novel-downloader 1.2.0__tar.gz → 1.2.1__tar.gz

{novel_downloader-1.2.0 → novel_downloader-1.2.1}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: novel-downloader
-Version: 1.2.0
+Version: 1.2.1
 Summary: A command-line tool for downloading Chinese web novels from Qidian and similar platforms.
 Author-email: Saudade Z <saudadez217@gmail.com>
 License: MIT License

{novel_downloader-1.2.0 → novel_downloader-1.2.1}/novel_downloader/__init__.py

@@ -7,7 +7,7 @@ novel_downloader
 Core package for the Novel Downloader project.
 """
 
-__version__ = "1.2.0"
+__version__ = "1.2.1"
 
 __author__ = "Saudade Z"
 __email__ = "saudadez217@gmail.com"

{novel_downloader-1.2.0 → novel_downloader-1.2.1}/novel_downloader/config/adapter.py

@@ -93,18 +93,21 @@ class ConfigAdapter:
         config["sites"][site] 中读取解析器相关配置, 返回 ParserConfig 实例
         """
         gen = self._config.get("general", {})
+        font_ocr = gen.get("font_ocr", {})
         site_cfg = self._config.get("sites", {}).get(self._site, {})
         return ParserConfig(
             cache_dir=gen.get("cache_dir", "./cache"),
-            decode_font=site_cfg.get("decode_font", False),
-            use_freq=site_cfg.get("use_freq", False),
-            use_ocr=site_cfg.get("use_ocr", True),
-            use_vec=site_cfg.get("use_vec", False),
-            ocr_version=site_cfg.get("ocr_version", "v1.0"),
-            save_font_debug=site_cfg.get("save_font_debug", False),
-            batch_size=site_cfg.get("batch_size", 32),
-            ocr_weight=site_cfg.get("ocr_weight", 0.6),
-            vec_weight=site_cfg.get("vec_weight", 0.4),
+            decode_font=font_ocr.get("decode_font", False),
+            use_freq=font_ocr.get("use_freq", False),
+            use_ocr=font_ocr.get("use_ocr", True),
+            use_vec=font_ocr.get("use_vec", False),
+            ocr_version=font_ocr.get("ocr_version", "v1.0"),
+            save_font_debug=font_ocr.get("save_font_debug", False),
+            batch_size=font_ocr.get("batch_size", 32),
+            gpu_mem=font_ocr.get("gpu_mem", 500),
+            gpu_id=font_ocr.get("gpu_id", None),
+            ocr_weight=font_ocr.get("ocr_weight", 0.6),
+            vec_weight=font_ocr.get("vec_weight", 0.4),
             mode=site_cfg.get("mode", "session"),
         )
 

{novel_downloader-1.2.0 → novel_downloader-1.2.1}/novel_downloader/config/models.py

@@ -24,10 +24,10 @@ from typing import Any, Dict, List, Literal, Optional, TypedDict
 # === Requesters ===
 @dataclass
 class RequesterConfig:
-    wait_time: int = 5
+    wait_time: float = 5.0
     retry_times: int = 3
-    retry_interval: int = 5
-    timeout: int = 30
+    retry_interval: float = 5.0
+    timeout: float = 30.0
     headless: bool = True
     user_data_folder: str = ""
     profile_name: str = ""
@@ -41,7 +41,7 @@ class RequesterConfig:
 # === Downloaders ===
 @dataclass
 class DownloaderConfig:
-    request_interval: int = 5
+    request_interval: float = 5.0
     raw_data_dir: str = "./raw_data"
     cache_dir: str = "./novel_cache"
     download_workers: int = 4
@@ -63,6 +63,8 @@ class ParserConfig:
     use_vec: bool = False
     ocr_version: str = "v1.0"
     batch_size: int = 32
+    gpu_mem: int = 500
+    gpu_id: Optional[int] = None
     ocr_weight: float = 0.6
     vec_weight: float = 0.4
     save_font_debug: bool = False

{novel_downloader-1.2.0 → novel_downloader-1.2.1}/novel_downloader/core/downloaders/base_async_downloader.py

@@ -94,7 +94,7 @@ class BaseAsyncDownloader(AsyncDownloaderProtocol, abc.ABC):
         return self._config.login_required
 
     @property
-    def request_interval(self) -> int:
+    def request_interval(self) -> float:
         return self._config.request_interval
 
     async def prepare(self) -> None:

{novel_downloader-1.2.0 → novel_downloader-1.2.1}/novel_downloader/core/interfaces/async_requester_protocol.py

@@ -28,7 +28,9 @@ class AsyncRequesterProtocol(Protocol):
         """
         ...
 
-    async def get_book_info(self, book_id: str, wait_time: Optional[int] = None) -> str:
+    async def get_book_info(
+        self, book_id: str, wait_time: Optional[float] = None
+    ) -> str:
         """
         Fetch the raw HTML (or JSON) of the book info page asynchronously.
 
@@ -39,7 +41,7 @@ class AsyncRequesterProtocol(Protocol):
         ...
 
     async def get_book_chapter(
-        self, book_id: str, chapter_id: str, wait_time: Optional[int] = None
+        self, book_id: str, chapter_id: str, wait_time: Optional[float] = None
     ) -> str:
         """
         Fetch the raw HTML (or JSON) of a single chapter asynchronously.
@@ -51,7 +53,7 @@ class AsyncRequesterProtocol(Protocol):
         """
         ...
 
-    async def get_bookcase(self, wait_time: Optional[int] = None) -> str:
+    async def get_bookcase(self, wait_time: Optional[float] = None) -> str:
         """
         Optional: Retrieve the HTML content of the authenticated
         user's bookcase page asynchronously.

{novel_downloader-1.2.0 → novel_downloader-1.2.1}/novel_downloader/core/interfaces/requester_protocol.py

@@ -26,7 +26,7 @@ class RequesterProtocol(Protocol):
         """
         ...
 
-    def get_book_info(self, book_id: str, wait_time: Optional[int] = None) -> str:
+    def get_book_info(self, book_id: str, wait_time: Optional[float] = None) -> str:
         """
         Fetch the raw HTML (or JSON) of the book info page.
 
@@ -37,7 +37,7 @@ class RequesterProtocol(Protocol):
         ...
 
     def get_book_chapter(
-        self, book_id: str, chapter_id: str, wait_time: Optional[int] = None
+        self, book_id: str, chapter_id: str, wait_time: Optional[float] = None
     ) -> str:
         """
         Fetch the raw HTML (or JSON) of a single chapter.
@@ -55,7 +55,7 @@ class RequesterProtocol(Protocol):
         """
         ...
 
-    def get_bookcase(self, wait_time: Optional[int] = None) -> str:
+    def get_bookcase(self, wait_time: Optional[float] = None) -> str:
         """
         Optional: Retrieve the HTML content of the authenticated user's bookcase page.
 

{novel_downloader-1.2.0 → novel_downloader-1.2.1}/novel_downloader/core/parsers/qidian_parser/browser/main_parser.py

@@ -60,6 +60,8 @@ class QidianBrowserParser(BaseParser):
                 use_ocr=config.use_ocr,
                 use_vec=config.use_vec,
                 batch_size=config.batch_size,
+                gpu_mem=config.gpu_mem,
+                gpu_id=config.gpu_id,
                 ocr_weight=config.ocr_weight,
                 vec_weight=config.vec_weight,
                 font_debug=config.save_font_debug,

{novel_downloader-1.2.0 → novel_downloader-1.2.1}/novel_downloader/core/parsers/qidian_parser/session/main_parser.py

@@ -63,6 +63,8 @@ class QidianSessionParser(BaseParser):
                 use_ocr=config.use_ocr,
                 use_vec=config.use_vec,
                 batch_size=config.batch_size,
+                gpu_mem=config.gpu_mem,
+                gpu_id=config.gpu_id,
                 ocr_weight=config.ocr_weight,
                 vec_weight=config.vec_weight,
                 font_debug=config.save_font_debug,

{novel_downloader-1.2.0 → novel_downloader-1.2.1}/novel_downloader/core/requesters/base_async_session.py

@@ -51,7 +51,7 @@ class BaseAsyncSession(AsyncRequesterProtocol, abc.ABC):
 
     Attributes:
         _session (ClientSession): The persistent aiohttp client session.
-        _timeout (int): Timeout for each request in seconds.
+        _timeout (float): Timeout for each request in seconds.
         _retry_times (int): Number of retry attempts on failure.
         _retry_interval (float): Delay (in seconds) between retries.
         _headers (Dict[str, str]): Default HTTP headers to send.
@@ -111,7 +111,9 @@ class BaseAsyncSession(AsyncRequesterProtocol, abc.ABC):
         )
 
     @abc.abstractmethod
-    async def get_book_info(self, book_id: str, wait_time: Optional[int] = None) -> str:
+    async def get_book_info(
+        self, book_id: str, wait_time: Optional[float] = None
+    ) -> str:
         """
         Fetch the raw HTML (or JSON) of the book info page asynchronously.
 
@@ -123,7 +125,7 @@ class BaseAsyncSession(AsyncRequesterProtocol, abc.ABC):
 
     @abc.abstractmethod
     async def get_book_chapter(
-        self, book_id: str, chapter_id: str, wait_time: Optional[int] = None
+        self, book_id: str, chapter_id: str, wait_time: Optional[float] = None
     ) -> str:
         """
         Fetch the raw HTML (or JSON) of a single chapter asynchronously.
@@ -135,7 +137,7 @@ class BaseAsyncSession(AsyncRequesterProtocol, abc.ABC):
         """
         ...
 
-    async def get_bookcase(self, wait_time: Optional[int] = None) -> str:
+    async def get_bookcase(self, wait_time: Optional[float] = None) -> str:
         """
         Optional: Retrieve the HTML content of the authenticated user's bookcase page.
         Subclasses that support user login/bookcase should override this.
@@ -238,7 +240,7 @@ class BaseAsyncSession(AsyncRequesterProtocol, abc.ABC):
         return self._session
 
     @property
-    def timeout(self) -> int:
+    def timeout(self) -> float:
         """Return the default timeout setting."""
         return self._timeout
 

{novel_downloader-1.2.0 → novel_downloader-1.2.1}/novel_downloader/core/requesters/base_browser.py

@@ -111,7 +111,7 @@ class BaseBrowser(RequesterProtocol, abc.ABC):
         )
 
     @abc.abstractmethod
-    def get_book_info(self, book_id: str, wait_time: Optional[int] = None) -> str:
+    def get_book_info(self, book_id: str, wait_time: Optional[float] = None) -> str:
         """
         Fetch the raw HTML (or JSON) of the book info page.
 
@@ -123,7 +123,7 @@ class BaseBrowser(RequesterProtocol, abc.ABC):
 
     @abc.abstractmethod
     def get_book_chapter(
-        self, book_id: str, chapter_id: str, wait_time: Optional[int] = None
+        self, book_id: str, chapter_id: str, wait_time: Optional[float] = None
     ) -> str:
         """
         Fetch the raw HTML (or JSON) of a single chapter.
@@ -135,7 +135,7 @@ class BaseBrowser(RequesterProtocol, abc.ABC):
         """
         ...
 
-    def get_bookcase(self, wait_time: Optional[int] = None) -> str:
+    def get_bookcase(self, wait_time: Optional[float] = None) -> str:
         """
         Optional: Retrieve the HTML content of the authenticated user's bookcase page.
 

{novel_downloader-1.2.0 → novel_downloader-1.2.1}/novel_downloader/core/requesters/base_session.py

@@ -28,7 +28,7 @@ class BaseSession(RequesterProtocol, abc.ABC):
 
     Attributes:
         _session (requests.Session): The persistent HTTP session.
-        _timeout (int): Timeout for each request in seconds.
+        _timeout (float): Timeout for each request in seconds.
     """
 
     def _init_session(
@@ -81,7 +81,7 @@ class BaseSession(RequesterProtocol, abc.ABC):
         )
 
     @abc.abstractmethod
-    def get_book_info(self, book_id: str, wait_time: Optional[int] = None) -> str:
+    def get_book_info(self, book_id: str, wait_time: Optional[float] = None) -> str:
         """
         Fetch the raw HTML (or JSON) of the book info page.
 
@@ -93,7 +93,7 @@ class BaseSession(RequesterProtocol, abc.ABC):
 
     @abc.abstractmethod
     def get_book_chapter(
-        self, book_id: str, chapter_id: str, wait_time: Optional[int] = None
+        self, book_id: str, chapter_id: str, wait_time: Optional[float] = None
     ) -> str:
         """
         Fetch the raw HTML (or JSON) of a single chapter.
@@ -105,7 +105,7 @@ class BaseSession(RequesterProtocol, abc.ABC):
         """
         ...
 
-    def get_bookcase(self, wait_time: Optional[int] = None) -> str:
+    def get_bookcase(self, wait_time: Optional[float] = None) -> str:
         """
         Optional: Retrieve the HTML content of the authenticated user's bookcase page.
 
@@ -171,7 +171,7 @@ class BaseSession(RequesterProtocol, abc.ABC):
         return self._session
 
     @property
-    def timeout(self) -> int:
+    def timeout(self) -> float:
         """Return the default timeout setting."""
         return self._timeout
 

{novel_downloader-1.2.0 → novel_downloader-1.2.1}/novel_downloader/core/requesters/common_requester/common_async_session.py

@@ -45,7 +45,9 @@ class CommonAsyncSession(BaseAsyncSession):
         self._site = site
         self._profile = profile
 
-    async def get_book_info(self, book_id: str, wait_time: Optional[int] = None) -> str:
+    async def get_book_info(
+        self, book_id: str, wait_time: Optional[float] = None
+    ) -> str:
         """
         Fetch the raw HTML of the book info page asynchronously.
 
@@ -62,7 +64,7 @@ class CommonAsyncSession(BaseAsyncSession):
         return html
 
     async def get_book_chapter(
-        self, book_id: str, chapter_id: str, wait_time: Optional[int] = None
+        self, book_id: str, chapter_id: str, wait_time: Optional[float] = None
     ) -> str:
         """
         Fetch the raw HTML of a single chapter asynchronously.

{novel_downloader-1.2.0 → novel_downloader-1.2.1}/novel_downloader/core/requesters/common_requester/common_session.py

@@ -47,7 +47,7 @@ class CommonSession(BaseSession):
         self._site = site
         self._profile = profile
 
-    def get_book_info(self, book_id: str, wait_time: Optional[int] = None) -> str:
+    def get_book_info(self, book_id: str, wait_time: Optional[float] = None) -> str:
         """
         Fetch the raw HTML (or JSON) of the book info page.
 
@@ -75,7 +75,7 @@ class CommonSession(BaseSession):
         raise RuntimeError("Unexpected error: get_book_info failed without returning")
 
     def get_book_chapter(
-        self, book_id: str, chapter_id: str, wait_time: Optional[int] = None
+        self, book_id: str, chapter_id: str, wait_time: Optional[float] = None
     ) -> str:
         """
         Fetch the raw HTML (or JSON) of a single chapter.

{novel_downloader-1.2.0 → novel_downloader-1.2.1}/novel_downloader/core/requesters/qidian_requester/qidian_broswer.py

@@ -266,7 +266,7 @@ class QidianBrowser(BaseBrowser):
         """
         return self.QIDIAN_BOOKCASE_URL
 
-    def get_book_info(self, book_id: str, wait_time: Optional[int] = None) -> str:
+    def get_book_info(self, book_id: str, wait_time: Optional[float] = None) -> str:
         """
         Retrieve the HTML of a Qidian book info page.
 
@@ -311,7 +311,7 @@ class QidianBrowser(BaseBrowser):
             time.sleep(pause)
 
     def get_book_chapter(
-        self, book_id: str, chapter_id: str, wait_time: Optional[int] = None
+        self, book_id: str, chapter_id: str, wait_time: Optional[float] = None
     ) -> str:
         """
         Retrieve the HTML content of a specific chapter.
@@ -347,7 +347,7 @@ class QidianBrowser(BaseBrowser):
             logger.warning("[fetch] Error fetching chapter from '%s': %s", url, e)
             return ""
 
-    def get_bookcase(self, wait_time: Optional[int] = None) -> str:
+    def get_bookcase(self, wait_time: Optional[float] = None) -> str:
         """
         Retrieve the HTML content of the logged‑in user's Qidian bookcase page.
 

{novel_downloader-1.2.0 → novel_downloader-1.2.1}/novel_downloader/core/requesters/qidian_requester/qidian_session.py

@@ -108,7 +108,7 @@ class QidianSession(BaseSession):
         self.get("https://www.qidian.com")
         return True
 
-    def get_book_info(self, book_id: str, wait_time: Optional[int] = None) -> str:
+    def get_book_info(self, book_id: str, wait_time: Optional[float] = None) -> str:
         """
         Fetch the raw HTML of the book info page.
 
@@ -140,7 +140,7 @@ class QidianSession(BaseSession):
         raise RuntimeError("Unexpected fall-through in get_book_info")
 
     def get_book_chapter(
-        self, book_id: str, chapter_id: str, wait_time: Optional[int] = None
+        self, book_id: str, chapter_id: str, wait_time: Optional[float] = None
     ) -> str:
         """
         Fetch the HTML of a single chapter.
@@ -174,7 +174,7 @@ class QidianSession(BaseSession):
 
         raise RuntimeError("Unexpected fall-through in get_book_chapter")
 
-    def get_bookcase(self, wait_time: Optional[int] = None) -> str:
+    def get_bookcase(self, wait_time: Optional[float] = None) -> str:
         """
         Retrieve the user's *bookcase* page.
 

{novel_downloader-1.2.0 → novel_downloader-1.2.1}/novel_downloader/resources/config/settings.yaml

@@ -1,9 +1,9 @@
 # 网络请求层设置
 requests:
-  wait_time: 5                      # 每次请求等待时间 (秒)
+  wait_time: 5.0                    # 每次请求等待时间 (秒)
   retry_times: 3                    # 请求失败重试次数
-  retry_interval: 5
-  timeout: 30                       # 页面加载超时时间 (秒)
+  retry_interval: 5.0
+  timeout: 30.0                     # 页面加载超时时间 (秒)
   max_rps: null                     # 最大请求速率 (requests per second), 为 null 则不限制
   # DrissionPage 专用设置
   headless: false                   # 是否以无头模式启动浏览器
@@ -15,7 +15,7 @@ requests:
 
 # 全局通用设置
 general:
-  request_interval: 5               # 同一本书各章节请求间隔 (秒)
+  request_interval: 5.0             # 同一本书各章节请求间隔 (秒)
   raw_data_dir: "./raw_data"        # 原始章节 HTML/JSON 存放目录
   output_dir: "./downloads"         # 最终输出文件存放目录
   cache_dir: "./novel_cache"        # 本地缓存目录 (字体 / 图片等)
@@ -26,6 +26,18 @@ general:
   debug:
     save_html: false                # 是否将抓取到的原始 HTML 保留到磁盘
     log_level: "INFO"               # 日志级别: DEBUG, INFO, WARNING, ERROR
+  font_ocr:
+    decode_font: false              # 是否尝试本地解码混淆字体
+    use_freq: false                 # 是否使用频率分析
+    ocr_version: "v2.0"             # "v1.0" / "v2.0"
+    use_ocr: true                   # 是否使用 OCR 辅助识别文本
+    use_vec: false                  # 是否使用 Vector 辅助识别文本
+    save_font_debug: false          # 是否保存字体解码调试数据
+    batch_size: 32
+    gpu_mem: 500                    # GPU 显存限制 (MB)
+    gpu_id: null                    # 使用哪个 GPU
+    ocr_weight: 0.6
+    vec_weight: 0.4
 
 # 各站点的特定配置
 sites:
@@ -38,15 +50,6 @@ sites:
       - "0000000000"
     mode: "browser"                 # browser / session
     login_required: true            # 是否需要登录才能访问
-    decode_font: false              # 是否尝试本地解码混淆字体
-    use_freq: false                 # 是否使用频率分析
-    ocr_version: "v2.0"             # "v1.0" / "v2.0"
-    use_ocr: true                   # 是否使用 OCR 辅助识别文本
-    use_vec: false                  # 是否使用 Vector 辅助识别文本
-    save_font_debug: false          # 是否保存字体解码调试数据
-    batch_size: 32
-    ocr_weight: 0.6
-    vec_weight: 0.4
     #
   sample_site:
     book_ids:

{novel_downloader-1.2.0 → novel_downloader-1.2.1}/novel_downloader/utils/fontocr/ocr_v2.py

@@ -221,6 +221,8 @@ class FontOCRV2:
         use_ocr: bool = True,
         use_vec: bool = False,
         batch_size: int = 32,
+        gpu_mem: int = 500,
+        gpu_id: Optional[int] = None,
         ocr_weight: float = 0.6,
         vec_weight: float = 0.4,
         ocr_version: str = "v1.0",
@@ -232,6 +234,8 @@ class FontOCRV2:
         self.use_ocr = use_ocr
         self.use_vec = use_vec
         self.batch_size = batch_size
+        self.gpu_mem = gpu_mem
+        self.gpu_id = gpu_id
         self.ocr_weight = ocr_weight
         self.vec_weight = vec_weight
         self.ocr_version = ocr_version
@@ -279,6 +283,8 @@ class FontOCRV2:
             rec_batch_num=self.batch_size,
             use_space_char=False,
             use_gpu=gpu_available,
+            gpu_mem=self.gpu_mem,
+            gpu_id=self.gpu_id,
         )
 
     def _load_char_freq_db(self) -> bool:

{novel_downloader-1.2.0 → novel_downloader-1.2.1}/novel_downloader.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: novel-downloader
-Version: 1.2.0
+Version: 1.2.1
 Summary: A command-line tool for downloading Chinese web novels from Qidian and similar platforms.
 Author-email: Saudade Z <saudadez217@gmail.com>
 License: MIT License

{novel_downloader-1.2.0 → novel_downloader-1.2.1}/pyproject.toml

@@ -104,7 +104,7 @@ testpaths = ["tests"]
 
 [tool.commitizen]
 name = "cz_conventional_commits"
-version = "1.2.0"
+version = "1.2.1"
 version_files = [
     "novel_downloader/__init__.py"
 ]