omni-split 0.0.3__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.
- omni_split/__init__.py +16 -0
- omni_split/base/__init__.py +0 -0
- omni_split/base/chonkie_base.py +139 -0
- omni_split/base/chonkie_tokenizer.py +285 -0
- omni_split/base/chonkie_types.py +519 -0
- omni_split/base/md2json_list.py +303 -0
- omni_split/base/md_json_list2chunk.py +348 -0
- omni_split/main.py +73 -0
- omni_split/model/text_chunker_tokenizer/qwen_tokenizer.json +303282 -0
- omni_split/omni_split.py +93 -0
- omni_split/sub_chunker/__init__.py +0 -0
- omni_split/sub_chunker/document_split.py +32 -0
- omni_split/sub_chunker/markdown_split.py +47 -0
- omni_split/sub_chunker/text_split.py +343 -0
- omni_split/test.py +80 -0
- omni_split/utils/__init__.py +0 -0
- omni_split/utils/base_utils.py +181 -0
- omni_split/utils/download_test_doc.py +61 -0
- omni_split-0.0.3.dist-info/METADATA +147 -0
- omni_split-0.0.3.dist-info/RECORD +23 -0
- omni_split-0.0.3.dist-info/WHEEL +5 -0
- omni_split-0.0.3.dist-info/licenses/LICENSE +21 -0
- omni_split-0.0.3.dist-info/top_level.txt +1 -0
omni_split/main.py
ADDED
|
@@ -0,0 +1,73 @@
|
|
|
1
|
+
import json
|
|
2
|
+
from omni_split import OmniSplit
|
|
3
|
+
from omni_split import word_preprocessing_and_return_bytesIO
|
|
4
|
+
|
|
5
|
+
omni_spliter = OmniSplit()
|
|
6
|
+
## note: test text split
|
|
7
|
+
test_text = True
|
|
8
|
+
if test_text:
|
|
9
|
+
with open("omni_split/test/text_test.txt", "r") as f:
|
|
10
|
+
text_content = "".join(f.readlines())
|
|
11
|
+
res = omni_spliter.text_chunk_func(text_content)
|
|
12
|
+
for item in res:
|
|
13
|
+
print(item)
|
|
14
|
+
print("------------")
|
|
15
|
+
print("=" * 10)
|
|
16
|
+
|
|
17
|
+
## note: test markdown json split
|
|
18
|
+
test_markdown = True
|
|
19
|
+
if test_markdown:
|
|
20
|
+
with open("omni_split/test/json_list_test.json", "r") as f:
|
|
21
|
+
md_content_json = json.load(f)
|
|
22
|
+
res = omni_spliter.markdown_json_chunk_func(md_content_json)
|
|
23
|
+
for item in res:
|
|
24
|
+
print(item)
|
|
25
|
+
print("------------")
|
|
26
|
+
print("=" * 10)
|
|
27
|
+
|
|
28
|
+
res = omni_spliter.markdown_json_chunk_func(md_content_json, clear_model=True)
|
|
29
|
+
for item in res:
|
|
30
|
+
print(item)
|
|
31
|
+
print("------------")
|
|
32
|
+
print("=" * 10)
|
|
33
|
+
|
|
34
|
+
## note: test markdown split
|
|
35
|
+
test_markdown = True
|
|
36
|
+
if test_markdown:
|
|
37
|
+
with open("omni_split/test/markdown_test.md", "r") as f:
|
|
38
|
+
md_content = f.read()
|
|
39
|
+
res = omni_spliter.markdown_chunk_func(md_content)
|
|
40
|
+
for item in res:
|
|
41
|
+
print(item)
|
|
42
|
+
print("------------")
|
|
43
|
+
print("=" * 10)
|
|
44
|
+
|
|
45
|
+
res = omni_spliter.markdown_chunk_func(md_content, clear_model=True)
|
|
46
|
+
for item in res:
|
|
47
|
+
print(item)
|
|
48
|
+
print("------------")
|
|
49
|
+
print("=" * 10)
|
|
50
|
+
|
|
51
|
+
|
|
52
|
+
## note: test word split
|
|
53
|
+
test_document = True
|
|
54
|
+
if test_document:
|
|
55
|
+
|
|
56
|
+
new_doc_io = word_preprocessing_and_return_bytesIO("omni_split/test/docx_test.docx")
|
|
57
|
+
res = omni_spliter.document_chunk_func(new_doc_io, txt_chunk_size=1000, clear_model=False)
|
|
58
|
+
for item in res:
|
|
59
|
+
print(item)
|
|
60
|
+
print("------------")
|
|
61
|
+
print("=" * 10)
|
|
62
|
+
|
|
63
|
+
res = omni_spliter.document_chunk_func(new_doc_io, txt_chunk_size=1000, clear_model=False, save_local_images_dir="./images")
|
|
64
|
+
for item in res:
|
|
65
|
+
print(item)
|
|
66
|
+
print("------------")
|
|
67
|
+
print("=" * 10)
|
|
68
|
+
|
|
69
|
+
res = omni_spliter.document_chunk_func(new_doc_io, txt_chunk_size=1000, clear_model=True)
|
|
70
|
+
for item in res:
|
|
71
|
+
print(item)
|
|
72
|
+
print("------------")
|
|
73
|
+
print("=" * 10)
|