# Alte edwiki.de-Tagesseiten galnet/2014/MM/TT.txt -> eine Meldung je Eintrag (DE) import re, sys, json, glob from pathlib import Path out = [] for f in sorted(glob.glob(sys.argv[1] + '/*/[0-9][0-9].txt')): mm, dd = f.split('/')[-2], Path(f).stem t = Path(f).read_text(encoding='utf-8') n = 0 for blk in re.findall(r'\n(={5,6} .+?)\n\n\n\n\n\{\{ ?([^?}]+)', t, re.S): text, img = blk m = re.match(r'={5,6} (.+?) ={5,6}\n', text) n += 1 body = text[m.end():].strip() out.append({'id': f'2014:{mm}:{dd}_{n:02d}', 'mm': mm, 'dd': dd, 'title': m.group(1).strip(), 'body': body, 'old_img': img.strip()}) json.dump(out, open(sys.argv[2], 'w'), ensure_ascii=False, indent=1) print(len(out))