItem(
+                    page_no=element.page_no + 1,
+                    charspan=(0, len(cap_text)),
+                    bbox=element.cluster.bbox.to_bottom_left_origin(page_height),
+                )
+                pic = out_doc.add_picture(prov=prov)
+
+                if rel.cid in el_to_captions_mapping.keys():
+                    for caption_cid in el_to_captions_mapping[rel.cid]:
+                        caption_elem = id_to_elem[cid_to_rels[caption_cid].ref.cref]
+                        new_cap_item = self._add_caption_or_footnote(
+                            caption_elem, out_doc, pic, page_height
+                        )
+
+                        pic.captions.append(new_cap_item.get_ref())
+
+                if rel.cid in el_to_footnotes_mapping.keys():
+                    for footnote_cid in el_to_footnotes_mapping[rel.cid]:
+                        footnote_elem = id_to_elem[cid_to_rels[footnote_cid].ref.cref]
+                        new_footnote_item = self._add_caption_or_footnote(
+                            footnote_elem, out_doc, pic, page_height
+                        )
+
+                        pic.footnotes.append(new_footnote_item.get_ref())
+
+                self._add_child_elements(element, pic, out_doc)
+
+            elif isinstance(element, ContainerElement):  # Form, KV region
+                label = element.label
+                group_label = GroupLabel.UNSPECIFIED
+                if label == DocItemLabel.FORM:
+                    group_label = GroupLabel.FORM_AREA
+                elif label == DocItemLabel.KEY_VALUE_REGION:
+                    group_label = GroupLabel.KEY_VALUE_AREA
+
+                container_el = out_doc.add_group(label=group_label)
+
+                self._add_child_elements(element, container_el, out_doc)
+
+        return out_doc
+
+    def _add_caption_or_footnote(self, elem, out_doc, parent, page_height):
+        assert isinstance(elem, TextElement)
+        text = elem.text
+        prov = ProvenanceItem(
+            page_no=elem.page_no + 1,
+            charspan=(0, len(text)),
+            bbox=elem.cluster.bbox.to_bottom_left_origin(page_height),
+        )
+        new_item = out_doc.add_text(
+            label=elem.label, text=text, prov=prov, parent=parent
+        )
+        return new_item
+
+    def _handle_text_element(self, element, out_doc, current_list, page_height):
+        cap_text = element.text
+
+        prov = ProvenanceItem(
+            page_no=element.page_no + 1,
+            charspan=(0, len(cap_text)),
+            bbox=element.cluster.bbox.to_bottom_left_origin(page_height),
+        )
+        label = element.label
+        if label == DocItemLabel.LIST_ITEM:
+            if current_list is None:
+                current_list = out_doc.add_group(label=GroupLabel.LIST, name="list")
+
+            # TODO: Infer if this is a numbered or a bullet list item
+            new_item = out_doc.add_list_item(
+                text=cap_text, enumerated=False, prov=prov, parent=current_list
+            )
+        elif label == DocItemLabel.SECTION_HEADER:
+            current_list = None
+
+            new_item = out_doc.add_heading(text=cap_text, prov=prov)
+        elif label == DocItemLabel.FORMULA:
+            current_list = None
+
+            new_item = out_doc.add_text(
+                label=DocItemLabel.FORMULA, text="", orig=cap_text, prov=prov
+            )
+        else:
+            current_list = None
+
+            content_layer = ContentLayer.BODY
+            if element.label in [DocItemLabel.PAGE_HEADER, DocItemLabel.PAGE_FOOTER]:
+                content_layer = ContentLayer.FURNITURE
+
+            new_item = out_doc.add_text(
+                label=element.label,
+                text=cap_text,
+                prov=prov,
+                content_layer=content_layer,
+            )
+        return new_item, current_list
+
+    def _merge_elements(self, element, merged_elem, new_item, page_height):
+        assert isinstance(
+            merged_elem, type(element)
+        ), "Merged element must be of same type as element."
+        assert (
+            merged_elem.label == new_item.label
+        ), "Labels of merged elements must match."
+        prov = ProvenanceItem(
+            page_no=element.page_no + 1,
+            charspan=(
+                len(new_item.text) + 1,
+                len(new_item.text) + 1 + len(merged_elem.text),
+            ),
+            bbox=element.cluster.bbox.to_bottom_left_origin(page_height),
+        )
+        new_item.text += f" {merged_elem.text}"
+        new_item.orig += f" {merged_elem.text}"  # TODO: This is incomplete, we don't have the `orig` field of the merged element.
+        new_item.prov.append(prov)
+
+    def __call__(self, conv_res: ConversionResult) -> DoclingDocument:
+        with TimeRecorder(conv_res, "glm", scope=ProfilingScope.DOCUMENT):
+            page_elements = self._assembled_to_readingorder_elements(conv_res)
+
+            # Apply reading order
+            sorted_elements = self.ro_model.predict_reading_order(
+                page_elements=page_elements
+            )
+            el_to_captions_mapping = self.ro_model.predict_to_captions(
+                sorted_elements=sorted_elements
+            )
+            el_to_footnotes_mapping = self.ro_model.predict_to_footnotes(
+                sorted_elements=sorted_elements
+            )
+            el_merges_mapping = self.ro_model.predict_merges(
+                sorted_elements=sorted_elements
+            )
+
+            docling_doc: DoclingDocument = self._readingorder_elements_to_docling_doc(
+                conv_res,
+                sorted_elements,
+                el_to_captions_mapping,
+                el_to_footnotes_mapping,
+                el_merges_mapping,
+            )
+
+        return docling_doc

docling/models/tesseract_ocr_cli_model.py

@@ -114,7 +114,9 @@ class TesseractOcrCliModel(BaseOcrModel):
         # _log.info("df: ", df.head())
 
         # Filter rows that contain actual text (ignore header or empty rows)
-        df_filtered = df[df["text"].notnull() & (df["text"].str.strip() != "")]
+        df_filtered = df[
+            df["text"].notnull() & (df["text"].apply(str).str.strip() != "")
+        ]
 
         return df_filtered
 

docling/pipeline/standard_pdf_pipeline.py

@@ -27,7 +27,6 @@ from docling.models.document_picture_classifier import (
     DocumentPictureClassifier,
     DocumentPictureClassifierOptions,
 )
-from docling.models.ds_glm_model import GlmModel, GlmOptions
 from docling.models.easyocr_model import EasyOcrModel
 from docling.models.layout_model import LayoutModel
 from docling.models.ocr_mac_model import OcrMacModel
@@ -40,6 +39,7 @@ from docling.models.picture_description_api_model import PictureDescriptionApiMo
 from docling.models.picture_description_base_model import PictureDescriptionBaseModel
 from docling.models.picture_description_vlm_model import PictureDescriptionVlmModel
 from docling.models.rapid_ocr_model import RapidOcrModel
+from docling.models.readingorder_model import ReadingOrderModel, ReadingOrderOptions
 from docling.models.table_structure_model import TableStructureModel
 from docling.models.tesseract_ocr_cli_model import TesseractOcrCliModel
 from docling.models.tesseract_ocr_model import TesseractOcrModel
@@ -76,7 +76,7 @@ class StandardPdfPipeline(PaginatedPipeline):
             or self.pipeline_options.generate_table_images
         )
 
-        self.glm_model = GlmModel(options=GlmOptions())
+        self.glm_model = ReadingOrderModel(options=ReadingOrderOptions())
 
         if (ocr_model := self.get_ocr_model(artifacts_path=artifacts_path)) is None:
             raise RuntimeError(

{docling-2.23.0.dist-info → docling-2.24.0.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: docling
-Version: 2.23.0
+Version: 2.24.0
 Summary: SDK and CLI for parsing PDF, DOCX, HTML, and more, to a unified document representation for powering downstream workflows such as gen AI applications.
 Home-page: https://github.com/DS4SD/docling
 License: MIT
@@ -25,11 +25,10 @@ Provides-Extra: ocrmac
 Provides-Extra: rapidocr
 Provides-Extra: tesserocr
 Provides-Extra: vlm
-Requires-Dist: beautifulsoup4 (>=4.12.3,<4.13.0)
+Requires-Dist: beautifulsoup4 (>=4.12.3,<5.0.0)
 Requires-Dist: certifi (>=2024.7.4)
-Requires-Dist: deepsearch-glm (>=1.0.0,<2.0.0)
 Requires-Dist: docling-core[chunking] (>=2.19.0,<3.0.0)
-Requires-Dist: docling-ibm-models (>=3.3.0,<4.0.0)
+Requires-Dist: docling-ibm-models (>=3.4.0,<4.0.0)
 Requires-Dist: docling-parse (>=3.3.0,<4.0.0)
 Requires-Dist: easyocr (>=1.7,<2.0)
 Requires-Dist: filetype (>=1.2.0,<2.0.0)

{docling-2.23.0.dist-info → docling-2.24.0.dist-info}/RECORD

@@ -5,7 +5,7 @@ docling/backend/asciidoc_backend.py,sha256=zyHxlG_BvlLwvpdNca3P6aopxOJZw8wbDFkJQ
 docling/backend/csv_backend.py,sha256=xuId4JGEXjoyPgO9Fy9hQ5C-ezXvJwv0TGB8fyFHgWM,4533
 docling/backend/docling_parse_backend.py,sha256=hEEJibI1oJS0LAnFoIs6gMshS3bCqGtVxHnDNvBGZuA,7649
 docling/backend/docling_parse_v2_backend.py,sha256=IpwrBrtLGwNRl5AYO-o3NjEfNRsAkuMhzvDt2HXb9Ko,8655
-docling/backend/html_backend.py,sha256=YTPLZiEEEuGaP6G62skK3wXJ0KftuqBCl8erNXeJyoE,15893
+docling/backend/html_backend.py,sha256=BxYvYmgcio6IqROMFKgyYyoankcNUccalCeYlmTE4fk,16094
 docling/backend/json/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 docling/backend/json/docling_json_backend.py,sha256=LlFMVoZrrCfVwbDuRbNN4Xg96Lujh4xxrTBt9jGhY9I,1984
 docling/backend/md_backend.py,sha256=NaVfcnEH-5bwVovjn76EobF6B6Wm8AhaTZ4E8k0TUPo,16826
@@ -15,8 +15,8 @@ docling/backend/msword_backend.py,sha256=V4miLIcOH8DDlSCm25F_DALBW60Uf9JoSS0TB4y
 docling/backend/pdf_backend.py,sha256=17Pr8dWsD1C4FYUprrwMM9trDGW-JYLjrcScx1Ul4io,2048
 docling/backend/pypdfium2_backend.py,sha256=QSPfp903ZtSpoNqPmcIek0HmvETrJ1kkwrdxnF5pjS0,9014
 docling/backend/xml/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-docling/backend/xml/jats_backend.py,sha256=JI1iibmrob9Gv9y7zoFncavQ0oJaGWnQoLkozAIiTQU,27513
-docling/backend/xml/uspto_backend.py,sha256=a5GxWLj2SUR5Of8TWJinhef1gKyaQSjHPVXvGiN8yG8,70324
+docling/backend/xml/jats_backend.py,sha256=HXailrDjiwu4swwFnXy3lNfRtLZmkBBp4yqafCvdr7s,24945
+docling/backend/xml/uspto_backend.py,sha256=IGUNeF2xpLeaVrX6nKb-jXgtSYD2ozULsrDPcrI1IbQ,71040
 docling/chunking/__init__.py,sha256=h83TDs0AuOV6oEPLAPrn9dpGKiU-2Vg6IRNo4cv6GDA,346
 docling/cli/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 docling/cli/main.py,sha256=pCJ_GFgxsgZ0soz32OhMl-CWi7YXIrvax_m9Qw4UhMs,16839
@@ -34,23 +34,23 @@ docling/models/base_model.py,sha256=q_lKeQ0FT70idXlZ3JgyAv8dA8J3bZWBSDBkqTzy0lo,
 docling/models/base_ocr_model.py,sha256=YiUMvdjnHw9SHjnfJKT5INrPMoIGEf_Z2OApfl_VRTE,6919
 docling/models/code_formula_model.py,sha256=6grbRPWaLljadheT5s4omdT6hmXfin4gJU17csWvhjY,8611
 docling/models/document_picture_classifier.py,sha256=6I_j6fG5fnhIV6rqN31LYikNTZyg5isXrVs0GIqHDaY,6235
-docling/models/ds_glm_model.py,sha256=1jLEM-B_oHFevKq23zDQpdifE3eJL7qiLr5YLpEf1kQ,15217
 docling/models/easyocr_model.py,sha256=ePg1exAXeOzkBRBT-6PBSmqKFmnNFkCEd4HNDsGVgLM,6860
 docling/models/layout_model.py,sha256=7fQWipGV1HDrvbP4uOKa9QAicQl89jp7lailQmbFL3w,7804
 docling/models/ocr_mac_model.py,sha256=bLP14UUmZcSzjDe-HLj-mtksTuBmsCTg2C1wCxUpan0,4502
-docling/models/page_assemble_model.py,sha256=c5KLKwkUIdW0JcDHizWsqrpb5x_3DK28x82Q8o-3VJM,5968
+docling/models/page_assemble_model.py,sha256=ivkCdbZJpFcGl7CazLegcP1tLK8ZixDfVhQXqsdW_UA,6359
 docling/models/page_preprocessing_model.py,sha256=1gVrZjObKxAvXkkKvXnIFApPOggzgiTFPtt1CGbMbSs,2763
 docling/models/picture_description_api_model.py,sha256=SKNoHpqzbfM8iO-DJJ4ccyNVqO0B2d9neLBnXqt50FY,3186
 docling/models/picture_description_base_model.py,sha256=rZLIW1_CaRAw_EP3zuI8ktC0ZxwO7yubhh2RkaC_8e8,1910
 docling/models/picture_description_vlm_model.py,sha256=a2vYUdlcA0--_8neY0tTiU8reCf29NCbVMKwWdMy2QQ,3653
 docling/models/rapid_ocr_model.py,sha256=2HXmurNRPP6qyqn7U5h9NQIs8zi0TMHf56CpcKQk0fU,5038
+docling/models/readingorder_model.py,sha256=hNWbBX3uZv1FxMwKNKn2JFQuQqTspBLsJBVEidXr6Wk,14869
 docling/models/table_structure_model.py,sha256=UIqWlw_9JNfGsO86c00rPb4GCg-yNliKEwyhCqlsZbM,11225
-docling/models/tesseract_ocr_cli_model.py,sha256=b2Is5x2gZLS6mQWnKe0y7p6UU6hRTHDfoH4D2RQ5mx0,9310
+docling/models/tesseract_ocr_cli_model.py,sha256=F5EhS4NDEmLkPq-a0P7o2LrzjmJgACzlYXTDvtD3NtY,9343
 docling/models/tesseract_ocr_model.py,sha256=ikGu6QNknLG64c9yYIb0Ix6MGhBzOoa1ODbNc8MT5r8,8508
 docling/pipeline/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 docling/pipeline/base_pipeline.py,sha256=9ABK-Cr235bxE5vweoIA5rgBZV_EF8qFxAqLI27H_Pg,8749
 docling/pipeline/simple_pipeline.py,sha256=mZqANqUtAOFAyqQEShErQnAUz6tJFOl6zVoazEDJ_wE,2254
-docling/pipeline/standard_pdf_pipeline.py,sha256=Zoe8GGPujha16_TGYBAxcPriEwgYPaJPkp3BwG5XowU,12862
+docling/pipeline/standard_pdf_pipeline.py,sha256=IQHktVYvueTrYnIgLonaMvfYKKsU3L-hC9dqrR-Lw8g,12904
 docling/py.typed,sha256=AbpHGcgLb-kRsJGnwFEktk7uzpZOCcBY74-YBdrKVGs,1
 docling/utils/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 docling/utils/accelerator_utils.py,sha256=ONNRrC8fH-8E93WUCNhfOq1t7WrQ1T7-YsmExTOY5f0,2292
@@ -62,8 +62,8 @@ docling/utils/ocr_utils.py,sha256=F7iOOjqolUcImUzir4qjDQd4QWSO3s6JC4WRn3U7uY4,26
 docling/utils/profiling.py,sha256=YaMGoB9MMZpagF9mb5ndoHj8Lpb9aIdb7El-Pl7IcFs,1753
 docling/utils/utils.py,sha256=0ozCk7zUkYzxRVmYoIB2zA1lqjQOuaQzxfGuf1wmKW4,1866
 docling/utils/visualization.py,sha256=4pn-80fVuE04ken7hUg5Ar47ndRSL9MWBgdHM-1g1zU,2735
-docling-2.23.0.dist-info/LICENSE,sha256=mBb7ErEcM8VS9OhiGHnQ2kk75HwPhr54W1Oiz3965MY,1088
-docling-2.23.0.dist-info/METADATA,sha256=O4EJYC_yjLCFfKnhnzgSW4qGLOHaatDWDXsQS2EJDjU,8720
-docling-2.23.0.dist-info/WHEEL,sha256=Nq82e9rUAnEjt98J6MlVmMCZb-t9cYE2Ir1kpBmnWfs,88
-docling-2.23.0.dist-info/entry_points.txt,sha256=cFrINXsORijdm2EWJzf1m9_rDxH9G9W1fP385-9atY4,84
-docling-2.23.0.dist-info/RECORD,,
+docling-2.24.0.dist-info/LICENSE,sha256=mBb7ErEcM8VS9OhiGHnQ2kk75HwPhr54W1Oiz3965MY,1088
+docling-2.24.0.dist-info/METADATA,sha256=0MJ5mBt0GwsZotaSpHnAWzdzWcu_BQFGqGzNR3gRpG4,8672
+docling-2.24.0.dist-info/WHEEL,sha256=Nq82e9rUAnEjt98J6MlVmMCZb-t9cYE2Ir1kpBmnWfs,88
+docling-2.24.0.dist-info/entry_points.txt,sha256=cFrINXsORijdm2EWJzf1m9_rDxH9G9W1fP385-9atY4,84
+docling-2.24.0.dist-info/RECORD,,