Fix/p0 p3 code review fixes

README.md

@@ -220,7 +220,7 @@ jmv 350234 -y
 ## 项目特点
 
 - **绕过Cloudflare的反爬虫**
-- **实现禁漫APP接口最新的加解密算法 (1.6.3)**
+- **实现禁漫APP接口最新的加解密算法**
 - 用法多样：
 
   - GitHub

pyproject.toml

@@ -25,17 +25,20 @@ classifiers=[
 ]
 dependencies = [
     "commonx>=0.6.38",
-    "curl-cffi",
-    "pillow",
-    "pycryptodome",
-    "pyyaml",
+    "curl-cffi>=0.5.10",
+    "pillow>=10.0.0",
+    "pycryptodome>=3.19.0",
+    "pyyaml>=6.0",
 ]
 dynamic = ["version"]
 
 [project.urls]
 Homepage = "https://github.com/hect0x7/JMComic-Crawler-Python"
 Documentation = "https://jmcomic.readthedocs.io"
 
+[project.optional-dependencies]
+extra = ["zhconv", "pyzipper", "py7zr", "img2pdf", "psutil"]
+
 [project.scripts]
 jmcomic = "jmcomic.cli:main"
 jmv = "jmcomic.cli:view_main"

src/jmcomic/api.py

@@ -78,31 +78,22 @@ def download_album(jm_album_id,
                    ) -> Union[__DOWNLOAD_API_RET, Set[__DOWNLOAD_API_RET]]:
     """
     下载一个本子（album），包含其所有的章节（photo）
-
-    当jm_album_id不是str或int时，视为批量下载，相当于调用 download_batch(download_album, jm_album_id, option, downloader)
-
+    当jm_album_id不是str或int时，视为批量下载
     :param jm_album_id: 本子的禁漫车号
     :param option: 下载选项
     :param downloader: 下载器类
     :param callback: 返回值回调函数，可以拿到 album 和 downloader
-    :param check_exception: 是否检查异常, 如果为True，会检查downloader是否有下载异常，并上抛PartialDownloadFailedException
-    :param extra: 下载特性（Feature），下载时动态挂载的附加行为上下文。会自动根据上下文（如 album/photo 来源）自适应参数行为。支持单个 Feature、FeatureChain、或列表
-    :return: 对于的本子实体类，下载器（如果是上述的批量情况，返回值为download_batch的返回值）
+    :param check_exception: 是否检查异常
+    :param extra: 下载特性（Feature）
+    :return: DownloadResult (如果是批量情况返回 BatchResult)
     """
-
     if not isinstance(jm_album_id, (str, int)):
         return download_batch(download_album, jm_album_id, option, downloader, extra=extra)
 
-    with new_downloader(option, downloader) as dler:
-        # 注册 Feature 及来源，由 downloader 在 after_album 钩子中自动执行
-        dler.add_features(extra, 'download_album')
-        album = dler.download_album(jm_album_id)
-
-        if callback is not None:
-            callback(album, dler)
-        if check_exception:
-            dler.raise_if_has_exception()
-        return DownloadResult(album, dler)
+    return _download_and_return(
+        jm_album_id, option, downloader, callback, check_exception, extra,
+        'download_album', lambda d: d.download_album(jm_album_id),
+    )
 
 
 def download_photo(jm_photo_id,
@@ -113,21 +104,35 @@ def download_photo(jm_photo_id,
                    extra=None,
                    ):
     """
-    下载一个章节（photo），参数同 download_album
+    下载一个章节（photo）
+    当jm_photo_id不是str或int时，视为批量下载
+    :param jm_photo_id: 章节的禁漫车号
+    :param option: 下载选项
+    :param downloader: 下载器类
+    :param callback: 返回值回调函数
+    :param check_exception: 是否检查异常
+    :param extra: 下载特性（Feature）
     """
     if not isinstance(jm_photo_id, (str, int)):
         return download_batch(download_photo, jm_photo_id, option, downloader, extra=extra)
 
+    return _download_and_return(
+        jm_photo_id, option, downloader, callback, check_exception, extra,
+        'download_photo', lambda d: d.download_photo(jm_photo_id),
+    )
+
+
+def _download_and_return(jm_id, option, downloader, callback, check_exception, extra,
+                         feature_source, download_fn):
     with new_downloader(option, downloader) as dler:
-        # 注册 Feature 及来源，由 downloader 在 after_photo 钩子中自动执行
-        dler.add_features(extra, 'download_photo')
-        photo = dler.download_photo(jm_photo_id)
+        dler.add_features(extra, feature_source)
+        entity = download_fn(dler)
 
         if callback is not None:
-            callback(photo, dler)
+            callback(entity, dler)
         if check_exception:
             dler.raise_if_has_exception()
-        return DownloadResult(photo, dler)
+        return DownloadResult(entity, dler)
 
 
 def new_downloader(option=None, downloader=None) -> JmDownloader:
@@ -182,11 +187,9 @@ async def download_album_async(jm_album_id,
                                extra=None,
                                ):
     """
-    异步下载一个本子（album），包含其所有的章节（photo）。
-
-    - 支持批量下载（当 jm_album_id 为可迭代对象时）
-    - callback 支持同步函数和异步函数
-    - 返回 (album, downloader) 元组，其中 downloader 的网络和线程池资源已关闭，仅用于读取下载结果
+    异步下载一个本子（album），包含其所有的章节（photo）
+    callback 支持同步函数和异步函数
+    返回的 downloader 已关闭网络和线程池资源，仅用于读取下载结果
     """
     if not isinstance(jm_album_id, (str, int)):
         return await download_batch_async(download_album_async,
@@ -196,15 +199,10 @@ async def download_album_async(jm_album_id,
                                           extra=extra
                                           )
 
-    async with new_async_downloader(option, downloader) as dler:
-        dler.add_features(extra, 'download_album')
-        album = await dler.download_album(jm_album_id)
-
-        await _invoke_async_callback(callback, album, dler)
-        if check_exception:
-            dler.raise_if_has_exception()
-
-        return DownloadResult(album, dler)
+    return await _download_async_and_return(
+        jm_album_id, option, downloader, callback, check_exception, extra,
+        'download_album', lambda d: d.download_album(jm_album_id),
+    )
 
 
 async def download_photo_async(jm_photo_id,
@@ -215,9 +213,9 @@ async def download_photo_async(jm_photo_id,
                                extra=None,
                                ):
     """
-    异步下载一个章节（photo）。
-    callback 支持同步函数和异步函数。
-    返回的 downloader 已关闭网络和线程池资源，仅用于读取下载结果。
+    异步下载一个章节（photo）
+    callback 支持同步函数和异步函数
+    返回的 downloader 已关闭网络和线程池资源，仅用于读取下载结果
     """
     if not isinstance(jm_photo_id, (str, int)):
         return await download_batch_async(download_photo_async,
@@ -227,15 +225,23 @@ async def download_photo_async(jm_photo_id,
                                           extra=extra
                                           )
 
+    return await _download_async_and_return(
+        jm_photo_id, option, downloader, callback, check_exception, extra,
+        'download_photo', lambda d: d.download_photo(jm_photo_id),
+    )
+
+
+async def _download_async_and_return(jm_id, option, downloader, callback, check_exception, extra,
+                                     feature_source, download_fn):
     async with new_async_downloader(option, downloader) as dler:
-        dler.add_features(extra, 'download_photo')
-        photo = await dler.download_photo(jm_photo_id)
+        dler.add_features(extra, feature_source)
+        entity = await download_fn(dler)
 
-        await _invoke_async_callback(callback, photo, dler)
+        await _invoke_async_callback(callback, entity, dler)
         if check_exception:
             dler.raise_if_has_exception()
 
-        return DownloadResult(photo, dler)
+        return DownloadResult(entity, dler)
 
 
 async def download_batch_async(download_api,

src/jmcomic/cli.py

@@ -102,6 +102,10 @@ def run(self, option):
         from .api import download_album, download_photo
         from common import MultiTaskLauncher
 
+        if len(self.album_id_list) == 0 and len(self.photo_id_list) == 0:
+            print('未指定任何 id，请提供 album 或 photo 的 id，例如: jmcomic 123')
+            return
+
         if len(self.album_id_list) == 0:
             download_photo(self.photo_id_list, option)
         elif len(self.photo_id_list) == 0:

src/jmcomic/jm_async_downloader.py

@@ -9,7 +9,6 @@
 from __future__ import annotations
 
 import asyncio
-import os
 from concurrent.futures import ThreadPoolExecutor
 
 from .jm_downloader import BaseDownloader
@@ -50,6 +49,7 @@ def __init__(self,
         self._photo_semaphore = asyncio.Semaphore(photo_concurrency)
 
         # 解密线程池（CPU 密集操作卸载）
+        decode_worker = int(decode_worker if decode_worker is not None else option.download.threading.decode_worker)
         self._decode_pool = ThreadPoolExecutor(max_workers=decode_worker, thread_name_prefix='jm-async-decode')
 
     # ======================================================================
@@ -142,8 +142,9 @@ async def _download_single_image(self, image: JmImageDetail):
         对齐 sync JmDownloader.download_by_image_detail 的逻辑。
         """
         img_save_path = self.option.decide_image_filepath(image)
+        from common import file_exists
         image.save_path = img_save_path
-        image.exists = os.path.exists(img_save_path)
+        image.exists = file_exists(img_save_path)
         image.cache = self.option.decide_download_cache(image)
 
         await self.before_image(image, img_save_path)

src/jmcomic/jm_client_impl.py

@@ -437,13 +437,7 @@ def favorite_folder(self,
 
         return JmPageTool.parse_html_to_favorite_page(resp.text)
 
-    # noinspection PyTypeChecker
-    def get_username_from_cookies(self) -> str:
-        # cookies = self.get_meta_data('cookies', None)
-        # if not cookies:
-        #     ExceptionTool.raises('未登录，无法获取到对应的用户名，请给favorite方法传入username参数')
-        # 解析cookies，可能需要用到 phpserialize，比较麻烦，暂不实现
-        pass
+
 
     def get_jm_html(self, url, require_200=True, **kwargs):
         """
@@ -698,12 +692,23 @@ def get_scramble_id(self, photo_id, album_id=None):
         if album_id is not None and album_id in cache:
             return cache[album_id]
 
-        scramble_id = self.fetch_scramble_id(photo_id)
-        cache[photo_id] = scramble_id
-        if album_id is not None:
-            cache[album_id] = scramble_id
+        if JmModuleConfig.SCRAMBLE_CACHE_LOCK is None:
+            from threading import Lock
+            JmModuleConfig.SCRAMBLE_CACHE_LOCK = Lock()
 
-        return scramble_id
+        with JmModuleConfig.SCRAMBLE_CACHE_LOCK:
+            # double-check after acquiring lock
+            if photo_id in cache:
+                return cache[photo_id]
+            if album_id is not None and album_id in cache:
+                return cache[album_id]
+
+            scramble_id = self.fetch_scramble_id(photo_id)
+            cache[photo_id] = scramble_id
+            if album_id is not None:
+                cache[album_id] = scramble_id
+
+            return scramble_id
 
     def fetch_detail_entity(self, jmid, clazz: Type[DetailType]) -> DetailType:
         """
@@ -963,16 +968,16 @@ def raise_if_resp_should_retry(self, resp, is_image):
             msg = JmModuleConfig.JM_ERROR_STATUS_CODE.get(code, f'HTTP状态码: {code}')
             ExceptionTool.raises_resp(f"禁漫API异常响应, {msg}", resp)
 
-        url = resp.request.url
+        url = getattr(resp, 'url', '') or getattr(getattr(resp, 'request', None), 'url', '')
 
         if self.API_SCRAMBLE in url:
             # /chapter_view_template 这个接口不是返回json数据，不做检查
             return resp
 
         text = resp.text
-        for char in text:
+        # 只检查前1024个字符，避免遍历大型HTML页面
+        for char in text[:1024]:
             if char not in (' ', '\n', '\t'):
-                # 找到第一个有效字符
                 ExceptionTool.require_true(
                     char == '{',
                     f'请求不是json格式，强制重试！响应文本: [{JmcomicText.limit_text(text, 200)}]'
@@ -1099,10 +1104,12 @@ def __init__(self, future, after_done_callback):
 
         def result(self):
             if not self.done:
-                result = self.future.result()
-                self._result = result
-                self.done = True
-                self.future = None  # help gc
+                try:
+                    result = self.future.result()
+                    self._result = result
+                finally:
+                    self.done = True
+                    self.future = None  # help gc
                 self.after_done_callback()
 
             return self._result

src/jmcomic/jm_config.py

@@ -135,6 +135,7 @@ class JmModuleConfig:
 
     # 图片分隔相关
     SCRAMBLE_CACHE = {}
+    SCRAMBLE_CACHE_LOCK = None  # threading.Lock 延迟初始化
 
     # 当本子没有作者名字时，顶替作者名字
     DEFAULT_AUTHOR = 'default_author'
@@ -465,6 +466,7 @@ def new_postman(cls, session=False, **kwargs):
             'threading': {
                 'image': 30,
                 'photo': None,
+                'decode_worker': None,
             },
         },
         'client': {
@@ -524,11 +526,14 @@ def option_default_dict(cls) -> dict:
             # use system proxy by default
             meta_data['proxies'] = cls.DEFAULT_PROXIES
 
-        # threading photo
+        # threading photo & decode_worker
         dt = option_dict['download']['threading']
         if dt['photo'] is None:
             import os
             dt['photo'] = os.cpu_count()
+        if dt['decode_worker'] is None:
+            import os
+            dt['decode_worker'] = min(4, os.cpu_count() or 1)
 
         return option_dict
 
@@ -619,3 +624,10 @@ def enable_pretty_log():
     handler.setFormatter(PrettyFormatter())
     jm_logger.addHandler(handler)
     jm_logger.setLevel(logging.INFO)
+
+
+def format_album_url(aid, domain='18comic.vip'):
+    """
+    把album_id变为可访问的URL，方便print打印后用浏览器访问
+    """
+    return f'{JmModuleConfig.PROT}{domain}/album/{aid}/'

src/jmcomic/jm_exception.py

@@ -4,6 +4,7 @@
 from typing import NoReturn
 
 from .jm_entity import *
+from .jm_config import format_album_url
 
 
 class JmcomicException(Exception):
@@ -156,8 +157,7 @@ def raise_missing(cls,
         :param resp: 响应对象
         :param jmid: 禁漫本子/章节id
         """
-        from .jm_toolkit import JmcomicText
-        url = JmcomicText.format_album_url(jmid)
+        url = format_album_url(jmid)
 
         req_type = "本子" if "album" in url else "章节"
         cls.raises(