extract-python 0.10.0__py3-none-any.whl → 0.10.2__py3-none-any.whl
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
extract_python/docling_.py
CHANGED
|
@@ -182,8 +182,6 @@ def _to_markdown_doc(
|
|
|
182
182
|
# nested in the tree structured
|
|
183
183
|
md_dir_name = path_to_artifacts_dirname(input_path)
|
|
184
184
|
md_dir = output_path / md_dir_name
|
|
185
|
-
if md_dir.exists():
|
|
186
|
-
raise FileExistsError(f"directory {md_dir} already exists")
|
|
187
185
|
# Let's avoid issue of duplicated input file names flattened top level
|
|
188
186
|
md_filename = md_dir_name + OutputFormat.MARKDOWN
|
|
189
187
|
with tempfile.TemporaryDirectory(ignore_cleanup_errors=True) as td:
|
|
@@ -1,13 +1,13 @@
|
|
|
1
1
|
Metadata-Version: 2.5
|
|
2
2
|
Name: extract-python
|
|
3
|
-
Version: 0.10.
|
|
3
|
+
Version: 0.10.2
|
|
4
4
|
Summary: Structured content extraction
|
|
5
5
|
Project-URL: Homepage, https://github.com/ICIJ/extract-python
|
|
6
6
|
Project-URL: Repository, https://github.com/ICIJ/extract-python
|
|
7
7
|
Project-URL: Issues, https://github.com/ICIJ/extract-python/issues
|
|
8
8
|
Author-email: Clément Doumouro <cdoumouro@icij.org>
|
|
9
9
|
Requires-Python: <3.15,>=3.13
|
|
10
|
-
Requires-Dist: extract-core~=0.10.
|
|
10
|
+
Requires-Dist: extract-core~=0.10.4
|
|
11
11
|
Requires-Dist: icij-common~=0.8.2
|
|
12
12
|
Requires-Dist: pympler~=1.1
|
|
13
13
|
Provides-Extra: benches
|
|
@@ -16,7 +16,7 @@ Requires-Dist: markdown2>=2.5.4; extra == 'benches'
|
|
|
16
16
|
Requires-Dist: notebook>=7.4.5; extra == 'benches'
|
|
17
17
|
Requires-Dist: pypdfium2>=4.30.0; extra == 'benches'
|
|
18
18
|
Provides-Extra: docling
|
|
19
|
-
Requires-Dist: docling-slim[feat-ocr-easyocr,feat-ocr-mac,feat-ocr-tesserocr,standard]~=2.127; extra == 'docling'
|
|
19
|
+
Requires-Dist: docling-slim[feat-ocr-easyocr,feat-ocr-mac,feat-ocr-nemotron,feat-ocr-rapidocr-onnx,feat-ocr-tesserocr,standard]~=2.127; extra == 'docling'
|
|
20
20
|
Provides-Extra: marker
|
|
21
21
|
Requires-Dist: marker-pdf~=1.10; extra == 'marker'
|
|
22
22
|
Provides-Extra: mineru
|
|
@@ -1,9 +1,9 @@
|
|
|
1
1
|
extract_python/__init__.py,sha256=DA2LUro6vMjfS8fb2MsqO95FbJEZHyZ7kFyn42q02Wk,759
|
|
2
2
|
extract_python/constants.py,sha256=659V40LcTWJhX3IbuJLSSvI5AsGJh9ciMrGCfzJn2zA,98
|
|
3
|
-
extract_python/docling_.py,sha256=
|
|
3
|
+
extract_python/docling_.py,sha256=zD5rIXV1pAQBNGZc0RliP9buHjttz4CdSOj21Z4FbFo,10054
|
|
4
4
|
extract_python/marker_.py,sha256=bb0DJXrSWOznIuPnw4-adR5_R9QiwtFGGBh5KlXFxjM,3323
|
|
5
5
|
extract_python/miner_u.py,sha256=sEU1PhPpfPVo4jP7flNlbGO9PfsLFLz973-ufZEd0ws,3966
|
|
6
6
|
extract_python/utils.py,sha256=LrR1tlOgpWEri6TLDlYYPRBmlka-AjOepWnbySf2Haw,9974
|
|
7
|
-
extract_python-0.10.
|
|
8
|
-
extract_python-0.10.
|
|
9
|
-
extract_python-0.10.
|
|
7
|
+
extract_python-0.10.2.dist-info/METADATA,sha256=FjrVsPTQOccD7fwPXGJWf89oncMILzBkJ-SyWHFflb8,1244
|
|
8
|
+
extract_python-0.10.2.dist-info/WHEEL,sha256=W3fkpkm7-wf9vBI5Z-7s0eWkeM-spu78I8Neb98DeEg,87
|
|
9
|
+
extract_python-0.10.2.dist-info/RECORD,,
|
|
File without changes
|