docling 2.28.2__py3-none-any.whl → 2.28.4__py3-none-any.whl

docling/models/table_structure_model.py

@@ -5,7 +5,11 @@ from typing import Iterable, Optional, Union
 
 import numpy
 from docling_core.types.doc import BoundingBox, DocItemLabel, TableCell
-from docling_core.types.doc.page import BoundingRectangle
+from docling_core.types.doc.page import (
+    BoundingRectangle,
+    SegmentedPdfPage,
+    TextCellUnit,
+)
 from docling_ibm_models.tableformer.data_management.tf_predictor import TFPredictor
 from PIL import ImageDraw
 
@@ -218,9 +222,21 @@ class TableStructureModel(BasePageModel):
 
                     if len(table_bboxes):
                         for table_cluster, tbl_box in in_tables:
-
+                            # Check if word-level cells are available from backend:
+                            sp = page._backend.get_segmented_page()
+                            if sp is not None:
+                                tcells = sp.get_cells_in_bbox(
+                                    cell_unit=TextCellUnit.WORD,
+                                    bbox=table_cluster.bbox,
+                                )
+                                if len(tcells) == 0:
+                                    # In case word-level cells yield empty
+                                    tcells = table_cluster.cells
+                            else:
+                                # Otherwise - we use normal (line/phrase) cells
+                                tcells = table_cluster.cells
                             tokens = []
-                            for c in table_cluster.cells:
+                            for c in tcells:
                                 # Only allow non empty stings (spaces) into the cells of a table
                                 if len(c.text.strip()) > 0:
                                     new_cell = copy.deepcopy(c)
@@ -229,7 +245,6 @@ class TableStructureModel(BasePageModel):
                                             scale=self.scale
                                         )
                                     )
-
                                     tokens.append(
                                         {
                                             "id": new_cell.index,

{docling-2.28.2.dist-info → docling-2.28.4.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: docling
-Version: 2.28.2
+Version: 2.28.4
 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/docling-project/docling
 License: MIT
@@ -28,7 +28,7 @@ Provides-Extra: vlm
 Requires-Dist: accelerate (>=1.2.1,<2.0.0) ; (sys_platform != "darwin" or platform_machine != "x86_64") and (extra == "vlm")
 Requires-Dist: beautifulsoup4 (>=4.12.3,<5.0.0)
 Requires-Dist: certifi (>=2024.7.4)
-Requires-Dist: docling-core[chunking] (>=2.23.1,<3.0.0)
+Requires-Dist: docling-core[chunking] (>=2.24.1,<3.0.0)
 Requires-Dist: docling-ibm-models (>=3.4.0,<4.0.0)
 Requires-Dist: docling-parse (>=4.0.0,<5.0.0)
 Requires-Dist: easyocr (>=1.7,<2.0)

{docling-2.28.2.dist-info → docling-2.28.4.dist-info}/RECORD

@@ -57,7 +57,7 @@ docling/models/plugins/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3
 docling/models/plugins/defaults.py,sha256=qslXGnRX07Z3GGttNriqaox0v0vXp4zs4KLurHCZjp4,858
 docling/models/rapid_ocr_model.py,sha256=C_I0Ek9mAPIyTFRHuNbqtXg1c15rLNDE1tJ6_hPIi4c,5869
 docling/models/readingorder_model.py,sha256=hNWbBX3uZv1FxMwKNKn2JFQuQqTspBLsJBVEidXr6Wk,14869
-docling/models/table_structure_model.py,sha256=_b6-2alzhzI19-thDGpM3mww54mxbHLkEiTYMU84d30,11773
+docling/models/table_structure_model.py,sha256=pvTsqUa5QIANBUfot0XXG1UUeku-eaUi04EPE-Yh2g0,12597
 docling/models/tesseract_ocr_cli_model.py,sha256=S-rCisPrVa3ASvOWycqQoria0PtmNqgdg8YxrLbG1ww,10067
 docling/models/tesseract_ocr_model.py,sha256=UpLAgKgJtBgbKtJELmKBNMcejJJKBCyFK0q-WgZN1Eg,9256
 docling/pipeline/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
@@ -77,8 +77,8 @@ 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=tY2ylE2aiQKkmzlSLnFW-HTfFyqUUMguW18ldd1PLfo,2868
-docling-2.28.2.dist-info/LICENSE,sha256=mBb7ErEcM8VS9OhiGHnQ2kk75HwPhr54W1Oiz3965MY,1088
-docling-2.28.2.dist-info/METADATA,sha256=ZeYjkP0ZzlpqoseGod2_iuJPW9d4B16JCeSo2b61KIw,9982
-docling-2.28.2.dist-info/WHEEL,sha256=Nq82e9rUAnEjt98J6MlVmMCZb-t9cYE2Ir1kpBmnWfs,88
-docling-2.28.2.dist-info/entry_points.txt,sha256=pIxel-UeVo1S7FhoNG5xgEfPjLZfBLi_N9TsGPtJSLo,144
-docling-2.28.2.dist-info/RECORD,,
+docling-2.28.4.dist-info/LICENSE,sha256=mBb7ErEcM8VS9OhiGHnQ2kk75HwPhr54W1Oiz3965MY,1088
+docling-2.28.4.dist-info/METADATA,sha256=bwT1X-5lpPBI8mpDI2-DeLloD2Rcf170xBdERgFsq-w,9982
+docling-2.28.4.dist-info/WHEEL,sha256=Nq82e9rUAnEjt98J6MlVmMCZb-t9cYE2Ir1kpBmnWfs,88
+docling-2.28.4.dist-info/entry_points.txt,sha256=pIxel-UeVo1S7FhoNG5xgEfPjLZfBLi_N9TsGPtJSLo,144
+docling-2.28.4.dist-info/RECORD,,