docling 1.6.3__py3-none-any.whl → 1.7.1__py3-none-any.whl

docling/backend/abstract_backend.py

@@ -39,8 +39,9 @@ class PdfPageBackend(ABC):
 
 class PdfDocumentBackend(ABC):
     @abstractmethod
-    def __init__(self, path_or_stream: Union[BytesIO, Path]):
-        pass
+    def __init__(self, path_or_stream: Union[BytesIO, Path], document_hash: str):
+        self.path_or_stream = path_or_stream
+        self.document_hash = document_hash
 
     @abstractmethod
     def load_page(self, page_no: int) -> PdfPageBackend:
@@ -56,4 +57,7 @@ class PdfDocumentBackend(ABC):
 
     @abstractmethod
     def unload(self):
-        pass
+        if isinstance(self.path_or_stream, BytesIO):
+            self.path_or_stream.close()
+
+        self.path_or_stream = None

docling/backend/docling_parse_backend.py

@@ -1,6 +1,5 @@
 import logging
 import random
-import time
 from io import BytesIO
 from pathlib import Path
 from typing import Iterable, Optional, Union
@@ -17,13 +16,26 @@ _log = logging.getLogger(__name__)
 
 
 class DoclingParsePageBackend(PdfPageBackend):
-    def __init__(self, page_obj: PdfPage, docling_page_obj):
+    def __init__(
+        self, parser: pdf_parser, document_hash: str, page_no: int, page_obj: PdfPage
+    ):
         super().__init__(page_obj)
         self._ppage = page_obj
-        self._dpage = docling_page_obj
-        self.text_page = None
+
+        parsed_page = parser.parse_pdf_from_key_on_page(document_hash, page_no)
+
+        self._dpage = None
+        self.broken_page = "pages" not in parsed_page
+        if not self.broken_page:
+            self._dpage = parsed_page["pages"][0]
+        else:
+            raise RuntimeError(
+                f"Page {page_no} of document {document_hash} could not be parsed."
+            )
 
     def get_text_in_rect(self, bbox: BoundingBox) -> str:
+        if self.broken_page:
+            return ""
         # Find intersecting cells on the page
         text_piece = ""
         page_size = self.get_size()
@@ -58,6 +70,9 @@ class DoclingParsePageBackend(PdfPageBackend):
         cells = []
         cell_counter = 0
 
+        if self.broken_page:
+            return cells
+
         page_size = self.get_size()
 
         parser_width = self._dpage["width"]
@@ -168,38 +183,39 @@ class DoclingParsePageBackend(PdfPageBackend):
     def unload(self):
         self._ppage = None
         self._dpage = None
-        self.text_page = None
 
 
 class DoclingParseDocumentBackend(PdfDocumentBackend):
-    def __init__(self, path_or_stream: Union[BytesIO, Path]):
-        super().__init__(path_or_stream)
-        self._pdoc = pdfium.PdfDocument(path_or_stream)
-        # Parsing cells with docling_parser call
-        parser = pdf_parser()
+    def __init__(self, path_or_stream: Union[BytesIO, Path], document_hash: str):
+        super().__init__(path_or_stream, document_hash)
 
-        start_pb_time = time.time()
+        self._pdoc = pdfium.PdfDocument(path_or_stream)
+        self.parser = pdf_parser()
 
+        success = False
         if isinstance(path_or_stream, BytesIO):
-            self._parser_doc = parser.find_cells_from_bytesio(path_or_stream)
-        else:
-            self._parser_doc = parser.find_cells(str(path_or_stream))
+            success = self.parser.load_document_from_bytesio(
+                document_hash, path_or_stream
+            )
+        elif isinstance(path_or_stream, Path):
+            success = self.parser.load_document(document_hash, str(path_or_stream))
 
-        end_pb_time = time.time() - start_pb_time
-        _log.info(f"Time to parse with docling-parse: time={end_pb_time:.3f}")
+        if not success:
+            raise RuntimeError("docling-parse could not load this document.")
 
     def page_count(self) -> int:
-        return len(self._parser_doc["pages"])
+        return len(self._pdoc)  # To be replaced with docling-parse API
 
     def load_page(self, page_no: int) -> DoclingParsePageBackend:
         return DoclingParsePageBackend(
-            self._pdoc[page_no], self._parser_doc["pages"][page_no]
+            self.parser, self.document_hash, page_no, self._pdoc[page_no]
         )
 
     def is_valid(self) -> bool:
         return self.page_count() > 0
 
     def unload(self):
+        super().unload()
+        self.parser.unload_document(self.document_hash)
         self._pdoc.close()
         self._pdoc = None
-        self._parser_doc = None

docling/backend/pypdfium2_backend.py

@@ -215,8 +215,8 @@ class PyPdfiumPageBackend(PdfPageBackend):
 
 
 class PyPdfiumDocumentBackend(PdfDocumentBackend):
-    def __init__(self, path_or_stream: Union[BytesIO, Path]):
-        super().__init__(path_or_stream)
+    def __init__(self, path_or_stream: Union[BytesIO, Path], document_hash: str):
+        super().__init__(path_or_stream, document_hash)
         self._pdoc = pdfium.PdfDocument(path_or_stream)
 
     def page_count(self) -> int:
@@ -229,5 +229,6 @@ class PyPdfiumDocumentBackend(PdfDocumentBackend):
         return self.page_count() > 0
 
     def unload(self):
+        super().unload()
         self._pdoc.close()
         self._pdoc = None

docling/datamodel/document.py

@@ -79,7 +79,9 @@ class InputDocument(BaseModel):
                     self.valid = False
                 else:
                     self.document_hash = create_file_hash(path_or_stream)
-                    self._backend = pdf_backend(path_or_stream=path_or_stream)
+                    self._backend = pdf_backend(
+                        path_or_stream=path_or_stream, document_hash=self.document_hash
+                    )
 
             elif isinstance(path_or_stream, BytesIO):
                 self.file = PurePath(filename)
@@ -89,7 +91,9 @@ class InputDocument(BaseModel):
                     self.valid = False
                 else:
                     self.document_hash = create_file_hash(path_or_stream)
-                    self._backend = pdf_backend(path_or_stream=path_or_stream)
+                    self._backend = pdf_backend(
+                        path_or_stream=path_or_stream, document_hash=self.document_hash
+                    )
 
             if self.document_hash and self._backend.page_count() > 0:
                 self.page_count = self._backend.page_count()

docling/document_converter.py

@@ -141,6 +141,8 @@ class DocumentConverter:
         start_doc_time = time.time()
         converted_doc = ConvertedDocument(input=in_doc)
 
+        _log.info(f"Processing document {in_doc.file.name}")
+
         if not in_doc.valid:
             converted_doc.status = ConversionStatus.FAILURE
             return converted_doc

{docling-1.6.3.dist-info → docling-1.7.1.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: docling
-Version: 1.6.3
+Version: 1.7.1
 Summary: Docling PDF conversion package
 Home-page: https://github.com/DS4SD/docling
 License: MIT
@@ -23,7 +23,7 @@ Requires-Dist: certifi (>=2024.7.4)
 Requires-Dist: deepsearch-glm (>=0.19.0,<1)
 Requires-Dist: docling-core (>=1.1.2,<2.0.0)
 Requires-Dist: docling-ibm-models (>=1.1.2,<2.0.0)
-Requires-Dist: docling-parse (>=0.2.0,<0.3.0)
+Requires-Dist: docling-parse (>=1.1.1,<2.0.0)
 Requires-Dist: easyocr (>=1.7,<2.0)
 Requires-Dist: filetype (>=1.2.0,<2.0.0)
 Requires-Dist: huggingface_hub (>=0.23,<1)

{docling-1.6.3.dist-info → docling-1.7.1.dist-info}/RECORD

@@ -1,13 +1,13 @@
 docling/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 docling/backend/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-docling/backend/abstract_backend.py,sha256=ZfEHaBPGM1cmqrhaEoU3MHhnHU11NhOnhtFEIbVMYDo,1221
-docling/backend/docling_parse_backend.py,sha256=ELDJeC0bHYWEtkMcvcPxTMIbTBLO1N9VLeqsardlXg4,6880
-docling/backend/pypdfium2_backend.py,sha256=xUiIYgd7i22YDx4-W2hfPUaQFszW0gcT6pavG5qZ8LE,8062
+docling/backend/abstract_backend.py,sha256=wvrywm1pPt79L5Dt_da5QGmf9dDzjBGj1rSaUQxqI8s,1432
+docling/backend/docling_parse_backend.py,sha256=hXyF2VPPdLs7APWEXTlfz0wI86rUGYa67Q73zgTB-Ug,7438
+docling/backend/pypdfium2_backend.py,sha256=84AnFah8Ztk-j8_9MTHalPU3a9fClrEz7A_rfzWDkFc,8122
 docling/datamodel/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 docling/datamodel/base_models.py,sha256=5VHit5h7OleKnbhvy-sWDxQLizEdNrGUBrypyzwHyAE,8604
-docling/datamodel/document.py,sha256=Dgi9pSwXCgIoR26MKiRDiVMyMaFKdvGSKq2Fm5Lef9M,13173
+docling/datamodel/document.py,sha256=J97KeT8fJRKijUorDky-xA2FoOGBXOjrReYjdeo8NK4,13333
 docling/datamodel/settings.py,sha256=t5g6wrEJnPa9gBzMMl8ppgBRUYz-8xgopEtfMS0ZH28,733
-docling/document_converter.py,sha256=UFSELvUSWsr8s0VByu4lNuzu7bn7zZauJTL3FTSLSBg,10371
+docling/document_converter.py,sha256=Tx4BwtOxpwizmXgJl2nK6I-6m0V3fygHwYFomJTH2Ns,10433
 docling/models/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 docling/models/base_ocr_model.py,sha256=Ipl82a3AV2OsgMQSMEMpnWJ6MXcmyIQzmp52PmTaB0g,4465
 docling/models/ds_glm_model.py,sha256=wmb--2JKFQby-kvidw6PyM8wURPXYPQ_Z_eKKCBAdYQ,3192
@@ -21,7 +21,7 @@ docling/pipeline/standard_model_pipeline.py,sha256=UTjyaEXvz9htYZz-IMTkn11cZwNjg
 docling/utils/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 docling/utils/layout_utils.py,sha256=FOFbL0hKzUoWXdZaeUvEtFqKv0IkPifIr4sdGW4suKs,31804
 docling/utils/utils.py,sha256=llhXSbIDNZ1MHOwBEfLHBAoJIAYI7QlPIonlI1jLUJ0,1208
-docling-1.6.3.dist-info/LICENSE,sha256=ACwmltkrXIz5VsEQcrqljq-fat6ZXAMepjXGoe40KtE,1069
-docling-1.6.3.dist-info/METADATA,sha256=DeRKK5TVCv9rp3eQfZkXfZXwKLi4df2l10qXKcm3ISQ,7229
-docling-1.6.3.dist-info/WHEEL,sha256=sP946D7jFCHeNz5Iq4fL4Lu-PrWrFsgfLXbbkciIZwg,88
-docling-1.6.3.dist-info/RECORD,,
+docling-1.7.1.dist-info/LICENSE,sha256=ACwmltkrXIz5VsEQcrqljq-fat6ZXAMepjXGoe40KtE,1069
+docling-1.7.1.dist-info/METADATA,sha256=ADdVabYgc4VEIGKhM-tI6XBU_CG9tzKl_au69TZ9LbY,7229
+docling-1.7.1.dist-info/WHEEL,sha256=sP946D7jFCHeNz5Iq4fL4Lu-PrWrFsgfLXbbkciIZwg,88
+docling-1.7.1.dist-info/RECORD,,