mirror of
https://github.com/AvengeMedia/DankMaterialShell.git
synced 2026-08-01 19:18:28 -04:00
bbedward
3c0f2cbc48
- Delete ~160-ish useless terms - Add context to more terms - Add a mechanism to duplicate the same terms with different contexts - sync
69 lines
2.3 KiB
Python
69 lines
2.3 KiB
Python
#!/usr/bin/env python3
|
|
"""Block near-variant translation terms (case/trailing-punctuation duplicates).
|
|
|
|
Unlike a term freeze, new terms are always allowed. This only fails when two
|
|
terms collapse to the same string after lowercasing and stripping trailing
|
|
".:?! " punctuation - e.g. "Signal" vs "Signal:", "Reset to Default?" vs
|
|
"Reset to default". Build punctuation outside tr() instead:
|
|
I18n.tr("Signal") + ":"
|
|
|
|
"Term" vs "Term..." pairs are NOT flagged: a trailing ellipsis is meaningful
|
|
(progress states, opens-a-dialog affordances).
|
|
"""
|
|
import sys
|
|
from collections import defaultdict
|
|
|
|
from extract_translations import extract_qstr_strings
|
|
from pathlib import Path
|
|
|
|
ROOT_DIR = Path(__file__).parent.parent
|
|
|
|
# Intentional pairs that survive normalization on purpose.
|
|
ALLOWED = [
|
|
{"PIN", "Pin"}, # WPS PIN acronym vs the verb "pin"
|
|
{"Device", "device"}, # label vs inline generic-noun fallback
|
|
{"Until %1", "until %1"}, # sentence-initial vs mid-sentence position
|
|
]
|
|
|
|
|
|
def normalize(term):
|
|
t = term.replace("…", "...")
|
|
ellipsis = t.rstrip().endswith("...")
|
|
t = t.lower().rstrip(".:?! ")
|
|
return t + "..." if ellipsis else t
|
|
|
|
|
|
def main():
|
|
translations = extract_qstr_strings(ROOT_DIR)
|
|
groups = defaultdict(set)
|
|
for term in translations:
|
|
groups[normalize(term)].add(term)
|
|
|
|
failures = []
|
|
for variants in groups.values():
|
|
if len(variants) < 2 or any(variants == a for a in ALLOWED):
|
|
continue
|
|
# term vs term+"..." is allowed; flag only same-ellipsis-class variants
|
|
plain = {v for v in variants if not v.replace("…", "...").rstrip().endswith("...")}
|
|
dotted = variants - plain
|
|
for cls in (plain, dotted):
|
|
if len(cls) > 1:
|
|
failures.append(sorted(cls))
|
|
|
|
if not failures:
|
|
print(f"No term variants ({len(translations)} terms checked)")
|
|
return 0
|
|
|
|
print("Near-variant terms found - reuse one exact term (or add to ALLOWED "
|
|
"in check_term_variants.py if genuinely intentional):", file=sys.stderr)
|
|
for variants in sorted(failures):
|
|
print(f" {variants}", file=sys.stderr)
|
|
for v in variants:
|
|
for occ in translations[v]["occurrences"][:3]:
|
|
print(f" {v!r}: {occ['file']}:{occ['line']}", file=sys.stderr)
|
|
return 1
|
|
|
|
|
|
if __name__ == "__main__":
|
|
sys.exit(main())
|