Upload 2 files
Browse files- convert_AKCES.py +72 -0
- test.jsonl +0 -0
convert_AKCES.py
ADDED
@@ -0,0 +1,72 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
import os
|
2 |
+
import json
|
3 |
+
import re
|
4 |
+
import subprocess
|
5 |
+
from docx import Document
|
6 |
+
|
7 |
+
FILE_PATH = ".data/AKCES/AKCES 1/kontroly-RTF/"
|
8 |
+
OUTPUT_FILE = ".data/hf_dataset/cnc_skript2012/test.jsonl"
|
9 |
+
|
10 |
+
|
11 |
+
def convert_to_docx(file_path):
|
12 |
+
"""Convert a given file to DOCX using LibreOffice."""
|
13 |
+
new_file_path = file_path + 'x'
|
14 |
+
command = ['libreoffice', '--headless', '--convert-to', 'docx', '--outdir', os.path.dirname(file_path), file_path]
|
15 |
+
subprocess.run(command, check=True)
|
16 |
+
return new_file_path
|
17 |
+
|
18 |
+
|
19 |
+
def read_docx_file(file_path):
|
20 |
+
"""Extract text from a DOCX file."""
|
21 |
+
doc = Document(file_path)
|
22 |
+
text = "\n".join([para.text for para in doc.paragraphs])
|
23 |
+
return text
|
24 |
+
|
25 |
+
|
26 |
+
def main():
|
27 |
+
# Step 1: Convert all RTF and DOC files to DOCX
|
28 |
+
for root, dirs, files in os.walk(FILE_PATH):
|
29 |
+
for file in files:
|
30 |
+
file_path = os.path.join(root, file)
|
31 |
+
if file.endswith('.rtf') or file.endswith('.doc'):
|
32 |
+
# if the docx file already exists, skip the conversion
|
33 |
+
docx_file_path = file_path[:-3] + 'docx'
|
34 |
+
if os.path.exists(docx_file_path):
|
35 |
+
continue
|
36 |
+
try:
|
37 |
+
convert_to_docx(file_path)
|
38 |
+
except subprocess.CalledProcessError as e:
|
39 |
+
raise ValueError(f"Error converting {file_path}: {e}")
|
40 |
+
|
41 |
+
# Step 2: Process all DOCX files to extract text
|
42 |
+
output_data = []
|
43 |
+
for root, dirs, files in os.walk(FILE_PATH):
|
44 |
+
for file in files:
|
45 |
+
file_path = os.path.join(root, file)
|
46 |
+
if file.endswith('.docx'):
|
47 |
+
try:
|
48 |
+
text = read_docx_file(file_path)
|
49 |
+
|
50 |
+
# postprocessing
|
51 |
+
text = text.replace("\xa0", " ")
|
52 |
+
text = text.replace("\n \n", "\n\n")
|
53 |
+
text = re.sub(r' +\n', '\n', text)
|
54 |
+
text = re.sub(r' +', ' ', text)
|
55 |
+
text = text.strip()
|
56 |
+
|
57 |
+
output_data.append({
|
58 |
+
"text": text,
|
59 |
+
"original_file": file[:-5], # remove the extension
|
60 |
+
})
|
61 |
+
except Exception as e:
|
62 |
+
print(f"Error reading {file_path}: {e}")
|
63 |
+
|
64 |
+
os.makedirs(os.path.dirname(OUTPUT_FILE), exist_ok=True)
|
65 |
+
# Write the output data to a JSONL file
|
66 |
+
with open(OUTPUT_FILE, 'w', encoding='utf-8') as jsonl_file:
|
67 |
+
for entry in output_data:
|
68 |
+
jsonl_file.write(json.dumps(entry, ensure_ascii=False) + '\n')
|
69 |
+
|
70 |
+
|
71 |
+
if __name__ == "__main__":
|
72 |
+
main()
|
test.jsonl
ADDED
The diff for this file is too large to render.
See raw diff
|
|