From 1ed64f2e56a3a8f26b9d59b23ed6e37f5cfe1114 Mon Sep 17 00:00:00 2001 From: DazedAnon Date: Fri, 27 Sep 2024 11:01:39 -0500 Subject: [PATCH] Changes to wolf and MV --- modules/images.py | 2 +- modules/rpgmakermvmz.py | 110 ++++++++++++++++++++------------------- modules/wolf.py | 111 +++++++++++++++++++++------------------- vocab.txt | 55 ++++++++++++++++++++ 4 files changed, 169 insertions(+), 109 deletions(-) diff --git a/modules/images.py b/modules/images.py index c0a82fa..6d8864a 100644 --- a/modules/images.py +++ b/modules/images.py @@ -192,7 +192,7 @@ def getImageDimensions(file_path): def processImagesDir(directory_path, imageList): for file_name in os.listdir(directory_path): # .png and Japanese - if '.png' in file_name and re.search(r'[一-龠ぁ-ゔァ-ヴー]+', file_name) and file_name.replace('.png', '') in VOCAB: + if '.png' in file_name and file_name.replace('.png', '') in VOCAB: file_path = os.path.join(directory_path, file_name) if os.path.isfile(file_path): # Check if the file is an image diff --git a/modules/rpgmakermvmz.py b/modules/rpgmakermvmz.py index cacb641..9d481b4 100644 --- a/modules/rpgmakermvmz.py +++ b/modules/rpgmakermvmz.py @@ -60,11 +60,11 @@ LEAVE = False # Dialogue / Scroll CODE401 = True -CODE405 = False +CODE405 = True CODE408 = False # Choices -CODE102 = False +CODE102 = True # Variables CODE122 = False @@ -488,7 +488,10 @@ def searchNames(data, pbar, context): tokensResponse = translateNote(data[i], r'') totalTokens[0] += tokensResponse[0] totalTokens[1] += tokensResponse[1] - + if 'PE拡張' in data[i]['note']: + tokensResponse = translateNote(data[i], r'') + totalTokens[0] += tokensResponse[0] + totalTokens[1] += tokensResponse[1] i += 1 else: batchFull = True @@ -611,7 +614,9 @@ def searchNames(data, pbar, context): else: # Get Text if data[j]['name'] != '': - data[j]['name'] = translatedNameBatch[0] + with open('translations.txt', 'a', encoding='utf-8') as file: + file.write(f'{data[j]['name']} ({translatedNameBatch[0]})\n') + data[j]['name'] = translatedNameBatch[0] translatedNameBatch.pop(0) if data[j]['nickname'] != '': data[j]['nickname'] = translatedNicknameBatch[0] @@ -644,26 +649,29 @@ def searchNames(data, pbar, context): # Set Data if len(nameList) == len(translatedNameBatch): j = k - while j < i: - # Empty Data - if data[j] is None or data[j]['name'] == "": - j += 1 - continue - else: - # Get Text - data[j]['name'] = translatedNameBatch[0] - if 'description' in data[j]: - data[j]['description'] = textwrap.fill(translatedDescriptionBatch[0], LISTWIDTH) - translatedNameBatch.pop(0) - translatedDescriptionBatch.pop(0) + with open('translations.txt', 'a', encoding='utf-8') as file: + file.write(f'# Items\n') + while j < i: + # Empty Data + if data[j] is None or data[j]['name'] == "": + j += 1 + continue + else: + # Get Text + file.write(f'{data[j]['name']} ({translatedNameBatch[0]})\n') + data[j]['name'] = translatedNameBatch[0] + if 'description' in data[j]: + data[j]['description'] = textwrap.fill(translatedDescriptionBatch[0], LISTWIDTH) + translatedNameBatch.pop(0) + translatedDescriptionBatch.pop(0) - # If Batch is empty. Move on. - if len(translatedNameBatch) == 0: - nameList.clear() - descriptionList.clear() - batchFull = False - filling = False - j += 1 + # If Batch is empty. Move on. + if len(translatedNameBatch) == 0: + nameList.clear() + descriptionList.clear() + batchFull = False + filling = False + j += 1 else: mismatch = True if context in ['Enemies', 'Classes', 'MapInfos']: @@ -720,14 +728,12 @@ def searchCodes(page, pbar, jobList, filename): list355655 = [] list108 = [] setData = False - currentGroup = [] textHistory = [] match = [] totalTokens = [0, 0] translatedText = '' speaker = '' speakerID = None - nametag = '' syncIndex = 0 CLFlag = False maxHistory = MAXHISTORY @@ -761,6 +767,10 @@ def searchCodes(page, pbar, jobList, filename): if len(codeList) <= i: break + # Declare Varss + currentGroup = [] + nametag = '' + ## Event Code: 401 Show Text if 'code' in codeList[i] and codeList[i]['code'] in [401, 405, -1] and (CODE401 or CODE405): # Save Code and starting index (j) @@ -810,8 +820,15 @@ def searchCodes(page, pbar, jobList, filename): if len(speakerList) == 0: speakerList = re.findall(r'^[\\]+[cC]\[[\d]+\](.+?)[\\]+[Cc]\[[\d]\]\\?\\?$', jaString) - # None + # First Line Speakers if len(speakerList) == 0 and FIRSTLINESPEAKERS is True: + # Remove any RPGMaker Code at start + ffMatch = re.search(r'^(\s*[\\]+[aAbBdDeEfFgGhHiIjJlLmMoOpPqQrRsStTuUvVwWxXyYzZ]+\[[\w\d\[\]\\]+\])', jaString) + if ffMatch != None: + jaString = jaString.replace(ffMatch.group(0), '') + nametag += ffMatch.group(0) + + # Test Speaker if len(jaString) < 40 \ and 'code' in codeList[i+1] \ and codeList[i+1]['code'] in [401, 405, -1] \ @@ -828,7 +845,8 @@ def searchCodes(page, pbar, jobList, filename): totalTokens[1] += response[1][1] # Set Data - codeList[i]['parameters'][0] = jaString.replace(speakerList[0], speaker) + codeList[i]['parameters'][0] = nametag + jaString.replace(speakerList[0], speaker) + nametag = '' # Iterate to next string i += 1 @@ -861,7 +879,8 @@ def searchCodes(page, pbar, jobList, filename): # Format String if len(currentGroup) > 0: - finalJAString = ' '.join(currentGroup).replace('？', '?') + finalJAString = '' + finalJAString = ''.join(currentGroup).replace('？', '?') oldjaString = finalJAString # Check if Empty @@ -956,7 +975,7 @@ def searchCodes(page, pbar, jobList, filename): finalJAString = finalJAString.replace(match, '') # Remove any RPGMaker Code at start - ffMatch = re.search(r'^([.\\]+[aAbBcCdDeEfFgGhHiIjJlLmMoOpPqQrRsStTuUvVwWxXyYzZ]+\[.+?\]\]?)+', finalJAString) + ffMatch = re.search(r'^(\s*[\\]+[aAbBdDeEfFgGhHiIjJlLmMoOpPqQrRsStTuUvVwWxXyYzZ]+\[[\w\d\[\]\\]+\])', finalJAString) if ffMatch != None: finalJAString = finalJAString.replace(ffMatch.group(0), '') nametag += ffMatch.group(0) @@ -997,6 +1016,7 @@ def searchCodes(page, pbar, jobList, filename): list401.append(speaker) speaker = '' match = [] + nametag = '' currentGroup = [] syncIndex = i + 1 @@ -2136,20 +2156,6 @@ def batchList(input_list, batch_size): return [input_list[i:i + batch_size] for i in range(0, len(input_list), batch_size)] def createContext(fullPromptFlag, subbedT, format): - characters = 'Game Characters:\n\ -圭太 (Keita) - Male\n\ -涼香 (Ryoka) - Female\n\ -咲 (Saki) - Female\n\ -大介 (Daisuke) - Male\n\ -浮浪者 (Vagrant) - Male\n\ -まさる (Masaru) - Male\n\ -ノブオ (Nobuo) - Male\n\ -安井 (Yasui) - Male\n\ -大山 (Oyama) - Male\n\ -関口 (Sekiguchi) - Male\n\ -黒崎 (Kurosaki) - Male\n\ -' - system = PROMPT + VOCAB if fullPromptFlag else \ f"\ You are an expert Eroge Game translator who translates Japanese text to {LANGUAGE}.\n\ @@ -2167,15 +2173,12 @@ Output ONLY the {LANGUAGE} translation in the following format: `Translation: <{ user = f'```json\n{subbedT}\n```' else: user = subbedT - return characters, system, user + return system, user -def translateText(characters, system, user, history, penalty, format, model=MODEL): +def translateText(system, user, history, penalty, format, model=MODEL): # Prompt msg = [{"role": "system", "content": system}] - # Characters - msg.append({"role": "system", "content": characters}) - # History if isinstance(history, list): msg.extend([{"role": "system", "content": h} for h in history]) @@ -2250,7 +2253,7 @@ def extractTranslation(translatedTextList, is_list): return None -def countTokens(characters, system, user, history): +def countTokens(system, user, history): inputTotalTokens = 0 outputTotalTokens = 0 enc = tiktoken.encoding_for_model('gpt-4') @@ -2262,7 +2265,6 @@ def countTokens(characters, system, user, history): else: inputTotalTokens += len(enc.encode(history)) inputTotalTokens += len(enc.encode(system)) - inputTotalTokens += len(enc.encode(characters)) inputTotalTokens += len(enc.encode(user)) # Output @@ -2306,17 +2308,17 @@ def translateGPT(text, history, fullPromptFlag): continue # Create Message - characters, system, user = createContext(fullPromptFlag, subbedT, format) + system, user = createContext(fullPromptFlag, subbedT, format) # Calculate Estimate if ESTIMATE: - estimate = countTokens(characters, system, user, history) + estimate = countTokens(system, user, history) totalTokens[0] += estimate[0] totalTokens[1] += estimate[1] continue # Translating - response = translateText(characters, system, user, history, 0.05, format) + response = translateText(system, user, history, 0.05, format) translatedText = response.choices[0].message.content totalTokens[0] += response.usage.prompt_tokens totalTokens[1] += response.usage.completion_tokens @@ -2327,7 +2329,7 @@ def translateGPT(text, history, fullPromptFlag): extractedTranslations = extractTranslation(translatedText, True) if extractedTranslations == None or len(tItem) != len(extractedTranslations): # Mismatch. Try Again - response = translateText(characters, system, user, history, 0.05, format, 'gpt-4o') + response = translateText(system, user, history, 0.05, format, 'gpt-4o') translatedText = response.choices[0].message.content totalTokens[0] += response.usage.prompt_tokens totalTokens[1] += response.usage.completion_tokens diff --git a/modules/wolf.py b/modules/wolf.py index e230d33..c1c7fe0 100644 --- a/modules/wolf.py +++ b/modules/wolf.py @@ -62,7 +62,7 @@ FILENAME = None # Dialogue / Scroll CODE101 = False CODE102 = False -CODE122 = False +CODE122 = True # Other CODE210 = False @@ -468,9 +468,10 @@ def searchCodes(events, pbar, jobList, filename): codeList[i]['stringArgs'][0] = translatedText ### Event Code: 300 Common Events - if codeList[i]['code'] == 300 and CODE300 == True and len(codeList[i]['stringArgs']) > 0: + if codeList[i]['code'] == 300 and CODE300 == True and 'stringArgs' in codeList[i] and len(codeList[i]['stringArgs']) > 1: # Choices - if codeList[i]['stringArgs'][0] == "[共]汎用ウィンドウ生成": + if codeList[i]['stringArgs'][0] == "[共]汎用ウィンドウ生成" \ + or codeList[i]['stringArgs'][0] == "[共]選択生成": # Grab String choiceList = codeList[i]['stringArgs'][1].split('\r\n') @@ -502,54 +503,56 @@ def searchCodes(events, pbar, jobList, filename): elif codeList[i]['stringArgs'][0] == "Hメッセージ" \ or codeList[i]['stringArgs'][0] == 'Hしらべる' \ or codeList[i]['stringArgs'][0] == '[共]Hメッセージ+' \ - or codeList[i]['stringArgs'][0] == '[共]ポップアップ表示' \ + or codeList[i]['stringArgs'][0] == 'd[共]ポップアップ表示' \ or codeList[i]['stringArgs'][0] == 'm_謎冒頭' \ - or (codeList[i]['codeStr'] == 'SetString' and '\n' in codeList[i]['stringArgs'][0])\ + or (codeList[i]['codeStr'] == 'SetString' and ('「' in codeList[i]['stringArgs'][0] or '*' in codeList[i]['stringArgs'][0]))\ or codeList[i]['stringArgs'][0] == '[移]サウンドノベル': - if len(codeList[i]['stringArgs']) > 1: + cleanedList = None + if len(codeList[i]['stringArgs']) > 1 and not re.search(r'^[\\]+cself\[\d+\]$', codeList[i]['stringArgs'][1]): cleanedList = formatDramon(codeList[i]['stringArgs'][1]) - else: + elif codeList[i]['code'] == 122: cleanedList = formatDramon(codeList[i]['stringArgs'][0]) - fontSize = 24 - translatedText = '' - - for str in cleanedList: - # Pass 1 - if not setData: - if all(x not in str for x in ['_', '@', '>', '/']) and str != '\r\n': - # Remove Textwrap and Font and Add to list - str = str.replace('\r\n', ' ') - str = str.replace(f'\\f[{fontSize}]', '') - list300.append(str) + if cleanedList: + fontSize = 24 + translatedText = '' + + for str in cleanedList: + # Pass 1 + if not setData: + if all(x not in str for x in ['_', '@', '>', '/']) and str != '\r\n': + # Remove Textwrap and Font and Add to list + str = str.replace('\r\n', ' ') + str = str.replace(f'\\f[{fontSize}]', '') + list300.append(str) - # Pass 2 - else: - if all(x not in str for x in ['_', '@', '>', '/',]) and str != '\r\n': - # Decide Wrap - if codeList[i]['stringArgs'][0] == '[移]サウンドノベル': - width = 40 - else: - width = WIDTH - - # Add Textwrap and Font - list300[0] = textwrap.fill(list300[0], width) - list300[0] = list300[0].replace('\n', f'\r\n\\f[{fontSize}]') - list300[0] = f'\\f[{fontSize}]{list300[0]}\r\n' - translatedText += list300[0] - list300.pop(0) + # Pass 2 else: - translatedText += str + if all(x not in str for x in ['_', '@', '>', '/',]) and str != '\r\n': + # Decide Wrap + if codeList[i]['stringArgs'][0] == '[移]サウンドノベル': + width = 40 + else: + width = WIDTH - # Write to File - if setData: - # Formatting Fixes - translatedText = translatedText.replace('*"', '* "') - translatedText = translatedText.replace('\r\n\r\n', '\r\n') - translatedText = re.sub(r'[^\S\r\n]+', ' ', translatedText) - if len(codeList[i]['stringArgs']) > 1: - codeList[i]['stringArgs'][1] = translatedText - else: - codeList[i]['stringArgs'][0] = translatedText + # Add Textwrap and Font + list300[0] = textwrap.fill(list300[0], width) + list300[0] = list300[0].replace('\n', f'\r\n\\f[{fontSize}]') + list300[0] = f'\\f[{fontSize}]{list300[0]}\r\n' + translatedText += list300[0] + list300.pop(0) + else: + translatedText += str + + # Write to File + if setData: + # Formatting Fixes + translatedText = translatedText.replace('*"', '* "') + translatedText = translatedText.replace('\r\n\r\n', '\r\n') + translatedText = re.sub(r'[^\S\r\n]+', ' ', translatedText) + if len(codeList[i]['stringArgs']) > 1: + codeList[i]['stringArgs'][1] = translatedText + else: + codeList[i]['stringArgs'][0] = translatedText ### Event Code: 250 Common Events if codeList[i]['code'] == 250 and CODE250 == True: @@ -820,7 +823,7 @@ def searchDB(events, pbar, jobList, filename): scenarioList[2].pop(0) # Grab Items - if table['name'] == 'm_ﾓﾝｽﾀｰ用名前候補' and ITEMFLAG == True: + if table['name'] == 'オーブ' and ITEMFLAG == True: with open('translations.txt', 'a', encoding='utf-8') as file: for item in table['data']: dataList = item['data'] @@ -828,7 +831,7 @@ def searchDB(events, pbar, jobList, filename): # Parse # for j in range(len(dataList)): # Name - if dataList[j].get('name') == '名前': + if dataList[j].get('name') == 'オーブの名前': # Pass 1 (Grab Data) if setData == False: if dataList[j].get('value') != '': @@ -842,9 +845,9 @@ def searchDB(events, pbar, jobList, filename): itemList[0].pop(0) # Description 1 (You are my specialz) - if dataList[j].get('name') == 'NULL': + if dataList[j].get('name') == 'オーブの説明': # Clean String - fontSize = 24 + fontSize = 18 translatedText = '' cleanedList = formatDramon(dataList[j].get('value')) for str in cleanedList: @@ -864,9 +867,9 @@ def searchDB(events, pbar, jobList, filename): # Add Textwrap and Font tempText = itemList[1][0] - # tempText = textwrap.fill(tempText, width) - # tempText = tempText.replace('\n', f'\r\n\\f[{fontSize}]') - # tempText = f'\\f[{fontSize}]{tempText}\r\n' + tempText = textwrap.fill(tempText, width) + tempText = tempText.replace('\n', f'\r\n\\f[{fontSize}]') + tempText = f'\\f[{fontSize}]{tempText}\r\n' translatedText += tempText itemList[1].pop(0) else: @@ -903,9 +906,9 @@ def searchDB(events, pbar, jobList, filename): # Add Textwrap and Font tempText = itemList[2][0] - # tempText = textwrap.fill(tempText, width) - # tempText = tempText.replace('\n', f'\r\n\\f[{fontSize}]') - # tempText = f'\\f[{fontSize}]{tempText}\r\n' + tempText = textwrap.fill(tempText, width) + tempText = tempText.replace('\n', f'\r\n\\f[{fontSize}]') + tempText = f'\\f[{fontSize}]{tempText}\r\n' translatedText += tempText itemList[2].pop(0) else: diff --git a/vocab.txt b/vocab.txt index 4a68b23..9357d5a 100644 --- a/vocab.txt +++ b/vocab.txt @@ -1,5 +1,54 @@ Here are some vocabulary and terms so that you know the proper spelling and translation. ``` +# Game Characters +朱雀 (Suzuku) - Female +紅 (Akane) - Female +エージェントシリウス (Agent Sirius) - Female +白虎 (Shiratora) - Female +撫子 (Nadeshiko) - Female +鬼塚 (Onizuka) - Female +百合花 (Yurika) - Female +犬飼 (Inukai) - Female +アンナ (Anna) - Female +メープル (Maple) - Female +鈴懸 (Suzukake) - Female +いるか (Iruka) - Female +安知理 (Ajiri) - Female +結 (Yuu) - Female +綾香 (Ayaka) - Female +鳥居 (Tori) - Female +鈴蘭 (Suzuran) - Female +宇宙 (Sora) - Female +青龍 (Aotatsu) - Female +桔梗 (Kikyo) - Female +獅堂 (Shidou) - Female +向日葵 (Himawari) - Female +大紫 (Omurasaki) - Female +菫 (Sumire) - Female +亀神 (Kamegami) - Female +柊 (Shu) - Female +鷲尾 (Washio) - Male +菊一 (Kikuichi) - Male +花村 (Hanamura) - Female +咲夜 (Sakuya) - Female +曽我井 (Sogai) - Male +四海 (Shikai) - Male +志倉　綿 (Shikura Wata) - Female +皐月　亜麻 (Satsuki Ama) - Female +安知理　結 (Ajiri Yuu) - Female +栗山　五十木 (Kuriyama Itsuki) - Male +天狐 橙 (Tenkou Daidai) - Female +天狐 蜜柑 (Tenkou Mikan) - Female +鷹本 椿 (Takamoto Tsubaki) - Female +青龍 桃葉 (Aotatsu Momoba) - Female +雨宮 星羅 (Amamiya Seira) - Female +山内 紫苑 (Yamauchi Shion) - Female +野沢 林檎 (Nozawa Ringo) - Female +城ヶ峰 柳 (Jougamine Yanagi) - Female +山角 翠 (Yamakaku Midori) - Female +来杉馬来 (Kisugiba Rai) - Male +狼煙蓮 (Noroshi Ren) - Male + # Lewd Terms マンコ (pussy) おまんこ (vagina) @@ -67,4 +116,10 @@ ME 音量 (ME Volume) 堕天使 (Fallen Angel) 鬼 (Oni) ローバー (Roper) + +# Skills +Highest Repeat Rate No.1 Girl +Veteran Courtesan +Vital Point Thrust +Majikite ``` \ No newline at end of file