37 lines
1.3 KiB
Python
37 lines
1.3 KiB
Python
|
"""Vergleicht die Wörter zweier epub2md-Ausgaben (Datenverlust-Check nach einer Codeänderung).
|
||
|
|
|
||
|
|
Aufruf: python tools/qa/worddiff.py ALT_AUSGABE NEU_AUSGABE
|
||
|
|
Gezählt werden die Wörter aller chapters/*.md ohne Frontmatter, Markup, Link- und Bildziele. Gibt die Wörter
|
||
|
|
aus, die nur in einer der beiden Ausgaben vorkommen (Multimengen-Differenz).
|
||
|
|
"""
|
||
|
|
|
||
|
|
from __future__ import annotations
|
||
|
|
|
||
|
|
import collections
|
||
|
|
import glob
|
||
|
|
import re
|
||
|
|
import sys
|
||
|
|
|
||
|
|
|
||
|
|
def words(out_dir: str) -> collections.Counter[str]:
|
||
|
|
counter: collections.Counter[str] = collections.Counter()
|
||
|
|
for path in sorted(glob.glob(out_dir + "/chapters/*.md")):
|
||
|
|
with open(path, encoding="utf-8") as fh:
|
||
|
|
text = fh.read()
|
||
|
|
text = re.sub(r"\A---\n.*?\n---\n", "", text, flags=re.S)
|
||
|
|
text = re.sub(r"\]\([^)]*\)|<[^>]+>|[*_`#>|\\\[\]↪➥]", " ", text)
|
||
|
|
counter.update(re.findall(r"\w+", text))
|
||
|
|
return counter
|
||
|
|
|
||
|
|
|
||
|
|
def main() -> None:
|
||
|
|
old, new = words(sys.argv[1]), words(sys.argv[2])
|
||
|
|
lost, gained = old - new, new - old
|
||
|
|
print(f"Wörter alt {sum(old.values())}, neu {sum(new.values())}; "
|
||
|
|
f"nur alt {sum(lost.values())}, nur neu {sum(gained.values())}")
|
||
|
|
print("nur alt:", lost.most_common(25))
|
||
|
|
print("nur neu:", gained.most_common(25))
|
||
|
|
|
||
|
|
|
||
|
|
if __name__ == "__main__":
|
||
|
|
main()
|