extract-python 0.8.2__py3-none-any.whl → 0.8.4__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/utils.py
CHANGED
|
@@ -230,8 +230,8 @@ def batch_per_pages(
|
|
|
230
230
|
max_pages = page_batch_size * max_page_batches
|
|
231
231
|
docs = itertools.batched(docs, chunk_size, strict=False)
|
|
232
232
|
for chunk in docs:
|
|
233
|
-
short_docs =
|
|
234
|
-
long_docs =
|
|
233
|
+
short_docs = (d for d in chunk if d.n_pages <= max_pages)
|
|
234
|
+
long_docs = (d for d in chunk if d.n_pages > max_pages)
|
|
235
235
|
# Bin fill for docs smaller than max_pages
|
|
236
236
|
offset = yield from _bin_fill(short_docs, max_pages=max_pages)
|
|
237
237
|
# otherwise we just yield chunks of max_pages except the last chunk which is
|
|
@@ -1,6 +1,6 @@
|
|
|
1
1
|
Metadata-Version: 2.5
|
|
2
2
|
Name: extract-python
|
|
3
|
-
Version: 0.8.
|
|
3
|
+
Version: 0.8.4
|
|
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
|
|
@@ -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.
|
|
19
|
+
Requires-Dist: docling-slim[feat-ocr-easyocr,feat-ocr-mac,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
|
|
@@ -3,7 +3,7 @@ extract_python/constants.py,sha256=659V40LcTWJhX3IbuJLSSvI5AsGJh9ciMrGCfzJn2zA,9
|
|
|
3
3
|
extract_python/docling_.py,sha256=H_A-6pMObQ0Q7XA62GCBv0nbGtW5rKjfOVW3giU4NCk,10146
|
|
4
4
|
extract_python/marker_.py,sha256=bb0DJXrSWOznIuPnw4-adR5_R9QiwtFGGBh5KlXFxjM,3323
|
|
5
5
|
extract_python/miner_u.py,sha256=OCIleNu-VvaJlx_zG0VuagJvQvElXV7e-aPXqHbp31A,7302
|
|
6
|
-
extract_python/utils.py,sha256=
|
|
7
|
-
extract_python-0.8.
|
|
8
|
-
extract_python-0.8.
|
|
9
|
-
extract_python-0.8.
|
|
6
|
+
extract_python/utils.py,sha256=LrR1tlOgpWEri6TLDlYYPRBmlka-AjOepWnbySf2Haw,9974
|
|
7
|
+
extract_python-0.8.4.dist-info/METADATA,sha256=iMMjpvJFarhTkbrco3oF1lNZSa2jhDLfFvAionIRJqU,1290
|
|
8
|
+
extract_python-0.8.4.dist-info/WHEEL,sha256=zOwg4jB6zX2kU910N-cMawjivD6tO8NEWvE12je1bVk,87
|
|
9
|
+
extract_python-0.8.4.dist-info/RECORD,,
|
|
File without changes
|