Add: Keyword extract

dev_tools/button_extract.py

@@ -206,11 +206,10 @@ def generate_code():
         path = os.path.join(AzurLaneConfig.ASSETS_MODULE, module.split('/', maxsplit=1)[0])
         output = os.path.join(path, 'assets')
         gen = CodeGenerator()
-        gen.add('from module.base.button import Button, ButtonWrapper')
-        gen.Empty()
-        gen.Comment('This file was auto-generated, do not modify it manually. To generate:')
-        gen.Comment('``` python -m dev_tools.button_extract ```')
-        gen.Empty()
+        gen.Import("""
+        from module.base.button import Button, ButtonWrapper
+        """)
+        gen.CommentAutoGenerage('dev_tools.button_extract')
         for assets, assets_data in module_data.items():
             has_share = SHARE_SERVER in assets_data
             with gen.Object(key=assets, object_class='ButtonWrapper'):

dev_tools/keyword_extract.py

@@ -0,0 +1,108 @@
+import os
+import typing as t
+from functools import cached_property
+
+from module.base.code_generator import CodeGenerator
+from module.config.utils import read_file
+from module.logger import logger
+from module.ocr.keyword import text_to_variable
+
+UI_LANGUAGES = ['cn', 'cht', 'en', 'jp']
+
+
+class TextMap:
+    DATA_FOLDER = ''
+
+    def __init__(self, lang: str):
+        self.lang = lang
+
+    @cached_property
+    def data(self) -> dict[int, str]:
+        if not TextMap.DATA_FOLDER:
+            logger.critical('`TextMap.DATA_FOLDER` is empty, please set it to your path to StarRailData')
+            exit(1)
+        file = os.path.join(TextMap.DATA_FOLDER, 'TextMap', f'TextMap{self.lang.upper()}.json')
+        data = {}
+        for id_, text in read_file(file).items():
+            data[int(id_)] = text
+        return data
+
+    def find(self, name: t.Union[int, str]) -> tuple[int, str]:
+        """
+        Args:
+            name:
+
+        Returns:
+            text id (hash in TextMap)
+            text
+        """
+        if isinstance(name, int) or (isinstance(name, str) and name.isdigit()):
+            name = int(name)
+            try:
+                return name, self.data[name]
+            except KeyError:
+                pass
+
+        name = str(name)
+        for row_id, row_name in self.data.items():
+            if row_id >= 0 and row_name == name:
+                return row_id, row_name
+        for row_id, row_name in self.data.items():
+            if row_name == name:
+                return row_id, row_name
+        logger.error(f'Cannot find name: "{name}" in language {self.lang}')
+        return 0, ''
+
+
+class KeywordExtract:
+    def __init__(self):
+        self.text_map: dict[str, TextMap] = {lang: TextMap(lang) for lang in UI_LANGUAGES}
+        self.keywords_id: list[int] = []
+
+    def find_keyword(self, keyword, lang):
+        text_map = self.text_map[lang]
+        return text_map.find(keyword)
+
+    def load_keywords(self, keywords: list[str], lang='cn'):
+        text_map = self.text_map[lang]
+        self.keywords_id = [text_map.find(keyword)[0] for keyword in keywords]
+        self.keywords_id = [keyword for keyword in self.keywords_id if keyword != 0]
+
+    def write_keywords(
+            self,
+            keyword_class,
+            output_file: str
+    ):
+        """
+        Args:
+            keyword_class:
+            keyword_import:
+            output_file:
+        """
+        gen = CodeGenerator()
+        gen.Import(f"""
+        from .classes import {keyword_class}
+        """)
+        gen.CommentAutoGenerage('dev_tools.keyword_extract')
+        for index, keyword in enumerate(self.keywords_id):
+            _, en = self.find_keyword(keyword, lang='en')
+            en = text_to_variable(en)
+            with gen.Object(key=en, object_class=keyword_class):
+                gen.ObjectAttr(key='id', value=index + 1)
+                for lang in UI_LANGUAGES:
+                    gen.ObjectAttr(key=lang, value=self.find_keyword(keyword, lang=lang)[1])
+
+        gen.write(output_file)
+
+
+def generate():
+    ex = KeywordExtract()
+    ex.load_keywords(['模拟宇宙', '拟造花萼（金）', '拟造花萼（赤）', '凝滞虚影', '侵蚀隧洞', '忘却之庭'])
+    ex.write_keywords(keyword_class='DungeonNav', output_file='./tasks/dungeon/keywords/nav.py')
+    ex.load_keywords(['行动摘要', '生存索引', '每日实训'])
+    ex.write_keywords(keyword_class='DungeonTab', output_file='./tasks/dungeon/keywords/tab.py')
+
+
+if __name__ == '__main__':
+    TextMap.DATA_FOLDER = r''
+    generate()

module/base/code_generator.py

@@ -100,6 +100,19 @@ class CodeGenerator:
             line = line.strip()
             self.add(line, comment=True)
 
+    def CommentAutoGenerage(self, file):
+        """
+        Args:
+            file: dev_tools.button_extract
+        """
+        # Only leave one blank line at above
+        if len(self.lines) >= 2:
+            if self.lines[-2:] == ['\n', '\n']:
+                self.lines.pop(-1)
+        self.Comment('This file was auto-generated, do not modify it manually. To generate:')
+        self.Comment(f'``` python -m {file} ```')
+        self.Empty()
+
     def List(self, key=None):
         if key is not None:
             return TabWrapper(self, prefix=str(key) + ' = [', suffix=']')

module/ocr/keyword.py

@@ -6,7 +6,7 @@ from typing import ClassVar
 from module.exception import ScriptError
 import module.config.server as server
 
-REGEX_PUNCTUATION = re.compile(r'[,.\'"，。\-—/\\\n\t]')
+REGEX_PUNCTUATION = re.compile(r'[ ,.\'"，。\-—/\\\n\t()（）]')
 
 
 def parse_name(n):
@@ -14,13 +14,19 @@ def parse_name(n):
     return n
 
 
+def text_to_variable(text):
+    text = re.sub('[ \-]', '_', text)
+    text = re.sub('[()]', '', text)
+    return text
+
+
 @dataclass
 class Keyword:
     id: int
     cn: str
     en: str
     jp: str
-    tw: str
+    cht: str
 
     """
     Instance attributes and methods
@@ -39,8 +45,26 @@ class Keyword:
         return parse_name(self.jp)
 
     @cached_property
-    def tw_parsed(self) -> str:
-        return parse_name(self.tw)
+    def cht_parsed(self) -> str:
+        return parse_name(self.cht)
+
+    @cached_property
+    def name(self) -> str:
+        return text_to_variable(self.en)
+
+    def __str__(self):
+        return self.name
+
+    __repr__ = __str__
+
+    def __eq__(self, other):
+        return str(self) == str(other)
+
+    def __hash__(self):
+        return hash(self.name)
+
+    def __bool__(self):
+        return True
 
     def _keywords_to_find(self, in_current_server=False, ignore_punctuation=True):
         if in_current_server:
@@ -62,23 +86,23 @@ class Keyword:
                         return [self.jp]
                 case 'tw':
                     if ignore_punctuation:
-                        return [self.tw_parsed]
+                        return [self.cht_parsed]
                     else:
-                        return [self.tw]
+                        return [self.cht]
         else:
             if ignore_punctuation:
                 return [
                     self.cn_parsed,
                     self.en_parsed,
                     self.jp_parsed,
-                    self.tw_parsed,
+                    self.cht_parsed,
                 ]
             else:
                 return [
                     self.cn,
                     self.en,
                     self.jp,
-                    self.tw,
+                    self.cht,
                 ]
 
     """
@@ -114,7 +138,10 @@ class Keyword:
             except KeyError:
                 pass
 
+        if ignore_punctuation:
             name = parse_name(name)
+        else:
+            name = str(name)
         instance: Keyword
         for instance in cls.instances.values():
             for keyword in instance._keywords_to_find(
@@ -123,27 +150,3 @@ class Keyword:
                     return instance
 
         raise ScriptError(f'Cannot find a {cls.__name__} instance that matches "{name}"')
-
-
-@dataclass
-class Zone(Keyword):
-    shape: str
-    hazard_level: int
-    region: int
-
-
-DIC = {
-    0: {'shape': 'J10', 'hazard_level': 1, 'cn': 'NY', 'en': 'NY City', 'jp': 'NYシティ', 'tw': '紐約',
-        'area_pos': (262, 567), 'offset_pos': (15, 15), 'region': 3},
-    1: {'shape': 'J10', 'hazard_level': 1, 'cn': '利维浦', 'en': 'Liverpool', 'jp': 'リバープール', 'tw': '利物浦',
-        'area_pos': (1446, 887), 'offset_pos': (15, 15), 'region': 2},
-    2: {'shape': 'J10', 'hazard_level': 1, 'cn': '直布罗特', 'en': 'Gibraltar', 'jp': 'ジブラルタル', 'tw': '直布羅陀',
-        'area_pos': (1418, 495), 'offset_pos': (15, 15), 'region': 4},
-    3: {'shape': 'J10', 'hazard_level': 1, 'cn': '圣彼得伯格', 'en': 'St. Petersburg', 'jp': 'ペテルブルク', 'tw': '聖彼得堡',
-        'area_pos': (1998, 1095), 'offset_pos': (15, 15), 'region': 2},
-}
-Zone(id=0, cn='NY', en='NY City', jp='NYシティ', tw='紐約', shape='J10', hazard_level=1, region=3)
-Zone(id=1, cn='利维浦', en='Liverpool', jp='リバープール', tw='利物浦', shape='J10', hazard_level=1, region=2)
-Zone(id=2, cn='直布罗特', en='Gibraltar', jp='ジブラルタル', tw='直布羅陀', shape='J10', hazard_level=1, region=3)
-
-print(Zone.find('直布羅陀'))

module/ocr/ocr.py

@@ -23,7 +23,7 @@ class OcrResultButton:
         try:
             self.matched_keyword = keyword_class.find(
                 boxed_result.ocr_text, in_current_server=True, ignore_punctuation=True)
-            self.name = f'{keyword_class.__name__}_{self.matched_keyword.id}'
+            self.name = str(self.matched_keyword)
         except ScriptError:
             self.matched_keyword = None
             self.name = boxed_result.ocr_text
@@ -148,5 +148,5 @@ class Ocr:
         ]
         results = [result for result in results if result.matched_keyword is not None]
         logger.attr(name=f'{self.name} matched',
-                    text='[' + ', '.join([result.name for result in results]) + ']')
+                    text=results)
         return results

tasks/dungeon/keywords/__init__.py

@@ -0,0 +1,3 @@
+import tasks.dungeon.keywords.nav as KEYWORDS_DUNGEON_NAV
+import tasks.dungeon.keywords.tab as KEYWORDS_DUNGEON_TAB
+from tasks.dungeon.keywords.classes import DungeonNav, DungeonTab

tasks/dungeon/keywords/classes.py

@@ -0,0 +1,13 @@
+from dataclasses import dataclass
+
+from module.ocr.keyword import Keyword
+
+
+@dataclass
+class DungeonNav(Keyword):
+    pass
+
+
+@dataclass
+class DungeonTab(Keyword):
+    pass

tasks/dungeon/keywords/nav.py

@@ -0,0 +1,47 @@
+from .classes import DungeonNav
+
+# This file was auto-generated, do not modify it manually. To generate:
+# ``` python -m dev_tools.keyword_extract ```
+
+Simulated_Universe = DungeonNav(
+    id=1,
+    cn='模拟宇宙',
+    cht='模擬宇宙',
+    en='Simulated Universe',
+    jp='模擬宇宙',
+)
+Calyx_Golden = DungeonNav(
+    id=2,
+    cn='拟造花萼（金）',
+    cht='擬造花萼（金）',
+    en='Calyx (Golden)',
+    jp='疑似花萼（金）',
+)
+Calyx_Crimson = DungeonNav(
+    id=3,
+    cn='拟造花萼（赤）',
+    cht='擬造花萼（赤）',
+    en='Calyx (Crimson)',
+    jp='疑似花萼（赤）',
+)
+Stagnant_Shadow = DungeonNav(
+    id=4,
+    cn='凝滞虚影',
+    cht='凝滯虛影',
+    en='Stagnant Shadow',
+    jp='凝結虚影',
+)
+Cavern_of_Corrosion = DungeonNav(
+    id=5,
+    cn='侵蚀隧洞',
+    cht='侵蝕隧洞',
+    en='Cavern of Corrosion',
+    jp='侵蝕トンネル',
+)
+Forgotten_Hall = DungeonNav(
+    id=6,
+    cn='忘却之庭',
+    cht='忘卻之庭',
+    en='Forgotten Hall',
+    jp='忘却の庭',
+)

tasks/dungeon/keywords/tab.py

@@ -0,0 +1,26 @@
+from .classes import DungeonTab
+
+# This file was auto-generated, do not modify it manually. To generate:
+# ``` python -m dev_tools.keyword_extract ```
+
+Operation_Briefing = DungeonTab(
+    id=1,
+    cn='行动摘要',
+    cht='行動摘要',
+    en='Operation Briefing',
+    jp='行動要旨',
+)
+Survival_Index = DungeonTab(
+    id=2,
+    cn='生存索引',
+    cht='生存索引',
+    en='Survival Index',
+    jp='生存手引書',
+)
+Daily_Training = DungeonTab(
+    id=3,
+    cn='每日实训',
+    cht='每日實訓',
+    en='Daily Training',
+    jp='デイリー訓練',
+)