dschlueter/youtube_abschrift_korrigieren
Watch
1
0
Fork
You've already forked youtube_abschrift_korrigieren
 0
Code Issues Pull requests Projects Releases Packages Wiki Activity

test: echte Reden liegen lokal in testdaten/<rede>/ mit ihrem Glossar

Browse source 
testdaten/ enthält je Rede einen Ordner mit .vtt, .txt und dem Glossar
der Rede, keine Ausgaben mehr; gemeinsame Glossare bleiben in glossare/.
Die lokalen Rede-Tests lesen von dort.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
This commit is contained in:
Dieter Schlüter 2026-10-11 23:27:11 +02:00
commit 775560a64c
2 changed files with 10 additions and 8 deletions
Download patch file Download diff file Expand all files Collapse all files

14
tests/test_abschrift_korrigieren.py
View file

@ -14,7 +14,8 @@ REPO = Path(__file__).resolve().parent.parent
# Own, made-up sample speech; real speeches are only kept locally in testdaten/ # Own, made-up sample speech; real speeches are only kept locally in testdaten/
SAMPLE = REPO / "tests" / "daten" / "beispiel.de.txt" SAMPLE = REPO / "tests" / "daten" / "beispiel.de.txt"
GLOSSARY = REPO / "tests" / "daten" / "beispiel.glossar.txt" GLOSSARY = REPO / "tests" / "daten" / "beispiel.glossar.txt"
# Real glossaries and speeches are only kept locally (see .gitignore) # Real glossaries and speeches are only kept locally (see .gitignore): shared
# glossaries in glossare/, each speech with its glossary in testdaten/<rede>/
GLOSSARIES = REPO / "glossare" GLOSSARIES = REPO / "glossare"
SHARED_GLOSSARY = GLOSSARIES / "krone-schmalz.glossar.txt" SHARED_GLOSSARY = GLOSSARIES / "krone-schmalz.glossar.txt"
SPEECHES = REPO / "testdaten" SPEECHES = REPO / "testdaten"
@ -127,17 +128,18 @@ def test_build_system_appends_terms():
assert "keep only the surname" in ak.SYSTEM_EN assert "keep only the surname" in ak.SYSTEM_EN
# testdaten/<rede>/: raw text and the glossary of that speech
SAMPLE_GLOSSARIES = [ SAMPLE_GLOSSARIES = [
("ramstein.glossar.txt", "ramstein.de.txt"), ("ramstein/ramstein.glossar.txt", "ramstein/ramstein.de.txt"),
("magdeburg.glossar.txt", "magdeburg.de-orig.txt"), ("magdeburg/magdeburg.glossar.txt", "magdeburg/magdeburg.de-orig.txt"),
("lahnstein.glossar.txt", "Lahnstein.de-orig.txt"), ("lahnstein/lahnstein.glossar.txt", "lahnstein/Lahnstein.de-orig.txt"),
] ]
# Real speeches and their glossaries are not in the repository; without them these skip # Real speeches and their glossaries are not in the repository; without them these skip
needs_speeches = pytest.mark.skipif( needs_speeches = pytest.mark.skipif(
not all( not all(
(SPEECHES / sample).is_file() and (GLOSSARIES / glossary).is_file() (SPEECHES / sample).is_file() and (SPEECHES / glossary).is_file()
for glossary, sample in SAMPLE_GLOSSARIES for glossary, sample in SAMPLE_GLOSSARIES
) or not SHARED_GLOSSARY.is_file(), ) or not SHARED_GLOSSARY.is_file(),
reason="echte Reden (testdaten/) und ihre Glossare nur lokal", reason="echte Reden (testdaten/) und ihre Glossare nur lokal",
@ -157,7 +159,7 @@ def test_sample_glossary_replacements_all_match(capsys, glossary, sample):
text = flat((SPEECHES / sample).read_text(encoding="utf-8")) text = flat((SPEECHES / sample).read_text(encoding="utf-8"))
text = ak.apply_replacements(text, ak.read_glossary(SHARED_GLOSSARY)[1]) text = ak.apply_replacements(text, ak.read_glossary(SHARED_GLOSSARY)[1])
capsys.readouterr() capsys.readouterr()
ak.apply_replacements(text, ak.read_glossary(GLOSSARIES / glossary)[1]) ak.apply_replacements(text, ak.read_glossary(SPEECHES / glossary)[1])
assert "ohne Treffer" not in capsys.readouterr().out assert "ohne Treffer" not in capsys.readouterr().out

4
tests/test_vtt2txt.py
View file

@ -124,8 +124,8 @@ def speech(name):
@pytest.mark.parametrize("vtt_path", [ @pytest.mark.parametrize("vtt_path", [
SAMPLE_VTT, SAMPLE_VTT,
speech("ramstein.de.vtt"), speech("magdeburg.de-orig.vtt"), speech("ramstein/ramstein.de.vtt"), speech("magdeburg/magdeburg.de-orig.vtt"),
speech("Lahnstein.de-orig.vtt"), speech("lahnstein/Lahnstein.de-orig.vtt"),
]) ])
def test_sample_vtt_gives_sample_txt(vtt_path): def test_sample_vtt_gives_sample_txt(vtt_path):
vtt = vtt_path.read_text(encoding="utf-8-sig") vtt = vtt_path.read_text(encoding="utf-8-sig")