dazed/DazedTL
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

substring check

Browse source
This commit is contained in:
dazedanon 2026-03-04 23:55:45 -06:00
parent b6df17d30b
commit 973b3322a9
2 changed files with 27 additions and 4 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

4
modules/rpgmakermvmz.py
View file

@ -2536,7 +2536,7 @@ def searchCodes(page, pbar, jobList, filename):
# Patterns: (regex, multiline) - multiline=True means it spans 355 + following 655 codes
patterns = {
# "テキスト-": (r"テキスト-(.+)", False),
"テキスト-": (r"テキスト-(.+)", False),
# "=": (r'=\s?(.*)",', False),
# "var text": (r"var\stext\d+\s=\s\"(.+)\"", False),
# "logtxt = ": (r"logtxt\s=\s'(.+)'", False),
@ -2554,7 +2554,7 @@ def searchCodes(page, pbar, jobList, filename):
# "this.Menu_SexTxtSet(": (r'"(.+)"', True),
# "Rn_RsltTxtArr": (r'"(.+)"', True),
# "_章切り替えStart": (r'_章切り替えStart\(\s*\\?"\s?,?.+?\\?"\s?,?\s?\\?"(.+?)\\?"', False),
"SkillLogAdd": (r'SkillLogAdd\((?:.+?\+\s*)?\\?"(?:\\\\+[A-Za-z]\[\d+\])?(.+?)\\?"', False),
# "SkillLogAdd": (r'SkillLogAdd\((?:.+?\+\s*)?\\?"(?:\\\\+[A-Za-z]\[\d+\])?(.+?)\\?"', False),
# "MobNameSet": (r'MobNameSet\(\\?"(.+?)\\?"\)', False),
# "AddAddress": (r'AddAddress\(\d+,\s*\\?"(.+?)\\?"', False),

27
util/translation.py
View file

@ -574,6 +574,29 @@ def parseVocabWithCategories(vocabText):
return pairs
def _japanese_term_in_text(term, text):
"""
Check if a Japanese term appears in text as a standalone word, not as a
substring of a longer run of the same script (katakana/hiragana/kanji).
E.g. 'キス' will NOT match inside 'テキスト' because both neighbours are katakana.
Falls back to plain substring check for non-Japanese or mixed terms.
"""
if term not in text:
return False
KATAKANA = r'ァ-ヴーｦ-ﾟ'
HIRAGANA = r'ぁ-ゔ'
KANJI = r'一-龠'
if re.search(rf'[{KATAKANA}]', term) and not re.search(rf'[{HIRAGANA}{KANJI}]', term):
pattern = rf'(?<![{KATAKANA}]){re.escape(term)}(?![{KATAKANA}])'
elif re.search(rf'[{HIRAGANA}]', term) and not re.search(rf'[{KATAKANA}{KANJI}]', term):
pattern = rf'(?<![{HIRAGANA}]){re.escape(term)}(?![{HIRAGANA}])'
elif re.search(rf'[{KANJI}]', term) and not re.search(rf'[{KATAKANA}{HIRAGANA}]', term):
pattern = rf'(?<![{KANJI}]){re.escape(term)}(?![{KANJI}])'
else:
return True # mixed-script term: plain substring match already confirmed above
return bool(re.search(pattern, text))
def buildMatchedVocabText(vocabPairs, subbedText, history=None):
"""Build formatted vocabulary text with matched terms organized by category."""
matchedCategories = {}
@ -593,11 +616,11 @@ def buildMatchedVocabText(vocabPairs, subbedText, history=None):
if isinstance(term, tuple):
# Check both Japanese and English terms
japanese_term, english_term = term
if japanese_term in textToSearch or english_term in textToSearch:
if _japanese_term_in_text(japanese_term, textToSearch) or english_term in textToSearch:
term_found = True
else:
# Single term check
if term in textToSearch:
if _japanese_term_in_text(term, textToSearch) if re.search(r'[一-龠ぁ-ゔァ-ヴーｦ-ﾟ｡-ﾟ]', str(term)) else term in textToSearch:
term_found = True
# Always include "# Game Characters" category and all its terms regardless of matches