docling 1.1.0__py3-none-any.whl → 1.1.2__py3-none-any.whl

docling/datamodel/document.py

@@ -125,7 +125,7 @@ class ConvertedDocument(BaseModel):
         desc = DsDocumentDescription(logs=[])
 
         page_hashes = [
-            PageReference(hash=p.page_hash, page=p.page_no, model="default")
+            PageReference(hash=p.page_hash, page=p.page_no + 1, model="default")
             for p in self.pages
         ]
 
@@ -159,7 +159,7 @@ class ConvertedDocument(BaseModel):
                         prov=[
                             Prov(
                                 bbox=target_bbox,
-                                page=element.page_no,
+                                page=element.page_no + 1,
                                 span=[0, len(element.text)],
                             )
                         ],
@@ -242,7 +242,7 @@ class ConvertedDocument(BaseModel):
                         prov=[
                             Prov(
                                 bbox=target_bbox,
-                                page=element.page_no,
+                                page=element.page_no + 1,
                                 span=[0, 0],
                             )
                         ],
@@ -264,7 +264,7 @@ class ConvertedDocument(BaseModel):
                         prov=[
                             Prov(
                                 bbox=target_bbox,
-                                page=element.page_no,
+                                page=element.page_no + 1,
                                 span=[0, 0],
                             )
                         ],
@@ -274,7 +274,7 @@ class ConvertedDocument(BaseModel):
                 )
 
         page_dimensions = [
-            PageDimensions(page=p.page_no, height=p.size.height, width=p.size.width)
+            PageDimensions(page=p.page_no + 1, height=p.size.height, width=p.size.width)
             for p in self.pages
         ]
 

docling/models/table_structure_model.py

@@ -114,12 +114,15 @@ class TableStructureModel:
                     for element in table_out["tf_responses"]:
 
                         if not self.do_cell_matching:
-                            the_bbox = BoundingBox.model_validate(element["bbox"])
+                            the_bbox = BoundingBox.model_validate(
+                                element["bbox"]
+                            ).scaled(1 / self.scale)
                             text_piece = page._backend.get_text_in_rect(the_bbox)
                             element["bbox"]["token"] = text_piece
 
                         tc = TableCell.model_validate(element)
-                        tc.bbox = tc.bbox.scaled(1 / self.scale)
+                        if self.do_cell_matching:
+                            tc.bbox = tc.bbox.scaled(1 / self.scale)
                         table_cells.append(tc)
 
                     # Retrieving cols/rows, after post processing:

{docling-1.1.0.dist-info → docling-1.1.2.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: docling
-Version: 1.1.0
+Version: 1.1.2
 Summary: Docling PDF conversion package
 Home-page: https://github.com/DS4SD/docling
 License: MIT
@@ -22,7 +22,7 @@ Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
 Provides-Extra: easyocr
 Provides-Extra: ocr
 Requires-Dist: deepsearch-glm (>=0.19.0,<1)
-Requires-Dist: docling-core (>=1.1.0,<2.0.0)
+Requires-Dist: docling-core (>=1.1.2,<2.0.0)
 Requires-Dist: docling-ibm-models (>=1.1.0,<2.0.0)
 Requires-Dist: easyocr (>=1.7,<2.0) ; extra == "easyocr" or extra == "ocr"
 Requires-Dist: filetype (>=1.2.0,<2.0.0)
@@ -92,7 +92,7 @@ print(doc.export_to_markdown())  # output: "## DocLayNet: A Large Human-Annotate
 
 ### Convert a batch of documents
 
-For an example of converting multiple documents, see [convert.py](https://github.com/DS4SD/docling/blob/main/examples/convert.py).
+For an example of batch-converting documents, see [convert.py](https://github.com/DS4SD/docling/blob/main/examples/convert.py).
 
 From a local repo clone, you can run it with:
 

{docling-1.1.0.dist-info → docling-1.1.2.dist-info}/RECORD

@@ -4,7 +4,7 @@ docling/backend/abstract_backend.py,sha256=dINr8oTax9Fq31Y1AR0CGWNZtAHN5aqB_M7TA
 docling/backend/pypdfium2_backend.py,sha256=cIQGFkwzceN57PzmACt06CytRo0A_t-im6rW804RC3M,7421
 docling/datamodel/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 docling/datamodel/base_models.py,sha256=k7gLFPnq3ArEMAFz6qUcp5qemlYzVhOmR9qtBTkAiX4,6862
-docling/datamodel/document.py,sha256=7caefzaii6itMQgtXfA4SJhB1TAF32v1c8zRwbiU03s,12497
+docling/datamodel/document.py,sha256=FG_ntDFRBWj-MhV52D0sC8XaZOwN3yryyXahsVHGnyI,12517
 docling/datamodel/settings.py,sha256=t5g6wrEJnPa9gBzMMl8ppgBRUYz-8xgopEtfMS0ZH28,733
 docling/document_converter.py,sha256=I9vjTLCLahsMrcs9ozM3C5r_CtBN-9qHk7-ANma7fkc,9895
 docling/models/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
@@ -12,14 +12,14 @@ docling/models/ds_glm_model.py,sha256=wmb--2JKFQby-kvidw6PyM8wURPXYPQ_Z_eKKCBAdY
 docling/models/easyocr_model.py,sha256=NaHVs8IN0eW9KB076E2Kae1s-bq74_4IMWueze9QqtE,2290
 docling/models/layout_model.py,sha256=4AfPFiu6pXc8wIQ1sQlEZnHRt7SnBmfzDdctiRveOWw,10944
 docling/models/page_assemble_model.py,sha256=8eoG2WiFxPxq9TPvM-wkngb2gkr0tdtCRVXg1JcTETo,5550
-docling/models/table_structure_model.py,sha256=ryZrmkNkCbw5SCpgdQabkmcRAEi_4VqOMv2VGdpvGZo,5499
+docling/models/table_structure_model.py,sha256=xUmfunZNYC30P0fRdESdztqy1FVlMzlhJjLBp-xcn4A,5638
 docling/pipeline/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 docling/pipeline/base_model_pipeline.py,sha256=ozHdJak0yQAxQf7pQN_C480vI35A2e5KL5Qq1xSkq5c,560
 docling/pipeline/standard_model_pipeline.py,sha256=UTwodKUKrisLoVcntbNUBDhjzRyFvpdUvyVw-gNmBlM,1541
 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-1.1.0.dist-info/LICENSE,sha256=ACwmltkrXIz5VsEQcrqljq-fat6ZXAMepjXGoe40KtE,1069
-docling-1.1.0.dist-info/METADATA,sha256=mUAryQOsHRejcJ3Qb4zFvRVWpcKX0e4aycnJM_OE0o0,6759
-docling-1.1.0.dist-info/WHEEL,sha256=sP946D7jFCHeNz5Iq4fL4Lu-PrWrFsgfLXbbkciIZwg,88
-docling-1.1.0.dist-info/RECORD,,
+docling-1.1.2.dist-info/LICENSE,sha256=ACwmltkrXIz5VsEQcrqljq-fat6ZXAMepjXGoe40KtE,1069
+docling-1.1.2.dist-info/METADATA,sha256=3uSlwJCchlMWLELW4Sr3L6apbAPt4sOZem3T7NlglU8,6756
+docling-1.1.2.dist-info/WHEEL,sha256=sP946D7jFCHeNz5Iq4fL4Lu-PrWrFsgfLXbbkciIZwg,88
+docling-1.1.2.dist-info/RECORD,,