docling 2.47.0__tar.gz → 2.48.0__tar.gz

{docling-2.47.0 → docling-2.48.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: docling
-Version: 2.47.0
+Version: 2.48.0
 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.
 Author-email: Christoph Auer <cau@zurich.ibm.com>, Michele Dolfi <dol@zurich.ibm.com>, Maxim Lysak <mly@zurich.ibm.com>, Nikos Livathinos <nli@zurich.ibm.com>, Ahmed Nassar <ahn@zurich.ibm.com>, Panos Vagenas <pva@zurich.ibm.com>, Peter Staar <taa@zurich.ibm.com>
 License-Expression: MIT
@@ -59,10 +59,11 @@ Provides-Extra: vlm
 Requires-Dist: transformers<5.0.0,>=4.46.0; extra == "vlm"
 Requires-Dist: accelerate<2.0.0,>=1.2.1; extra == "vlm"
 Requires-Dist: mlx-vlm<1.0.0,>=0.3.0; (python_version >= "3.10" and sys_platform == "darwin" and platform_machine == "arm64") and extra == "vlm"
-Requires-Dist: vllm<1.0.0,>=0.10.0; (python_version >= "3.10" and sys_platform == "linux") and extra == "vlm"
+Requires-Dist: vllm<1.0.0,>=0.10.0; (python_version >= "3.10" and sys_platform == "linux" and platform_machine == "x86_64") and extra == "vlm"
 Provides-Extra: rapidocr
-Requires-Dist: rapidocr-onnxruntime<2.0.0,>=1.4.0; python_version < "3.13" and extra == "rapidocr"
+Requires-Dist: rapidocr<4.0.0,>=3.3; python_version < "3.14" and extra == "rapidocr"
 Requires-Dist: onnxruntime<2.0.0,>=1.7.0; extra == "rapidocr"
+Requires-Dist: modelscope>=1.29.0; extra == "rapidocr"
 Provides-Extra: asr
 Requires-Dist: openai-whisper>=20250625; extra == "asr"
 Dynamic: license-file

{docling-2.47.0 → docling-2.48.0}/docling/backend/html_backend.py

@@ -1,5 +1,6 @@
 import logging
 import re
+import traceback
 from contextlib import contextmanager
 from copy import deepcopy
 from io import BytesIO
@@ -45,20 +46,22 @@ _BLOCK_TAGS: Final = {
     "h4",
     "h5",
     "h6",
+    "ol",
     "p",
     "pre",
-    "code",
-    "ul",
-    "ol",
     "summary",
     "table",
+    "ul",
 }
 
+_CODE_TAG_SET: Final = {"code", "kbd", "samp"}
+
 _FORMAT_TAG_MAP: Final = {
     "b": {"bold": True},
     "strong": {"bold": True},
     "i": {"italic": True},
     "em": {"italic": True},
+    "var": {"italic": True},
     # "mark",
     # "small",
     "s": {"strikethrough": True},
@@ -67,6 +70,7 @@ _FORMAT_TAG_MAP: Final = {
     "ins": {"underline": True},
     "sub": {"script": Script.SUB},
     "sup": {"script": Script.SUPER},
+    **{k: {} for k in _CODE_TAG_SET},
 }
 
 
@@ -79,6 +83,7 @@ class AnnotatedText(BaseModel):
     text: str
     hyperlink: Union[AnyUrl, Path, None] = None
     formatting: Union[Formatting, None] = None
+    code: bool = False
 
 
 class AnnotatedTextList(list):
@@ -86,10 +91,12 @@ class AnnotatedTextList(list):
         current_h = None
         current_text = ""
         current_f = None
+        current_code = False
         for at in self:
             t = at.text
             h = at.hyperlink
             f = at.formatting
+            c = at.code
             current_text += t.strip() + " "
             if f is not None and current_f is None:
                 current_f = f
@@ -103,8 +110,13 @@ class AnnotatedTextList(list):
                 _log.warning(
                     f"Clashing hyperlinks: '{h}' and '{current_h}'! Chose '{current_h}'"
                 )
+            current_code = c if c else current_code
+
         return AnnotatedText(
-            text=current_text.strip(), hyperlink=current_h, formatting=current_f
+            text=current_text.strip(),
+            hyperlink=current_h,
+            formatting=current_f,
+            code=current_code,
         )
 
     def simplify_text_elements(self) -> "AnnotatedTextList":
@@ -114,9 +126,14 @@ class AnnotatedTextList(list):
         text = self[0].text
         hyperlink = self[0].hyperlink
         formatting = self[0].formatting
+        code = self[0].code
         last_elm = text
         for i in range(1, len(self)):
-            if hyperlink == self[i].hyperlink and formatting == self[i].formatting:
+            if (
+                hyperlink == self[i].hyperlink
+                and formatting == self[i].formatting
+                and code == self[i].code
+            ):
                 sep = " "
                 if not self[i].text.strip() or not last_elm.strip():
                     sep = ""
@@ -124,15 +141,20 @@ class AnnotatedTextList(list):
                 last_elm = self[i].text
             else:
                 simplified.append(
-                    AnnotatedText(text=text, hyperlink=hyperlink, formatting=formatting)
+                    AnnotatedText(
+                        text=text, hyperlink=hyperlink, formatting=formatting, code=code
+                    )
                 )
                 text = self[i].text
                 last_elm = text
                 hyperlink = self[i].hyperlink
                 formatting = self[i].formatting
+                code = self[i].code
         if text:
             simplified.append(
-                AnnotatedText(text=text, hyperlink=hyperlink, formatting=formatting)
+                AnnotatedText(
+                    text=text, hyperlink=hyperlink, formatting=formatting, code=code
+                )
             )
         return simplified
 
@@ -174,7 +196,7 @@ class HTMLDocumentBackend(DeclarativeDocumentBackend):
         self.ctx = _Context()
         for i in range(self.max_levels):
             self.parents[i] = None
-        self.hyperlink = None
+        self.hyperlink: Union[AnyUrl, Path, None] = None
         self.original_url = original_url
         self.format_tags: list[str] = []
 
@@ -235,9 +257,13 @@ class HTMLDocumentBackend(DeclarativeDocumentBackend):
                 orig=title_text,
                 content_layer=ContentLayer.FURNITURE,
             )
-        # remove scripts/styles
+        # remove script and style tags
         for tag in self.soup(["script", "style"]):
             tag.decompose()
+        # remove any hidden tag
+        for tag in self.soup(hidden=True):
+            tag.decompose()
+
         content = self.soup.body or self.soup
         # normalize <br> tags
         for br in content("br"):
@@ -268,7 +294,7 @@ class HTMLDocumentBackend(DeclarativeDocumentBackend):
         def flush_buffer():
             if not buffer:
                 return
-            annotated_text_list = buffer.simplify_text_elements()
+            annotated_text_list: AnnotatedTextList = buffer.simplify_text_elements()
             parts = annotated_text_list.split_by_newline()
             buffer.clear()
 
@@ -276,20 +302,29 @@ class HTMLDocumentBackend(DeclarativeDocumentBackend):
                 return
 
             for annotated_text_list in parts:
-                with self.use_inline_group(annotated_text_list, doc):
+                with self._use_inline_group(annotated_text_list, doc):
                     for annotated_text in annotated_text_list:
                         if annotated_text.text.strip():
                             seg_clean = HTMLDocumentBackend._clean_unicode(
                                 annotated_text.text.strip()
                             )
-                            doc.add_text(
-                                parent=self.parents[self.level],
-                                label=DocItemLabel.TEXT,
-                                text=seg_clean,
-                                content_layer=self.content_layer,
-                                formatting=annotated_text.formatting,
-                                hyperlink=annotated_text.hyperlink,
-                            )
+                            if annotated_text.code:
+                                doc.add_code(
+                                    parent=self.parents[self.level],
+                                    text=seg_clean,
+                                    content_layer=self.content_layer,
+                                    formatting=annotated_text.formatting,
+                                    hyperlink=annotated_text.hyperlink,
+                                )
+                            else:
+                                doc.add_text(
+                                    parent=self.parents[self.level],
+                                    label=DocItemLabel.TEXT,
+                                    text=seg_clean,
+                                    content_layer=self.content_layer,
+                                    formatting=annotated_text.formatting,
+                                    hyperlink=annotated_text.hyperlink,
+                                )
 
         for node in element.contents:
             if isinstance(node, Tag):
@@ -298,10 +333,10 @@ class HTMLDocumentBackend(DeclarativeDocumentBackend):
                     flush_buffer()
                     self._emit_image(node, doc)
                 elif name in _FORMAT_TAG_MAP:
-                    with self.use_format([name]):
+                    with self._use_format([name]):
                         self._walk(node, doc)
                 elif name == "a":
-                    with self.use_hyperlink(node):
+                    with self._use_hyperlink(node):
                         self._walk(node, doc)
                 elif name in _BLOCK_TAGS:
                     flush_buffer()
@@ -367,8 +402,8 @@ class HTMLDocumentBackend(DeclarativeDocumentBackend):
             this_parent = item.parent
             while this_parent is not None:
                 if this_parent.name == "a" and this_parent.get("href"):
-                    with self.use_format(format_tags):
-                        with self.use_hyperlink(this_parent):
+                    with self._use_format(format_tags):
+                        with self._use_hyperlink(this_parent):
                             return self._extract_text_and_hyperlink_recursively(
                                 item, ignore_list
                             )
@@ -379,6 +414,7 @@ class HTMLDocumentBackend(DeclarativeDocumentBackend):
 
         if isinstance(item, NavigableString):
             text = item.strip()
+            code = any(code_tag in self.format_tags for code_tag in _CODE_TAG_SET)
             if text:
                 return AnnotatedTextList(
                     [
@@ -386,6 +422,7 @@ class HTMLDocumentBackend(DeclarativeDocumentBackend):
                             text=text,
                             hyperlink=self.hyperlink,
                             formatting=self._formatting,
+                            code=code,
                         )
                     ]
                 )
@@ -396,6 +433,7 @@ class HTMLDocumentBackend(DeclarativeDocumentBackend):
                             text="\n",
                             hyperlink=self.hyperlink,
                             formatting=self._formatting,
+                            code=code,
                         )
                     ]
                 )
@@ -405,14 +443,14 @@ class HTMLDocumentBackend(DeclarativeDocumentBackend):
         if not ignore_list or (tag.name not in ["ul", "ol"]):
             for child in tag:
                 if isinstance(child, Tag) and child.name in _FORMAT_TAG_MAP:
-                    with self.use_format([child.name]):
+                    with self._use_format([child.name]):
                         result.extend(
                             self._extract_text_and_hyperlink_recursively(
                                 child, ignore_list, keep_newlines=keep_newlines
                             )
                         )
                 elif isinstance(child, Tag) and child.name == "a":
-                    with self.use_hyperlink(child):
+                    with self._use_hyperlink(child):
                         result.extend(
                             self._extract_text_and_hyperlink_recursively(
                                 child, ignore_list, keep_newlines=keep_newlines
@@ -428,29 +466,30 @@ class HTMLDocumentBackend(DeclarativeDocumentBackend):
         return result
 
     @contextmanager
-    def use_hyperlink(self, tag):
+    def _use_hyperlink(self, tag: Tag):
         this_href = tag.get("href")
         if this_href is None:
             yield None
         else:
-            if this_href:
-                old_hyperlink = self.hyperlink
+            if isinstance(this_href, str) and this_href:
+                old_hyperlink: Union[AnyUrl, Path, None] = self.hyperlink
+                new_hyperlink: Union[AnyUrl, Path, None] = None
                 if self.original_url is not None:
-                    this_href = urljoin(self.original_url, this_href)
+                    this_href = urljoin(str(self.original_url), str(this_href))
                 # ugly fix for relative links since pydantic does not support them.
                 try:
-                    AnyUrl(this_href)
+                    new_hyperlink = AnyUrl(this_href)
                 except ValidationError:
-                    this_href = Path(this_href)
-                self.hyperlink = this_href
+                    new_hyperlink = Path(this_href)
+                self.hyperlink = new_hyperlink
             try:
                 yield None
             finally:
-                if this_href:
+                if new_hyperlink:
                     self.hyperlink = old_hyperlink
 
     @contextmanager
-    def use_format(self, tags: list[str]):
+    def _use_format(self, tags: list[str]):
         if not tags:
             yield None
         else:
@@ -461,7 +500,7 @@ class HTMLDocumentBackend(DeclarativeDocumentBackend):
                 self.format_tags = self.format_tags[: -len(tags)]
 
     @contextmanager
-    def use_inline_group(
+    def _use_inline_group(
         self, annotated_text_list: AnnotatedTextList, doc: DoclingDocument
     ):
         """Create an inline group for annotated texts.
@@ -473,9 +512,6 @@ class HTMLDocumentBackend(DeclarativeDocumentBackend):
         Args:
             annotated_text_list (AnnotatedTextList): Annotated text
             doc (DoclingDocument): Currently used document
-
-        Yields:
-            None: _description_
         """
         if len(annotated_text_list) > 1:
             inline_fmt = doc.add_group(
@@ -493,6 +529,57 @@ class HTMLDocumentBackend(DeclarativeDocumentBackend):
         else:
             yield None
 
+    @contextmanager
+    def _use_details(self, tag: Tag, doc: DoclingDocument):
+        """Create a group with the content of a details tag.
+
+        While the context manager is active, the hierarchy level is set one
+        level higher as the cuurent parent.
+
+        Args:
+            tag: The details tag.
+            doc: Currently used document.
+        """
+        self.parents[self.level + 1] = doc.add_group(
+            name=tag.name,
+            label=GroupLabel.SECTION,
+            parent=self.parents[self.level],
+            content_layer=self.content_layer,
+        )
+        self.level += 1
+        try:
+            yield None
+        finally:
+            self.parents[self.level + 1] = None
+            self.level -= 1
+
+    @contextmanager
+    def _use_footer(self, tag: Tag, doc: DoclingDocument):
+        """Create a group with a footer.
+
+        Create a group with the content of a footer tag. While the context manager
+        is active, the hierarchy level is set one level higher as the cuurent parent.
+
+        Args:
+            tag: The footer tag.
+            doc: Currently used document.
+        """
+        current_layer = self.content_layer
+        self.content_layer = ContentLayer.FURNITURE
+        self.parents[self.level + 1] = doc.add_group(
+            name=tag.name,
+            label=GroupLabel.SECTION,
+            parent=self.parents[self.level],
+            content_layer=self.content_layer,
+        )
+        self.level += 1
+        try:
+            yield None
+        finally:
+            self.parents[self.level + 1] = None
+            self.level -= 1
+            self.content_layer = current_layer
+
     def _handle_heading(self, tag: Tag, doc: DoclingDocument) -> None:
         tag_name = tag.name.lower()
         # set default content layer to BODY as soon as we encounter a heading
@@ -611,20 +698,29 @@ class HTMLDocumentBackend(DeclarativeDocumentBackend):
                             content_layer=self.content_layer,
                         )
                         self.level += 1
-                        with self.use_inline_group(min_parts, doc):
+                        with self._use_inline_group(min_parts, doc):
                             for annotated_text in min_parts:
                                 li_text = re.sub(
                                     r"\s+|\n+", " ", annotated_text.text
                                 ).strip()
                                 li_clean = HTMLDocumentBackend._clean_unicode(li_text)
-                                doc.add_text(
-                                    parent=self.parents[self.level],
-                                    label=DocItemLabel.TEXT,
-                                    text=li_clean,
-                                    content_layer=self.content_layer,
-                                    formatting=annotated_text.formatting,
-                                    hyperlink=annotated_text.hyperlink,
-                                )
+                                if annotated_text.code:
+                                    doc.add_code(
+                                        parent=self.parents[self.level],
+                                        text=li_clean,
+                                        content_layer=self.content_layer,
+                                        formatting=annotated_text.formatting,
+                                        hyperlink=annotated_text.hyperlink,
+                                    )
+                                else:
+                                    doc.add_text(
+                                        parent=self.parents[self.level],
+                                        label=DocItemLabel.TEXT,
+                                        text=li_clean,
+                                        content_layer=self.content_layer,
+                                        formatting=annotated_text.formatting,
+                                        hyperlink=annotated_text.hyperlink,
+                                    )
 
                         # 4) recurse into any nested lists, attaching them to this <li> item
                         for sublist in li({"ul", "ol"}, recursive=False):
@@ -687,20 +783,29 @@ class HTMLDocumentBackend(DeclarativeDocumentBackend):
             text_list = self._extract_text_and_hyperlink_recursively(
                 tag, find_parent_annotation=True
             )
-            annotated_texts = text_list.simplify_text_elements()
+            annotated_texts: AnnotatedTextList = text_list.simplify_text_elements()
             for part in annotated_texts.split_by_newline():
-                with self.use_inline_group(part, doc):
+                with self._use_inline_group(part, doc):
                     for annotated_text in part:
                         if seg := annotated_text.text.strip():
                             seg_clean = HTMLDocumentBackend._clean_unicode(seg)
-                            doc.add_text(
-                                parent=self.parents[self.level],
-                                label=DocItemLabel.TEXT,
-                                text=seg_clean,
-                                content_layer=self.content_layer,
-                                formatting=annotated_text.formatting,
-                                hyperlink=annotated_text.hyperlink,
-                            )
+                            if annotated_text.code:
+                                doc.add_code(
+                                    parent=self.parents[self.level],
+                                    text=seg_clean,
+                                    content_layer=self.content_layer,
+                                    formatting=annotated_text.formatting,
+                                    hyperlink=annotated_text.hyperlink,
+                                )
+                            else:
+                                doc.add_text(
+                                    parent=self.parents[self.level],
+                                    label=DocItemLabel.TEXT,
+                                    text=seg_clean,
+                                    content_layer=self.content_layer,
+                                    formatting=annotated_text.formatting,
+                                    hyperlink=annotated_text.hyperlink,
+                                )
 
             for img_tag in tag("img"):
                 if isinstance(img_tag, Tag):
@@ -718,13 +823,13 @@ class HTMLDocumentBackend(DeclarativeDocumentBackend):
                     content_layer=self.content_layer,
                 )
 
-        elif tag_name in {"pre", "code"}:
+        elif tag_name in {"pre"}:
             # handle monospace code snippets (pre).
             text_list = self._extract_text_and_hyperlink_recursively(
-                tag, find_parent_annotation=True
+                tag, find_parent_annotation=True, keep_newlines=True
             )
             annotated_texts = text_list.simplify_text_elements()
-            with self.use_inline_group(annotated_texts, doc):
+            with self._use_inline_group(annotated_texts, doc):
                 for annotated_text in annotated_texts:
                     text_clean = HTMLDocumentBackend._clean_unicode(
                         annotated_text.text.strip()
@@ -737,22 +842,13 @@ class HTMLDocumentBackend(DeclarativeDocumentBackend):
                         hyperlink=annotated_text.hyperlink,
                     )
 
-        elif tag_name in {"details", "footer"}:
-            if tag_name == "footer":
-                current_layer = self.content_layer
-                self.content_layer = ContentLayer.FURNITURE
-            self.parents[self.level + 1] = doc.add_group(
-                name=tag_name,
-                label=GroupLabel.SECTION,
-                parent=self.parents[self.level],
-                content_layer=self.content_layer,
-            )
-            self.level += 1
-            self._walk(tag, doc)
-            self.parents[self.level + 1] = None
-            self.level -= 1
-            if tag_name == "footer":
-                self.content_layer = current_layer
+        elif tag_name == "footer":
+            with self._use_footer(tag, doc):
+                self._walk(tag, doc)
+
+        elif tag_name == "details":
+            with self._use_details(tag, doc):
+                self._walk(tag, doc)
 
     def _emit_image(self, img_tag: Tag, doc: DoclingDocument) -> None:
         figure = img_tag.find_parent("figure")

{docling-2.47.0 → docling-2.48.0}/docling/datamodel/pipeline_options.py

@@ -99,6 +99,8 @@ class RapidOcrOptions(OcrOptions):
     # For more details on the following options visit
     # https://rapidai.github.io/RapidOCRDocs/install_usage/api/RapidOCR/
 
+    # https://rapidai.github.io/RapidOCRDocs/main/install_usage/rapidocr/usage/#__tabbed_3_4
+    backend: Literal["onnxruntime", "openvino", "paddle", "torch"] = "onnxruntime"
     text_score: float = 0.5  # same default as rapidocr
 
     use_det: Optional[bool] = None  # same default as rapidocr

{docling-2.47.0 → docling-2.48.0}/docling/models/rapid_ocr_model.py

@@ -42,10 +42,10 @@ class RapidOcrModel(BaseOcrModel):
 
         if self.enabled:
             try:
-                from rapidocr_onnxruntime import RapidOCR  # type: ignore
+                from rapidocr import EngineType, RapidOCR  # type: ignore
             except ImportError:
                 raise ImportError(
-                    "RapidOCR is not installed. Please install it via `pip install rapidocr_onnxruntime` to use this OCR engine. "
+                    "RapidOCR is not installed. Please install it via `pip install rapidocr onnxruntime` to use this OCR engine. "
                     "Alternatively, Docling has support for other OCR engines. See the documentation."
                 )
 
@@ -54,21 +54,39 @@ class RapidOcrModel(BaseOcrModel):
             use_cuda = str(AcceleratorDevice.CUDA.value).lower() in device
             use_dml = accelerator_options.device == AcceleratorDevice.AUTO
             intra_op_num_threads = accelerator_options.num_threads
+            _ALIASES = {
+                "onnxruntime": EngineType.ONNXRUNTIME,
+                "openvino": EngineType.OPENVINO,
+                "paddle": EngineType.PADDLE,
+                "torch": EngineType.TORCH,
+            }
+            backend_enum = _ALIASES.get(self.options.backend, EngineType.ONNXRUNTIME)
 
             self.reader = RapidOCR(
-                text_score=self.options.text_score,
-                cls_use_cuda=use_cuda,
-                rec_use_cuda=use_cuda,
-                det_use_cuda=use_cuda,
-                det_use_dml=use_dml,
-                cls_use_dml=use_dml,
-                rec_use_dml=use_dml,
-                intra_op_num_threads=intra_op_num_threads,
-                print_verbose=self.options.print_verbose,
-                det_model_path=self.options.det_model_path,
-                cls_model_path=self.options.cls_model_path,
-                rec_model_path=self.options.rec_model_path,
-                rec_keys_path=self.options.rec_keys_path,
+                params={
+                    # Global settings (these are still correct)
+                    "Global.text_score": self.options.text_score,
+                    # "Global.verbose": self.options.print_verbose,
+                    # Detection model settings
+                    "Det.model_path": self.options.det_model_path,
+                    "Det.use_cuda": use_cuda,
+                    "Det.use_dml": use_dml,
+                    "Det.intra_op_num_threads": intra_op_num_threads,
+                    # Classification model settings
+                    "Cls.model_path": self.options.cls_model_path,
+                    "Cls.use_cuda": use_cuda,
+                    "Cls.use_dml": use_dml,
+                    "Cls.intra_op_num_threads": intra_op_num_threads,
+                    # Recognition model settings
+                    "Rec.model_path": self.options.rec_model_path,
+                    "Rec.keys_path": self.options.rec_keys_path,
+                    "Rec.use_cuda": use_cuda,
+                    "Rec.use_dml": use_dml,
+                    "Rec.intra_op_num_threads": intra_op_num_threads,
+                    "Det.engine_type": backend_enum,
+                    "Cls.engine_type": backend_enum,
+                    "Rec.engine_type": backend_enum,
+                }
             )
 
     def __call__(
@@ -95,12 +113,15 @@ class RapidOcrModel(BaseOcrModel):
                             scale=self.scale, cropbox=ocr_rect
                         )
                         im = numpy.array(high_res_image)
-                        result, _ = self.reader(
+                        result = self.reader(
                             im,
                             use_det=self.options.use_det,
                             use_cls=self.options.use_cls,
                             use_rec=self.options.use_rec,
                         )
+                        result = list(
+                            zip(result.boxes.tolist(), result.txts, result.scores)
+                        )
 
                         del high_res_image
                         del im

{docling-2.47.0 → docling-2.48.0}/docling/pipeline/base_pipeline.py

@@ -146,6 +146,7 @@ class PaginatedPipeline(BasePipeline):  # TODO this is a bad name.
                     conv_res.pages.append(Page(page_no=i))
 
             try:
+                total_pages_processed = 0
                 # Iterate batches of pages (page_batch_size) in the doc
                 for page_batch in chunkify(
                     conv_res.pages, settings.perf.page_batch_size
@@ -186,9 +187,9 @@ class PaginatedPipeline(BasePipeline):  # TODO this is a bad name.
                         )
                         conv_res.status = ConversionStatus.PARTIAL_SUCCESS
                         break
-
+                    total_pages_processed += len(page_batch)
                     _log.debug(
-                        f"Finished converting page batch time={end_batch_time:.3f}"
+                        f"Finished converting pages {total_pages_processed}/{len(conv_res.pages)} time={end_batch_time:.3f}"
                     )
 
             except Exception as e:

{docling-2.47.0 → docling-2.48.0}/docling.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: docling
-Version: 2.47.0
+Version: 2.48.0
 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.
 Author-email: Christoph Auer <cau@zurich.ibm.com>, Michele Dolfi <dol@zurich.ibm.com>, Maxim Lysak <mly@zurich.ibm.com>, Nikos Livathinos <nli@zurich.ibm.com>, Ahmed Nassar <ahn@zurich.ibm.com>, Panos Vagenas <pva@zurich.ibm.com>, Peter Staar <taa@zurich.ibm.com>
 License-Expression: MIT
@@ -59,10 +59,11 @@ Provides-Extra: vlm
 Requires-Dist: transformers<5.0.0,>=4.46.0; extra == "vlm"
 Requires-Dist: accelerate<2.0.0,>=1.2.1; extra == "vlm"
 Requires-Dist: mlx-vlm<1.0.0,>=0.3.0; (python_version >= "3.10" and sys_platform == "darwin" and platform_machine == "arm64") and extra == "vlm"
-Requires-Dist: vllm<1.0.0,>=0.10.0; (python_version >= "3.10" and sys_platform == "linux") and extra == "vlm"
+Requires-Dist: vllm<1.0.0,>=0.10.0; (python_version >= "3.10" and sys_platform == "linux" and platform_machine == "x86_64") and extra == "vlm"
 Provides-Extra: rapidocr
-Requires-Dist: rapidocr-onnxruntime<2.0.0,>=1.4.0; python_version < "3.13" and extra == "rapidocr"
+Requires-Dist: rapidocr<4.0.0,>=3.3; python_version < "3.14" and extra == "rapidocr"
 Requires-Dist: onnxruntime<2.0.0,>=1.7.0; extra == "rapidocr"
+Requires-Dist: modelscope>=1.29.0; extra == "rapidocr"
 Provides-Extra: asr
 Requires-Dist: openai-whisper>=20250625; extra == "asr"
 Dynamic: license-file

{docling-2.47.0 → docling-2.48.0}/docling.egg-info/requires.txt

@@ -35,9 +35,10 @@ ocrmac<2.0.0,>=1.0.0
 
 [rapidocr]
 onnxruntime<2.0.0,>=1.7.0
+modelscope>=1.29.0
 
-[rapidocr:python_version < "3.13"]
-rapidocr-onnxruntime<2.0.0,>=1.4.0
+[rapidocr:python_version < "3.14"]
+rapidocr<4.0.0,>=3.3
 
 [tesserocr]
 tesserocr<3.0.0,>=2.7.1
@@ -49,5 +50,5 @@ accelerate<2.0.0,>=1.2.1
 [vlm:python_version >= "3.10" and sys_platform == "darwin" and platform_machine == "arm64"]
 mlx-vlm<1.0.0,>=0.3.0
 
-[vlm:python_version >= "3.10" and sys_platform == "linux"]
+[vlm:python_version >= "3.10" and sys_platform == "linux" and platform_machine == "x86_64"]
 vllm<1.0.0,>=0.10.0

{docling-2.47.0 → docling-2.48.0}/pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "docling"
-version = "2.47.0"  # DO NOT EDIT, updated automatically
+version = "2.48.0"  # DO NOT EDIT, updated automatically
 description = "SDK and CLI for parsing PDF, DOCX, HTML, and more, to a unified document representation for powering downstream workflows such as gen AI applications."
 license = "MIT"
 keywords = [
@@ -93,11 +93,12 @@ vlm = [
   'transformers (>=4.46.0,<5.0.0)',
   'accelerate (>=1.2.1,<2.0.0)',
   'mlx-vlm (>=0.3.0,<1.0.0) ; python_version >= "3.10" and sys_platform == "darwin" and platform_machine == "arm64"',
-  'vllm (>=0.10.0,<1.0.0) ; python_version >= "3.10" and sys_platform == "linux"',
+  'vllm (>=0.10.0,<1.0.0) ; python_version >= "3.10" and sys_platform == "linux" and platform_machine == "x86_64"',
 ]
 rapidocr = [
-  'rapidocr-onnxruntime (>=1.4.0,<2.0.0) ; python_version < "3.13"',
+  'rapidocr (>=3.3,<4.0.0) ; python_version < "3.14"',
   'onnxruntime (>=1.7.0,<2.0.0)',
+    "modelscope>=1.29.0",
   # 'onnxruntime (>=1.7.0,<2.0.0) ; python_version >= "3.10"',
   # 'onnxruntime (>=1.7.0,<1.20.0) ; python_version < "3.10"',
 ]

{docling-2.47.0 → docling-2.48.0}/tests/test_backend_webp.py

@@ -55,8 +55,8 @@ def test_e2e_webp_conversions():
         TesseractCliOcrOptions(force_full_page_ocr=True, lang=["auto"]),
     ]
 
-    # rapidocr is only available for Python >=3.6,<3.13
-    if sys.version_info < (3, 13):
+    # rapidocr is only available for Python >=3.6,<3.14
+    if sys.version_info < (3, 14):
         engines.append(RapidOcrOptions())
         engines.append(RapidOcrOptions(force_full_page_ocr=True))