docling 2.22.0__py3-none-any.whl → 2.23.0__py3-none-any.whl

docling/datamodel/base_models.py

@@ -34,7 +34,6 @@ class InputFormat(str, Enum):
     DOCX = "docx"
     PPTX = "pptx"
     HTML = "html"
-    XML_PUBMED = "xml_pubmed"
     IMAGE = "image"
     PDF = "pdf"
     ASCIIDOC = "asciidoc"
@@ -42,6 +41,7 @@ class InputFormat(str, Enum):
     CSV = "csv"
     XLSX = "xlsx"
     XML_USPTO = "xml_uspto"
+    XML_JATS = "xml_jats"
     JSON_DOCLING = "json_docling"
 
 
@@ -59,7 +59,7 @@ FormatToExtensions: Dict[InputFormat, List[str]] = {
     InputFormat.PDF: ["pdf"],
     InputFormat.MD: ["md"],
     InputFormat.HTML: ["html", "htm", "xhtml"],
-    InputFormat.XML_PUBMED: ["xml", "nxml"],
+    InputFormat.XML_JATS: ["xml", "nxml"],
     InputFormat.IMAGE: ["jpg", "jpeg", "png", "tif", "tiff", "bmp"],
     InputFormat.ASCIIDOC: ["adoc", "asciidoc", "asc"],
     InputFormat.CSV: ["csv"],
@@ -79,7 +79,7 @@ FormatToMimeType: Dict[InputFormat, List[str]] = {
         "application/vnd.openxmlformats-officedocument.presentationml.presentation",
     ],
     InputFormat.HTML: ["text/html", "application/xhtml+xml"],
-    InputFormat.XML_PUBMED: ["application/xml"],
+    InputFormat.XML_JATS: ["application/xml"],
     InputFormat.IMAGE: [
         "image/png",
         "image/jpeg",

docling/datamodel/document.py

@@ -333,11 +333,11 @@ class _DocumentConversionInput(BaseModel):
                 ):
                     input_format = InputFormat.XML_USPTO
 
-                if (
-                    InputFormat.XML_PUBMED in formats
-                    and "/NLM//DTD JATS" in xml_doctype
+                if InputFormat.XML_JATS in formats and (
+                    "JATS-journalpublishing" in xml_doctype
+                    or "JATS-archive" in xml_doctype
                 ):
-                    input_format = InputFormat.XML_PUBMED
+                    input_format = InputFormat.XML_JATS
 
         elif mime == "text/plain":
             if InputFormat.XML_USPTO in formats and content_str.startswith("PATN\r\n"):

docling/datamodel/pipeline_options.py

@@ -1,11 +1,26 @@
 import logging
 import os
+import re
+import warnings
 from enum import Enum
 from pathlib import Path
 from typing import Annotated, Any, Dict, List, Literal, Optional, Union
 
-from pydantic import AnyUrl, BaseModel, ConfigDict, Field, model_validator
-from pydantic_settings import BaseSettings, SettingsConfigDict
+from pydantic import (
+    AnyUrl,
+    BaseModel,
+    ConfigDict,
+    Field,
+    field_validator,
+    model_validator,
+    validator,
+)
+from pydantic_settings import (
+    BaseSettings,
+    PydanticBaseSettingsSource,
+    SettingsConfigDict,
+)
+from typing_extensions import deprecated
 
 _log = logging.getLogger(__name__)
 
@@ -25,7 +40,18 @@ class AcceleratorOptions(BaseSettings):
     )
 
     num_threads: int = 4
-    device: AcceleratorDevice = AcceleratorDevice.AUTO
+    device: Union[str, AcceleratorDevice] = "auto"
+
+    @field_validator("device")
+    def validate_device(cls, value):
+        # "auto", "cpu", "cuda", "mps", or "cuda:N"
+        if value in {d.value for d in AcceleratorDevice} or re.match(
+            r"^cuda(:\d+)?$", value
+        ):
+            return value
+        raise ValueError(
+            "Invalid device option. Use 'auto', 'cpu', 'mps', 'cuda', or 'cuda:N'."
+        )
 
     @model_validator(mode="before")
     @classmethod
@@ -41,7 +67,6 @@ class AcceleratorOptions(BaseSettings):
         """
         if isinstance(data, dict):
             input_num_threads = data.get("num_threads")
-
             # Check if to set the num_threads from the alternative envvar
             if input_num_threads is None:
                 docling_num_threads = os.getenv("DOCLING_NUM_THREADS")

docling/document_converter.py

@@ -18,7 +18,7 @@ from docling.backend.md_backend import MarkdownDocumentBackend
 from docling.backend.msexcel_backend import MsExcelDocumentBackend
 from docling.backend.mspowerpoint_backend import MsPowerpointDocumentBackend
 from docling.backend.msword_backend import MsWordDocumentBackend
-from docling.backend.xml.pubmed_backend import PubMedDocumentBackend
+from docling.backend.xml.jats_backend import JatsDocumentBackend
 from docling.backend.xml.uspto_backend import PatentUsptoDocumentBackend
 from docling.datamodel.base_models import (
     ConversionStatus,
@@ -102,9 +102,9 @@ class PatentUsptoFormatOption(FormatOption):
     backend: Type[PatentUsptoDocumentBackend] = PatentUsptoDocumentBackend
 
 
-class XMLPubMedFormatOption(FormatOption):
+class XMLJatsFormatOption(FormatOption):
     pipeline_cls: Type = SimplePipeline
-    backend: Type[AbstractDocumentBackend] = PubMedDocumentBackend
+    backend: Type[AbstractDocumentBackend] = JatsDocumentBackend
 
 
 class ImageFormatOption(FormatOption):
@@ -143,8 +143,8 @@ def _get_default_option(format: InputFormat) -> FormatOption:
         InputFormat.XML_USPTO: FormatOption(
             pipeline_cls=SimplePipeline, backend=PatentUsptoDocumentBackend
         ),
-        InputFormat.XML_PUBMED: FormatOption(
-            pipeline_cls=SimplePipeline, backend=PubMedDocumentBackend
+        InputFormat.XML_JATS: FormatOption(
+            pipeline_cls=SimplePipeline, backend=JatsDocumentBackend
         ),
         InputFormat.IMAGE: FormatOption(
             pipeline_cls=StandardPdfPipeline, backend=DoclingParseV2DocumentBackend

docling/utils/accelerator_utils.py

@@ -7,36 +7,62 @@ from docling.datamodel.pipeline_options import AcceleratorDevice
 _log = logging.getLogger(__name__)
 
 
-def decide_device(accelerator_device: AcceleratorDevice) -> str:
+def decide_device(accelerator_device: str) -> str:
     r"""
-    Resolve the device based on the acceleration options and the available devices in the system
+    Resolve the device based on the acceleration options and the available devices in the system.
+
     Rules:
     1. AUTO: Check for the best available device on the system.
     2. User-defined: Check if the device actually exists, otherwise fall-back to CPU
     """
-    cuda_index = 0
     device = "cpu"
 
     has_cuda = torch.backends.cuda.is_built() and torch.cuda.is_available()
     has_mps = torch.backends.mps.is_built() and torch.backends.mps.is_available()
 
-    if accelerator_device == AcceleratorDevice.AUTO:
+    if accelerator_device == AcceleratorDevice.AUTO.value:  # Handle 'auto'
         if has_cuda:
-            device = f"cuda:{cuda_index}"
+            device = "cuda:0"
         elif has_mps:
             device = "mps"
 
-    else:
-        if accelerator_device == AcceleratorDevice.CUDA:
-            if has_cuda:
-                device = f"cuda:{cuda_index}"
-            else:
-                _log.warning("CUDA is not available in the system. Fall back to 'CPU'")
-        elif accelerator_device == AcceleratorDevice.MPS:
-            if has_mps:
-                device = "mps"
+    elif accelerator_device.startswith("cuda"):
+        if has_cuda:
+            # if cuda device index specified extract device id
+            parts = accelerator_device.split(":")
+            if len(parts) == 2 and parts[1].isdigit():
+                # select cuda device's id
+                cuda_index = int(parts[1])
+                if cuda_index < torch.cuda.device_count():
+                    device = f"cuda:{cuda_index}"
+                else:
+                    _log.warning(
+                        "CUDA device 'cuda:%d' is not available. Fall back to 'CPU'.",
+                        cuda_index,
+                    )
+            elif len(parts) == 1:  # just "cuda"
+                device = "cuda:0"
             else:
-                _log.warning("MPS is not available in the system. Fall back to 'CPU'")
+                _log.warning(
+                    "Invalid CUDA device format '%s'. Fall back to 'CPU'",
+                    accelerator_device,
+                )
+        else:
+            _log.warning("CUDA is not available in the system. Fall back to 'CPU'")
+
+    elif accelerator_device == AcceleratorDevice.MPS.value:
+        if has_mps:
+            device = "mps"
+        else:
+            _log.warning("MPS is not available in the system. Fall back to 'CPU'")
+
+    elif accelerator_device == AcceleratorDevice.CPU.value:
+        device = "cpu"
+
+    else:
+        _log.warning(
+            "Unknown device option '%s'. Fall back to 'CPU'", accelerator_device
+        )
 
     _log.info("Accelerator device: '%s'", device)
     return device

{docling-2.22.0.dist-info → docling-2.23.0.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: docling
-Version: 2.22.0
+Version: 2.23.0
 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
@@ -28,7 +28,7 @@ Provides-Extra: vlm
 Requires-Dist: beautifulsoup4 (>=4.12.3,<4.13.0)
 Requires-Dist: certifi (>=2024.7.4)
 Requires-Dist: deepsearch-glm (>=1.0.0,<2.0.0)
-Requires-Dist: docling-core[chunking] (>=2.18.0,<3.0.0)
+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-parse (>=3.3.0,<4.0.0)
 Requires-Dist: easyocr (>=1.7,<2.0)

{docling-2.22.0.dist-info → docling-2.23.0.dist-info}/RECORD

@@ -15,7 +15,7 @@ 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/pubmed_backend.py,sha256=LMnpowjnxa5SydfNC00Ll840BYraL8dCJu-FfC9iSKk,20447
+docling/backend/xml/jats_backend.py,sha256=JI1iibmrob9Gv9y7zoFncavQ0oJaGWnQoLkozAIiTQU,27513
 docling/backend/xml/uspto_backend.py,sha256=a5GxWLj2SUR5Of8TWJinhef1gKyaQSjHPVXvGiN8yG8,70324
 docling/chunking/__init__.py,sha256=h83TDs0AuOV6oEPLAPrn9dpGKiU-2Vg6IRNo4cv6GDA,346
 docling/cli/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
@@ -23,11 +23,11 @@ docling/cli/main.py,sha256=pCJ_GFgxsgZ0soz32OhMl-CWi7YXIrvax_m9Qw4UhMs,16839
 docling/cli/models.py,sha256=Z4IEuaXE9el5PuI6_6mR4D5Sn3y8WZzBtoIJPi6jL_s,3188
 docling/cli/tools.py,sha256=QhtRxQG0TVrfsMqdv5i7J0_qQy1ZZyWYnHPwJl7b5oY,322
 docling/datamodel/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-docling/datamodel/base_models.py,sha256=_TPj-ADts3Qsc6vx1dpwZZnrOQCelqXOYIBCkK7A8FM,7107
-docling/datamodel/document.py,sha256=Aeqpm7d_CCV_2mwMhvNGVeGPWtWN9DJ5WAE4sjqN-dw,14530
-docling/datamodel/pipeline_options.py,sha256=pWCGtK0HEfltTR9Z14BYdS1-Zg6gZq9RlIHA014DpAk,9683
+docling/datamodel/base_models.py,sha256=b_8LiDCC4MkpqnKfsJjduH2DSsjADCllBLNB83Tpamw,7099
+docling/datamodel/document.py,sha256=DbJifyMgBEkAk80BMYXTuSgqH2vijDENDkU7Fmr6j_g,14567
+docling/datamodel/pipeline_options.py,sha256=5jXSVNGyOy6Ha18Wd80e7pYFmvRZk-2Lkgx0bwMOuq8,10234
 docling/datamodel/settings.py,sha256=bNMdowIKv7RUchabQTo4rFNEsxfB6pGg2LoZSY634zo,1869
-docling/document_converter.py,sha256=DX_bMqYyVO6rQvpf2JEy95HDR1QXT51v3T3Xn40pwjE,13196
+docling/document_converter.py,sha256=AeiSmKzWcnOkZm8O-KIBG72g3l4W2CAsq3yEbfC1tiE,13184
 docling/exceptions.py,sha256=K1WnCS1leK2JtMB5ewZWKkb0EaijFgl-tRzrO9ntgPM,134
 docling/models/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 docling/models/base_model.py,sha256=q_lKeQ0FT70idXlZ3JgyAv8dA8J3bZWBSDBkqTzy0lo,2679
@@ -53,7 +53,7 @@ docling/pipeline/simple_pipeline.py,sha256=mZqANqUtAOFAyqQEShErQnAUz6tJFOl6zVoaz
 docling/pipeline/standard_pdf_pipeline.py,sha256=Zoe8GGPujha16_TGYBAxcPriEwgYPaJPkp3BwG5XowU,12862
 docling/py.typed,sha256=AbpHGcgLb-kRsJGnwFEktk7uzpZOCcBY74-YBdrKVGs,1
 docling/utils/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-docling/utils/accelerator_utils.py,sha256=ZjULCn-qhxqx3frF-rJmAlWdzqgUMxH5utLHbSPev80,1367
+docling/utils/accelerator_utils.py,sha256=ONNRrC8fH-8E93WUCNhfOq1t7WrQ1T7-YsmExTOY5f0,2292
 docling/utils/export.py,sha256=KyGF1BVDHPFfHVXZc8vegsWlFfOgGPP2YckWpTadyI8,4694
 docling/utils/glm_utils.py,sha256=W4JRoP0xQ6SJmhhIoAfcKxm5dr1CFvLHp8pqI1kdhxs,12250
 docling/utils/layout_postprocessor.py,sha256=urRzeF9PrKiMBvA6DdHHwyLxG06CMhelgJeV5B1l6l0,24258
@@ -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-2.22.0.dist-info/LICENSE,sha256=mBb7ErEcM8VS9OhiGHnQ2kk75HwPhr54W1Oiz3965MY,1088
-docling-2.22.0.dist-info/METADATA,sha256=eKFbLHbqOA9xMt4c0Pdqwh7tVBOXSqdSWh_MP4ztkeU,8720
-docling-2.22.0.dist-info/WHEEL,sha256=Nq82e9rUAnEjt98J6MlVmMCZb-t9cYE2Ir1kpBmnWfs,88
-docling-2.22.0.dist-info/entry_points.txt,sha256=cFrINXsORijdm2EWJzf1m9_rDxH9G9W1fP385-9atY4,84
-docling-2.22.0.dist-info/RECORD,,
+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,,