feat: Arbeitsordner je Video, Quellenangabe und automatisches Video-Glossar
- Ohne -o/-d schreiben yt_untertitel.py, extract_text.py und abschrift_korrigieren.py -u in abschriften/<Name>/ (ein Ordner je Video, auch für alle Ausschnitte). - Beim Download entsteht neben dem Rohtext <Rohtext>.quelle.json mit Titel, Kanal, Datum und URL, für den Quellenkopf des fertigen Textes. - Ein Glossar, das wie sein Ordner heißt (abschriften/<Name>/<Name>.glossar.txt), nutzen abschrift_korrigieren.py, glossar_vorschlag.py und aenderungen.py automatisch, nach den mit -g angegebenen gemeinsamen Glossaren. - .gitignore: abschriften/, fertig/, *.quelle.json. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
This commit is contained in:
parent
acf9e33382
commit
3aa04fff4b
11 changed files with 195 additions and 34 deletions
5
.gitignore
vendored
5
.gitignore
vendored
|
|
@ -13,6 +13,11 @@ __pycache__/
|
|||
# Eigene Ergebnisse und Testläufe
|
||||
output/
|
||||
|
||||
# Arbeitsordner je Video, fertige Abschriften und Quellenangaben: nur lokal
|
||||
abschriften/
|
||||
fertig/
|
||||
*.quelle.json
|
||||
|
||||
# Echte Reden (Urheberrecht): nur lokal; die Tests nutzen tests/daten/
|
||||
testdaten/
|
||||
|
||||
|
|
|
|||
|
|
@ -21,7 +21,9 @@ from vtt2txt import (
|
|||
add_time_options, join_segments, range_argv, resolve_time_ranges, split_segments,
|
||||
time_suffix, vtt_to_text,
|
||||
)
|
||||
from yt_untertitel import YT_DLP, base_language, default_name, original_language
|
||||
from yt_untertitel import (
|
||||
YT_DLP, base_language, default_name, original_language, work_folder, write_source,
|
||||
)
|
||||
|
||||
MODEL = "qwen3:32b"
|
||||
OLLAMA_URL = "http://127.0.0.1:11434/api/chat"
|
||||
|
|
@ -264,6 +266,22 @@ def read_glossary(path):
|
|||
return terms, replacements
|
||||
|
||||
|
||||
def video_glossary(folder):
|
||||
"""Glossary of the video in `folder`, named like the folder:
|
||||
abschriften/<Name>/<Name>.glossar.txt; None if there is none."""
|
||||
path = folder / f"{folder.resolve().name}.glossar.txt"
|
||||
return path if path.is_file() else None
|
||||
|
||||
|
||||
def with_video_glossary(glossaries, folder):
|
||||
"""The given glossaries plus the video glossary of `folder` (last, after the
|
||||
shared ones), unless it was given; returns (glossaries, added or None)."""
|
||||
found = video_glossary(folder)
|
||||
if found is None or any(g.resolve() == found.resolve() for g in glossaries):
|
||||
return list(glossaries), None
|
||||
return [*glossaries, found], found
|
||||
|
||||
|
||||
def apply_replacements(text, replacements):
|
||||
"""Replace whole words in file order; report rules without any match."""
|
||||
unmatched = []
|
||||
|
|
@ -633,8 +651,8 @@ def parse_args():
|
|||
parser.add_argument(
|
||||
"-o", "--output", type=Path,
|
||||
help="Ausgabedatei (Standard: <Eingabe ohne Endung>.korrigiert.txt, "
|
||||
"mit -u <Kanal>_<Titel>_<Datum>.korrigiert.txt; mit -z zusätzlich "
|
||||
"die Bereiche)",
|
||||
"mit -u abschriften/<Name>/<Name>.korrigiert.txt mit "
|
||||
"<Name> = <Kanal>_<Titel>_<Datum>; mit -z zusätzlich die Bereiche)",
|
||||
)
|
||||
parser.add_argument(
|
||||
"-g", "--glossar", type=Path, action="append", default=[],
|
||||
|
|
@ -709,9 +727,12 @@ def read_source(args):
|
|||
f"Der Beitrag ist auf „{lang}“; unterstützt: {', '.join(PROMPTS)} (-l)."
|
||||
)
|
||||
suffix = time_suffix(args.bereiche)
|
||||
target = args.output or Path(f"{default_name(info)}{suffix}.korrigiert.txt")
|
||||
name = default_name(info)
|
||||
target = args.output or work_folder(name) / f"{name}{suffix}.korrigiert.txt"
|
||||
raw = text + "\n"
|
||||
save_raw_text(raw_text_path(target), raw)
|
||||
raw_path = raw_text_path(target)
|
||||
save_raw_text(raw_path, raw)
|
||||
write_source(info, raw_path.with_suffix(".quelle.json"))
|
||||
return raw, target, lang
|
||||
|
||||
raw = args.input.read_text(encoding="utf-8-sig")
|
||||
|
|
@ -762,10 +783,12 @@ def main():
|
|||
parts_dir = target.with_suffix(".parts")
|
||||
|
||||
terms, replacements = [], []
|
||||
for glossary in args.glossar:
|
||||
glossaries, automatic = with_video_glossary(args.glossar, target.parent)
|
||||
for glossary in glossaries:
|
||||
more_terms, more_replacements = read_glossary(glossary)
|
||||
note = " (Video-Glossar, automatisch)" if glossary == automatic else ""
|
||||
print(
|
||||
f"Glossar {glossary}: {len(more_terms)} Begriffe, "
|
||||
f"Glossar {glossary}{note}: {len(more_terms)} Begriffe, "
|
||||
f"{len(more_replacements)} feste Ersetzungen"
|
||||
)
|
||||
terms += more_terms
|
||||
|
|
|
|||
|
|
@ -473,13 +473,16 @@ def parse_args():
|
|||
|
||||
def main():
|
||||
# abschrift_korrigieren imports this module: import it only when needed
|
||||
from abschrift_korrigieren import apply_replacements, read_glossary
|
||||
from abschrift_korrigieren import apply_replacements, read_glossary, with_video_glossary
|
||||
from yt_untertitel import base_language
|
||||
|
||||
args = parse_args()
|
||||
raw = read_raw(args.roh)
|
||||
terms, replacements = [], []
|
||||
for glossary in args.glossar:
|
||||
glossaries, automatic = with_video_glossary(args.glossar, args.korrigiert.parent)
|
||||
if automatic is not None:
|
||||
print(f"Video-Glossar (automatisch): {automatic}")
|
||||
for glossary in glossaries:
|
||||
more_terms, more_replacements = read_glossary(glossary)
|
||||
terms += more_terms
|
||||
replacements += more_replacements
|
||||
|
|
|
|||
|
|
@ -14,7 +14,7 @@ from vtt2txt import (
|
|||
)
|
||||
from yt_untertitel import (
|
||||
YT_DLP, choose_track, default_name, download, fetch_info, no_track_message,
|
||||
original_language,
|
||||
original_language, work_folder, write_source,
|
||||
)
|
||||
|
||||
FLOW_WIDTH = 80 # Zeilenbreite für --fliesstext
|
||||
|
|
@ -31,8 +31,8 @@ def parse_args():
|
|||
)
|
||||
parser.add_argument(
|
||||
"-o", "--output", type=Path,
|
||||
help="Ausgabedatei (Standard: <Kanal>_<Titel>_<Datum>.txt, "
|
||||
"mit -z zusätzlich die Bereiche)",
|
||||
help="Ausgabedatei (Standard: abschriften/<Name>/<Name>.txt mit "
|
||||
"<Name> = <Kanal>_<Titel>_<Datum>, mit -z zusätzlich die Bereiche)",
|
||||
)
|
||||
parser.add_argument(
|
||||
"-f", "--fliesstext", action="store_true",
|
||||
|
|
@ -87,7 +87,8 @@ def main():
|
|||
args = parse_args()
|
||||
info, text = fetch_text(args.url, args.lang, args.bereiche, args.exakt)
|
||||
suffix = time_suffix(args.bereiche)
|
||||
output = args.output or Path(f"{default_name(info)}{suffix}.txt")
|
||||
name = default_name(info)
|
||||
output = args.output or work_folder(name) / f"{name}{suffix}.txt"
|
||||
segments = split_segments(text)
|
||||
|
||||
if args.fliesstext:
|
||||
|
|
@ -104,6 +105,7 @@ def main():
|
|||
|
||||
output.parent.mkdir(parents=True, exist_ok=True)
|
||||
output.write_text(text + "\n", encoding="utf-8")
|
||||
write_source(info, output.with_suffix(".quelle.json"))
|
||||
print(
|
||||
f"Gespeichert: {output} "
|
||||
f"({len(text.splitlines())} Zeilen, "
|
||||
|
|
|
|||
|
|
@ -13,7 +13,7 @@ import re
|
|||
from aenderungen import compare, read_raw
|
||||
from abschrift_korrigieren import (
|
||||
MODEL, apply_replacements, chat_request, check_ollama, ollama_open, read_glossary,
|
||||
split_chunks,
|
||||
split_chunks, with_video_glossary,
|
||||
)
|
||||
from vtt2txt import split_segments
|
||||
|
||||
|
|
@ -226,7 +226,10 @@ def main():
|
|||
|
||||
raw = text
|
||||
known, replacements = set(), []
|
||||
for glossary in args.glossar:
|
||||
glossaries, automatic = with_video_glossary(args.glossar, args.input.parent)
|
||||
if automatic is not None:
|
||||
print(f"Video-Glossar (automatisch): {automatic}")
|
||||
for glossary in glossaries:
|
||||
_, more = read_glossary(glossary)
|
||||
replacements += more
|
||||
known.update(wrong for wrong, _ in more)
|
||||
|
|
|
|||
|
|
@ -636,13 +636,63 @@ def kreml_glossary(folder):
|
|||
|
||||
def test_main_url_default_names(from_url, tmp_path, monkeypatch):
|
||||
monkeypatch.chdir(tmp_path)
|
||||
work = tmp_path / "abschriften" / "Kanal_Rede_2026-03-28"
|
||||
run_main(monkeypatch, "-u", URL, "-g", kreml_glossary(tmp_path))
|
||||
|
||||
raw = (tmp_path / "Kanal_Rede_2026-03-28.txt").read_text(encoding="utf-8")
|
||||
raw = (work / "Kanal_Rede_2026-03-28.txt").read_text(encoding="utf-8")
|
||||
assert raw == "Der Kremmel ist in Moskau.\nDas ist bekannt.\n"
|
||||
result = (tmp_path / "Kanal_Rede_2026-03-28.korrigiert.txt").read_text(encoding="utf-8")
|
||||
result = (work / "Kanal_Rede_2026-03-28.korrigiert.txt").read_text(encoding="utf-8")
|
||||
assert result == "Der Kreml ist in Moskau. Das ist bekannt.\n"
|
||||
assert (tmp_path / "Kanal_Rede_2026-03-28.aenderungen.html").is_file()
|
||||
assert (work / "Kanal_Rede_2026-03-28.aenderungen.html").is_file()
|
||||
source = json.loads((work / "Kanal_Rede_2026-03-28.quelle.json").read_text(encoding="utf-8"))
|
||||
assert source["titel"] == "Rede" and source["kanal"] == "Kanal"
|
||||
assert source["datum"] == "20260328"
|
||||
assert source["url"] == "https://www.youtube.com/watch?v=abc123"
|
||||
|
||||
|
||||
def video_folder(tmp_path):
|
||||
"""abschriften/rede/ with raw text and its video glossary."""
|
||||
folder = tmp_path / "abschriften" / "rede"
|
||||
folder.mkdir(parents=True)
|
||||
(folder / "rede.txt").write_text("Der Kremmel ist in Moskau.\n", encoding="utf-8")
|
||||
(folder / "rede.glossar.txt").write_text("Kremmel => Kreml\n", encoding="utf-8")
|
||||
return folder
|
||||
|
||||
|
||||
def test_main_uses_video_glossary(tmp_path, monkeypatch, capsys):
|
||||
folder = video_folder(tmp_path)
|
||||
monkeypatch.setattr(ak, "correct_chunk", identity_correct)
|
||||
monkeypatch.setattr(ak, "find_paragraph_starts", every_nth_start(5))
|
||||
run_main(monkeypatch, "-i", folder / "rede.txt")
|
||||
|
||||
result = (folder / "rede.korrigiert.txt").read_text(encoding="utf-8")
|
||||
assert result == "Der Kreml ist in Moskau.\n"
|
||||
assert "rede.glossar.txt (Video-Glossar, automatisch)" in capsys.readouterr().out
|
||||
|
||||
|
||||
def test_main_video_glossary_given_explicitly_counts_once(tmp_path, monkeypatch, capsys):
|
||||
folder = video_folder(tmp_path)
|
||||
shared = tmp_path / "gemeinsam.glossar.txt"
|
||||
shared.write_text("Moskau\n", encoding="utf-8")
|
||||
monkeypatch.setattr(ak, "correct_chunk", identity_correct)
|
||||
monkeypatch.setattr(ak, "find_paragraph_starts", every_nth_start(5))
|
||||
run_main(monkeypatch, "-i", folder / "rede.txt", "-g", shared, "-g", folder / "rede.glossar.txt")
|
||||
|
||||
out = capsys.readouterr().out
|
||||
assert out.count("rede.glossar.txt") == 1
|
||||
assert "automatisch" not in out
|
||||
# Shared glossaries first, as given
|
||||
assert out.index("gemeinsam.glossar.txt") < out.index("rede.glossar.txt")
|
||||
|
||||
|
||||
def test_with_video_glossary(tmp_path):
|
||||
folder = video_folder(tmp_path)
|
||||
shared = tmp_path / "gemeinsam.glossar.txt"
|
||||
assert ak.with_video_glossary([shared], folder) == (
|
||||
[shared, folder / "rede.glossar.txt"], folder / "rede.glossar.txt"
|
||||
)
|
||||
# Without a glossary named like the folder nothing is added
|
||||
assert ak.with_video_glossary([shared], tmp_path) == ([shared], None)
|
||||
|
||||
|
||||
def test_main_url_output_option_and_kept_raw_text(from_url, tmp_path, monkeypatch, capsys):
|
||||
|
|
@ -762,19 +812,21 @@ def test_cli_errors(sample_copy, monkeypatch, capsys, args, message):
|
|||
|
||||
def test_main_url_time_range_default_names(from_url, tmp_path, monkeypatch):
|
||||
monkeypatch.chdir(tmp_path)
|
||||
work = tmp_path / "abschriften" / "Kanal_Rede_2026-03-28"
|
||||
run_main(monkeypatch, "-u", URL, "-z", "0:00:02")
|
||||
|
||||
name = "Kanal_Rede_2026-03-28_ab_00-00-02"
|
||||
raw = (tmp_path / f"{name}.txt").read_text(encoding="utf-8")
|
||||
raw = (work / f"{name}.txt").read_text(encoding="utf-8")
|
||||
assert raw == "[00:00:02 - 00:00:03]\nDas ist bekannt.\n"
|
||||
result = (tmp_path / f"{name}.korrigiert.txt").read_text(encoding="utf-8")
|
||||
result = (work / f"{name}.korrigiert.txt").read_text(encoding="utf-8")
|
||||
assert result == "[00:00:02 - 00:00:03]\n\nDas ist bekannt.\n"
|
||||
|
||||
|
||||
def test_main_url_range_from_beginning(from_url, tmp_path, monkeypatch):
|
||||
monkeypatch.chdir(tmp_path)
|
||||
work = tmp_path / "abschriften" / "Kanal_Rede_2026-03-28"
|
||||
run_main(monkeypatch, "-u", URL, "-z", "-0:00:02")
|
||||
raw = (tmp_path / "Kanal_Rede_2026-03-28_bis_00-00-02.txt").read_text(encoding="utf-8")
|
||||
raw = (work / "Kanal_Rede_2026-03-28_bis_00-00-02.txt").read_text(encoding="utf-8")
|
||||
assert raw == "[00:00:01 - 00:00:02]\nDer Kremmel ist in Moskau.\n"
|
||||
|
||||
|
||||
|
|
@ -926,14 +978,15 @@ def test_main_url_several_ranges(fake_yt_dlp, yt_dlp_installed, tmp_path, monkey
|
|||
monkeypatch.setattr(subprocess, "run", fake_yt_dlp(VIDEO, vtt=MULTI_VTT))
|
||||
sent = chunk_recorder(monkeypatch)
|
||||
monkeypatch.chdir(tmp_path)
|
||||
work = tmp_path / "abschriften" / "Kanal_Rede_2026-03-28"
|
||||
run_main(monkeypatch, "-u", URL, "-z", "0:00:00-0:00:30 0:02", "-g", kreml_glossary(tmp_path))
|
||||
|
||||
name = "Kanal_Rede_2026-03-28_bis_00-00-30+ab_00-02-00"
|
||||
assert (tmp_path / f"{name}.txt").read_text(encoding="utf-8") == (
|
||||
assert (work / f"{name}.txt").read_text(encoding="utf-8") == (
|
||||
"[00:00:01 - 00:00:02]\nDer Kremmel ist in Moskau.\n\n"
|
||||
"[00:02:00 - 00:02:01]\nDas ist bekannt.\n"
|
||||
)
|
||||
assert (tmp_path / f"{name}.korrigiert.txt").read_text(encoding="utf-8") == (
|
||||
assert (work / f"{name}.korrigiert.txt").read_text(encoding="utf-8") == (
|
||||
"[00:00:01 - 00:00:02]\n\nDer Kreml ist in Moskau.\n\n"
|
||||
"[00:02:00 - 00:02:01]\n\nDas ist bekannt.\n"
|
||||
)
|
||||
|
|
|
|||
|
|
@ -321,3 +321,19 @@ def test_report_and_summary_put_word_forms_last(tmp_path, capsys):
|
|||
out = capsys.readouterr().out.split("Zu prüfen: ")[1]
|
||||
assert out.startswith("2 Stellen (davon 1 Wortformen)")
|
||||
assert out.index("Professorin") < out.index("„das“ → „dass“")
|
||||
|
||||
|
||||
def test_cli_uses_video_glossary(tmp_path, monkeypatch, capsys):
|
||||
folder = tmp_path / "rede"
|
||||
folder.mkdir()
|
||||
raw = folder / "rede.txt"
|
||||
raw.write_text("das Kremmel und\nBar heute\n", encoding="utf-8")
|
||||
corrected = folder / "rede.korrigiert.txt"
|
||||
corrected.write_text("Das Kreml und Bahr heute.\n", encoding="utf-8")
|
||||
(folder / "rede.glossar.txt").write_text("Kremmel => Kreml\nEgon Bahr\n", encoding="utf-8")
|
||||
|
||||
monkeypatch.setattr(sys, "argv", ["aenderungen.py", "-a", str(raw), "-b", str(corrected)])
|
||||
ae.main()
|
||||
out = capsys.readouterr().out
|
||||
assert "Video-Glossar (automatisch)" in out
|
||||
assert "Zu prüfen: 0 Stellen" in out
|
||||
|
|
|
|||
|
|
@ -1,4 +1,5 @@
|
|||
"""Tests for extract_text.py; yt-dlp is always replaced by fakes (no network)."""
|
||||
import json
|
||||
import sys
|
||||
|
||||
import pytest
|
||||
|
|
@ -38,14 +39,19 @@ def test_default_language_is_original_audio(fake_yt_dlp, tmp_path, monkeypatch,
|
|||
monkeypatch.chdir(tmp_path)
|
||||
run_main(monkeypatch, "-u", URL)
|
||||
|
||||
text = (tmp_path / "Kanal_Eine_Rede_2026-03-28.txt").read_text(encoding="utf-8")
|
||||
work = tmp_path / "abschriften" / "Kanal_Eine_Rede_2026-03-28"
|
||||
text = (work / "Kanal_Eine_Rede_2026-03-28.txt").read_text(encoding="utf-8")
|
||||
assert text == "Guten Abend.\nSchön, dass Sie da sind.\n"
|
||||
assert "de-orig" in run.calls[1]
|
||||
assert "--write-auto-subs" in run.calls[1]
|
||||
out = capsys.readouterr().out
|
||||
assert "Kanal_Eine_Rede_2026-03-28.txt (2 Zeilen, 7 Wörter)" in out
|
||||
# Only the text file remains, no VTT
|
||||
assert [p.name for p in tmp_path.iterdir()] == ["Kanal_Eine_Rede_2026-03-28.txt"]
|
||||
# Only the text file and where it comes from remain, no VTT
|
||||
assert sorted(p.name for p in work.iterdir()) == [
|
||||
"Kanal_Eine_Rede_2026-03-28.quelle.json", "Kanal_Eine_Rede_2026-03-28.txt",
|
||||
]
|
||||
source = json.loads((work / "Kanal_Eine_Rede_2026-03-28.quelle.json").read_text())
|
||||
assert source["kanal"] == "Kanal" and source["url"].endswith("v=abc123")
|
||||
|
||||
|
||||
def test_default_language_ignores_dubbed_tracks(fake_yt_dlp, tmp_path, monkeypatch):
|
||||
|
|
@ -133,7 +139,8 @@ def test_time_range_default_name(fake_yt_dlp, tmp_path, monkeypatch):
|
|||
run_main(monkeypatch, "-u", URL, "-z", "0:00:02-0:00:03")
|
||||
|
||||
name = "Kanal_Eine_Rede_2026-03-28_00-00-02_bis_00-00-03.txt"
|
||||
assert (tmp_path / name).read_text(encoding="utf-8") == (
|
||||
work = tmp_path / "abschriften" / "Kanal_Eine_Rede_2026-03-28"
|
||||
assert (work / name).read_text(encoding="utf-8") == (
|
||||
"[00:00:02 - 00:00:03]\nSchön, dass Sie da sind.\n"
|
||||
)
|
||||
|
||||
|
|
@ -183,7 +190,8 @@ def test_several_ranges_flow_text_keeps_headers(fake_yt_dlp, tmp_path, monkeypat
|
|||
run_main(monkeypatch, "-u", URL, "-z", "0:00:00-0:00:30 0:02", "-f")
|
||||
|
||||
name = "Kanal_Eine_Rede_2026-03-28_bis_00-00-30+ab_00-02-00.txt"
|
||||
assert (tmp_path / name).read_text(encoding="utf-8") == (
|
||||
work = tmp_path / "abschriften" / "Kanal_Eine_Rede_2026-03-28"
|
||||
assert (work / name).read_text(encoding="utf-8") == (
|
||||
"[00:00:01 - 00:00:02]\nErster Satz der\nRede.\n\n"
|
||||
"[00:02:00 - 00:02:01]\nDritter Satz, der\netwas länger ist.\n"
|
||||
)
|
||||
|
|
|
|||
|
|
@ -160,3 +160,18 @@ def test_time_headers_never_reach_model_or_changes(tmp_path, monkeypatch):
|
|||
draft = out.read_text(encoding="utf-8")
|
||||
assert "# ? Kremmel => Kreml (1×)" in draft
|
||||
assert "00:0" not in draft.replace(str(raw), "")
|
||||
|
||||
|
||||
def test_main_uses_video_glossary(tmp_path, monkeypatch, capsys):
|
||||
folder = tmp_path / "rede"
|
||||
folder.mkdir()
|
||||
raw = folder / "rede.txt"
|
||||
raw.write_text(RAW + "\n", encoding="utf-8")
|
||||
(folder / "rede.glossar.txt").write_text("Kremmel => Kreml\n", encoding="utf-8")
|
||||
model = fake_model(ENTRIES)
|
||||
monkeypatch.setattr(urllib.request, "urlopen", model)
|
||||
run_main(monkeypatch, "-i", raw)
|
||||
|
||||
sent = model.calls[0]["messages"][1]["content"]
|
||||
assert "Kreml" in sent and "Kremmel" not in sent
|
||||
assert "Video-Glossar (automatisch)" in capsys.readouterr().out
|
||||
|
|
|
|||
|
|
@ -140,8 +140,11 @@ def test_yt_dlp_runs_in_temporary_directory(fake_yt_dlp, tmp_path, monkeypatch):
|
|||
|
||||
assert all(cwd is not None and Path(cwd) != tmp_path for cwd in run.cwds)
|
||||
template = run.calls[1][run.calls[1].index("-o") + 1]
|
||||
assert template == f"{tmp_path / 'Rede'}.%(ext)s"
|
||||
assert (tmp_path / "Rede.de-orig.vtt").is_file()
|
||||
# Without -d: working folder abschriften/<Name>/, named after -n
|
||||
work = tmp_path / "abschriften" / "Rede"
|
||||
assert template == f"{work / 'Rede'}.%(ext)s"
|
||||
assert (work / "Rede.de-orig.vtt").is_file()
|
||||
assert (work / "Rede.quelle.json").is_file()
|
||||
|
||||
|
||||
def test_main_prefers_manual_and_default_name(fake_yt_dlp, tmp_path, monkeypatch):
|
||||
|
|
|
|||
|
|
@ -1,6 +1,7 @@
|
|||
#!/usr/bin/env python3
|
||||
"""Lädt die Untertitel eines YouTube-Videos als VTT-Datei herunter (über yt-dlp),
|
||||
als Eingabe für vtt2txt.py bzw. abschrift_korrigieren.py."""
|
||||
from datetime import date
|
||||
from pathlib import Path
|
||||
import argparse
|
||||
import json
|
||||
|
|
@ -14,6 +15,7 @@ YT_DLP = "yt-dlp"
|
|||
NAME_WORD = re.compile(r"\w+(?:[-.]\w+)*") # keeps "Krone-Schmalz", "28.03.2026"
|
||||
MAX_CHANNEL_CHARS = 40
|
||||
MAX_TITLE_CHARS = 60
|
||||
WORK_DIR = Path("abschriften") # ein Arbeitsordner je Video: abschriften/<Name>/
|
||||
|
||||
|
||||
def run_yt_dlp(arguments):
|
||||
|
|
@ -112,6 +114,31 @@ def default_name(info):
|
|||
return "_".join(p for p in parts if p) or info.get("id") or "video"
|
||||
|
||||
|
||||
def work_folder(name):
|
||||
"""Working folder of one video: abschriften/<Name>/"""
|
||||
return WORK_DIR / name
|
||||
|
||||
|
||||
def write_source(info, path):
|
||||
"""Store where a transcript comes from (<Name>.quelle.json), for the
|
||||
header of the finished result."""
|
||||
url = info.get("webpage_url") or (
|
||||
f"https://www.youtube.com/watch?v={info['id']}" if info.get("id") else None
|
||||
)
|
||||
source = {
|
||||
"titel": info.get("title"),
|
||||
"kanal": info.get("channel") or info.get("uploader"),
|
||||
"datum": info.get("upload_date"),
|
||||
"url": url,
|
||||
"id": info.get("id"),
|
||||
"dauer": info.get("duration"),
|
||||
"abgerufen": date.today().isoformat(),
|
||||
}
|
||||
path.parent.mkdir(parents=True, exist_ok=True)
|
||||
path.write_text(json.dumps(source, ensure_ascii=False, indent=2) + "\n", encoding="utf-8")
|
||||
return path
|
||||
|
||||
|
||||
def available_tracks(info):
|
||||
"""Manual tracks and the automatic ones usable for this program."""
|
||||
manual = sorted((info.get("subtitles") or {}).keys())
|
||||
|
|
@ -164,8 +191,8 @@ def parse_args():
|
|||
"-l", "--lang", default="de", help="Sprache der Untertitel (Standard: de)"
|
||||
)
|
||||
parser.add_argument(
|
||||
"-d", "--dir", type=Path, default=Path("."),
|
||||
help="Zielverzeichnis (Standard: aktuelles Verzeichnis)",
|
||||
"-d", "--dir", type=Path,
|
||||
help="Zielverzeichnis (Standard: Arbeitsordner abschriften/<Name>/)",
|
||||
)
|
||||
args = parser.parse_args()
|
||||
|
||||
|
|
@ -189,9 +216,12 @@ def main():
|
|||
kind = "automatische" if is_auto else "manuelle"
|
||||
print(f"Lade {kind} Untertitel „{track}“ …", flush=True)
|
||||
|
||||
args.dir.mkdir(parents=True, exist_ok=True)
|
||||
target_base = args.dir / (args.name or default_name(info))
|
||||
name = args.name or default_name(info)
|
||||
folder = args.dir or work_folder(name)
|
||||
folder.mkdir(parents=True, exist_ok=True)
|
||||
target_base = folder / name
|
||||
download(args.url, track, is_auto, target_base)
|
||||
write_source(info, folder / f"{name}.quelle.json")
|
||||
|
||||
target = Path(f"{target_base}.{track}.vtt")
|
||||
if not target.is_file():
|
||||
|
|
|
|||
Loading…
Reference in a new issue