feat(tools): scripts/add_pronunciation.py zum bequemen Pflegen des Aussprache-Lexikons
Fuegt 'wort:aussprache' in config/pronunciation.<lang>.yaml ein (Default Sektion terms). Erhaelt Kommentare/Struktur, aktualisiert vorhandene Eintraege statt zu duplizieren, validiert das YAML vor dem Schreiben. Optional --verify zeigt die espeak-Phoneme vorher/nachher und warnt, wenn die Umschreibung nichts aendert. Doku-Hinweis in BEDIENUNGSANLEITUNG ergaenzt. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
This commit is contained in:
parent
449a62a88e
commit
37b14773f2
2 changed files with 149 additions and 0 deletions
142
scripts/add_pronunciation.py
Normal file
142
scripts/add_pronunciation.py
Normal file
|
|
@ -0,0 +1,142 @@
|
|||
#!/usr/bin/env python3
|
||||
"""Fügt ein Wort + Aussprache ins Aussprache-Lexikon ein.
|
||||
|
||||
Das Lexikon (config/pronunciation.<lang>.yaml) wird vor dem lokalen TTS (piper)
|
||||
angewandt: der geschriebene Begriff wird durch eine "sprechbare" Umschreibung
|
||||
ersetzt (z. B. langer Vokal per eingefügtem 'h').
|
||||
|
||||
Beispiele:
|
||||
python scripts/add_pronunciation.py strömen ströhmen
|
||||
python scripts/add_pronunciation.py "strömen:ströhmen"
|
||||
python scripts/add_pronunciation.py CPU "Ze Pe U" --section terms
|
||||
python scripts/add_pronunciation.py kWh "Kilowattstunden" --section units
|
||||
python scripts/add_pronunciation.py strömen ströhmen --verify # espeak-Phoneme zeigen
|
||||
|
||||
Hinweis: Damit der laufende Server die Änderung nutzt, einmal neu starten.
|
||||
"""
|
||||
|
||||
from __future__ import annotations
|
||||
|
||||
import argparse
|
||||
import shutil
|
||||
import subprocess
|
||||
import sys
|
||||
from pathlib import Path
|
||||
|
||||
try:
|
||||
import yaml
|
||||
except ModuleNotFoundError:
|
||||
sys.exit("Fehlt: Python-Paket 'PyYAML' (pip install pyyaml).")
|
||||
|
||||
BASE_DIR = Path(__file__).resolve().parent.parent
|
||||
SECTIONS = ("terms", "abbreviations", "units")
|
||||
|
||||
|
||||
def _quote(s: str) -> str:
|
||||
"""YAML-sicher als doppelt-gequoteter String."""
|
||||
return '"' + s.replace("\\", "\\\\").replace('"', '\\"') + '"'
|
||||
|
||||
|
||||
def _espeak(word: str, lang: str) -> str | None:
|
||||
if not shutil.which("espeak-ng"):
|
||||
return None
|
||||
try:
|
||||
out = subprocess.run(
|
||||
["espeak-ng", "-v", lang, "-q", "-x", word],
|
||||
capture_output=True, text=True, timeout=10,
|
||||
)
|
||||
return out.stdout.strip() or None
|
||||
except (subprocess.SubprocessError, OSError):
|
||||
return None
|
||||
|
||||
|
||||
def _entry_line_index(lines: list[str], section: str, word: str) -> int | None:
|
||||
"""Index einer vorhandenen, aktiven Zeile ' "word": ...' in der Sektion."""
|
||||
in_section = False
|
||||
target = word.lower()
|
||||
for i, line in enumerate(lines):
|
||||
stripped = line.strip()
|
||||
if not line.startswith(" ") and stripped.rstrip(":") in SECTIONS:
|
||||
in_section = stripped.rstrip(":") == section
|
||||
continue
|
||||
if in_section and stripped and not stripped.startswith("#"):
|
||||
key = stripped.split(":", 1)[0].strip().strip('"').strip("'")
|
||||
if key.lower() == target:
|
||||
return i
|
||||
return None
|
||||
|
||||
|
||||
def _section_header_index(lines: list[str], section: str) -> int | None:
|
||||
for i, line in enumerate(lines):
|
||||
if not line.startswith(" ") and line.strip().rstrip(":") == section \
|
||||
and line.rstrip().endswith(":"):
|
||||
return i
|
||||
return None
|
||||
|
||||
|
||||
def upsert(path: Path, section: str, word: str, pron: str) -> str:
|
||||
text = path.read_text(encoding="utf-8") if path.exists() else ""
|
||||
lines = text.splitlines()
|
||||
new_line = f" {_quote(word)}: {_quote(pron)}"
|
||||
|
||||
existing = _entry_line_index(lines, section, word)
|
||||
if existing is not None:
|
||||
action = "aktualisiert"
|
||||
lines[existing] = new_line
|
||||
else:
|
||||
action = "hinzugefügt"
|
||||
header = _section_header_index(lines, section)
|
||||
if header is None: # Sektion fehlt -> am Ende anlegen
|
||||
if lines and lines[-1].strip():
|
||||
lines.append("")
|
||||
lines.append(f"{section}:")
|
||||
lines.append(new_line)
|
||||
else:
|
||||
lines.insert(header + 1, new_line)
|
||||
|
||||
result = "\n".join(lines) + "\n"
|
||||
# Vor dem Schreiben validieren -> niemals kaputtes YAML hinterlassen.
|
||||
parsed = yaml.safe_load(result) or {}
|
||||
if (parsed.get(section) or {}).get(word) != pron:
|
||||
raise SystemExit("Abbruch: Eintrag nach dem Schreiben nicht wie erwartet (YAML-Problem).")
|
||||
path.write_text(result, encoding="utf-8")
|
||||
return action
|
||||
|
||||
|
||||
def main() -> None:
|
||||
p = argparse.ArgumentParser(description="Wort + Aussprache ins Lexikon eintragen.")
|
||||
p.add_argument("word", help="Wort wie geschrieben (oder 'wort:aussprache').")
|
||||
p.add_argument("pron", nargs="?", help="sprechbare Umschreibung.")
|
||||
p.add_argument("--section", choices=SECTIONS, default="terms")
|
||||
p.add_argument("--lang", default="de")
|
||||
p.add_argument("--file", default=None, help="Lexikon-Pfad (Default: config/pronunciation.<lang>.yaml)")
|
||||
p.add_argument("--verify", action="store_true", help="espeak-Phoneme vorher/nachher zeigen")
|
||||
args = p.parse_args()
|
||||
|
||||
word, pron = args.word, args.pron
|
||||
if pron is None:
|
||||
if ":" not in word:
|
||||
p.error("Bitte 'wort aussprache' oder 'wort:aussprache' angeben.")
|
||||
word, pron = (s.strip() for s in word.split(":", 1))
|
||||
if not word or not pron:
|
||||
p.error("Wort und Aussprache dürfen nicht leer sein.")
|
||||
|
||||
path = Path(args.file) if args.file else BASE_DIR / "config" / f"pronunciation.{args.lang}.yaml"
|
||||
|
||||
if args.verify:
|
||||
before, after = _espeak(word, args.lang), _espeak(pron, args.lang)
|
||||
if before is None:
|
||||
print(" (espeak-ng nicht gefunden – Verifikation übersprungen)")
|
||||
else:
|
||||
print(f" espeak {word!r:14} -> {before}")
|
||||
print(f" espeak {pron!r:14} -> {after}")
|
||||
if before == after:
|
||||
print(" ⚠ Phoneme identisch – die Umschreibung ändert die Aussprache NICHT.")
|
||||
|
||||
action = upsert(path, args.section, word, pron)
|
||||
print(f"✓ {action}: [{args.section}] {word!r} -> {pron!r} ({path})")
|
||||
print(" Hinweis: laufenden Server neu starten, damit die Änderung greift.")
|
||||
|
||||
|
||||
if __name__ == "__main__":
|
||||
main()
|
||||
Loading…
Add table
Add a link
Reference in a new issue