30 lines
934 B
Python
30 lines
934 B
Python
import re
|
|
from pathlib import Path
|
|
|
|
def natural_key(text):
|
|
return [int(c) if c.isdigit() else c.lower() for c in re.split(r'(\d+)', str(text))]
|
|
|
|
def read_all_labels(base_dir):
|
|
# return sorted(list(filter(None,
|
|
# (Path(base_dir) / "labels").read_text().splitlines())),
|
|
# key = natural_key)
|
|
return list(filter(None, (Path(base_dir) / "labels").read_text().splitlines()))
|
|
|
|
def enonce_total(base_dir):
|
|
text_dir = Path(base_dir) / 'Text'
|
|
if not text_dir.is_dir():
|
|
return ""
|
|
|
|
# Exclude .tex and .pdf files
|
|
files = [f for f in text_dir.iterdir()
|
|
if f.is_file() and f.suffix.lower() not in ('.tex', '.pdf')]
|
|
|
|
files.sort(key=lambda f: natural_key(f.name))
|
|
|
|
output = []
|
|
for filepath in files:
|
|
content = filepath.read_text(encoding='utf-8')
|
|
output.append(f"{filepath.name}\n{content}\n\n\n")
|
|
|
|
return "".join(output)
|