64 lines
1.8 KiB
Python
64 lines
1.8 KiB
Python
#!/usr/bin/env python3
|
||
# -*- coding: utf-8 -*-
|
||
"""
|
||
Convierte el TTML/XML (archivo .srt en este repositorio) a texto plano
|
||
agruparndo cada <p> como un párrafo y juntando los <s> en una sola línea.
|
||
Guarda el resultado en el mismo directorio con extensión .txt
|
||
"""
|
||
import sys
|
||
from pathlib import Path
|
||
import re
|
||
from xml.etree import ElementTree as ET
|
||
|
||
in_path = Path('/media/kyman/SSD2TB/git.copilot/OpoTests/Public/data/srt/Real Decreto Legislativo 5⁄2015, TREBEP - Art 1 a 7 - 1a parte. (1080p_30fps_H264-128kbit_AAC).español.srt')
|
||
if not in_path.exists():
|
||
print('ERROR: input file not found:', in_path)
|
||
sys.exit(2)
|
||
|
||
out_path = in_path.with_suffix('.txt')
|
||
|
||
try:
|
||
tree = ET.parse(in_path)
|
||
root = tree.getroot()
|
||
except Exception as e:
|
||
# try to recover by reading text and parsing from <body>
|
||
txt = in_path.read_text(encoding='utf-8')
|
||
idx = txt.find('<body>')
|
||
if idx != -1:
|
||
txt = txt[idx:]
|
||
root = ET.fromstring('<root>' + txt + '</root>')
|
||
else:
|
||
raise
|
||
|
||
paragraphs = []
|
||
for p in root.findall('.//p'):
|
||
parts = []
|
||
for s in p.findall('.//s'):
|
||
t = s.text or ''
|
||
t = t.strip()
|
||
if t:
|
||
parts.append(t)
|
||
if not parts:
|
||
# fallback to any text content
|
||
text = ''.join(p.itertext()).strip()
|
||
if text:
|
||
parts = [text]
|
||
para = ' '.join(parts)
|
||
# normalize whitespace
|
||
para = re.sub(r"\s+", ' ', para).strip()
|
||
if para:
|
||
paragraphs.append(para)
|
||
|
||
full_text = '\n\n'.join(paragraphs)
|
||
full_text = re.sub(r'<[^>]+>', '', full_text)
|
||
|
||
out_path.write_text(full_text, encoding='utf-8')
|
||
print(f'WROTE: {out_path}')
|
||
print(f'PARAGRAPHS: {len(paragraphs)}')
|
||
# print a short preview to confirm
|
||
preview = full_text[:4000]
|
||
print('\n---PREVIEW---\n')
|
||
print(preview)
|
||
print('\n---END PREVIEW---\n')
|
||
print('Done')
|