up

Author: lgc2333

README.md

@@ -126,6 +126,8 @@ plugins = [
 |       `ASCII2D_BASE_URL`       |              否              | `https://ascii2d.net` |                                                                       Ascii2D Base URL \([#139](https://github.com/lgc-NB2Dev/YetAnotherPicSearch/issues/139)\)                                                                       |
 |    `HIDE_ASCII2D_BASE_URL`     |              否              |        `True`         |                                                                                    当自定义 Ascii2D Base URL 后，是否将自定义的 URL 替换回官方 URL                                                                                    |
 |       `EXHENTAI_COOKIES`       |              否              |        `None`         |             ExHentai 的 Cookies，没有的情况下自动改用 E-Hentai 搜图，获取方式参考 请参考 [PicImageSearch 文档](https://pic-image-search.kituin.fun/wiki/picimagesearch/E-hentai/DataStructure/#cookies%E8%8E%B7%E5%8F%96)             |
+|       `NHENTAI_BASE_URL`       |              否              | `https://ascii2d.net` |                                                                       NHentai Base URL \([#139](https://github.com/lgc-NB2Dev/YetAnotherPicSearch/issues/139)\)                                                                       |
+|    `HIDE_NHENTAI_BASE_URL`     |              否              |        `True`         |                                                                                    当自定义 NHentai Base URL 后，是否将自定义的 URL 替换回官方 URL                                                                                    |
 |      `NHENTAI_USERAGENT`       |              否              |        `None`         |     用来绕过 NHentai Cloudflare 拦截的 User Agent，配置后在 E-Hentai 标题搜索无结果时会自动调用 NHentai 标题搜索<br />先用配置的 `PROXY` 做代理，使用浏览器访问 NHentai 通过 CloudFlare 检测后，获取 UA 和 Cookies 填到对应配置项     |
 |       `NHENTAI_COOKIES`        |              否              |        `None`         |                                                                                           用来绕过 NHentai Cloudflare 拦截的 Cookies，同上                                                                                            |
 |          **行为配置**          |                              |                       |                                                                                                                                                                                                                                       |
@@ -194,6 +196,14 @@ Telegram：[@lgc2333](https://t.me/lgc2333)
 
 ## 📝 更新日志
 
+### 2.0.6
+
+- 添加配置 `NHENTAI_BASE_URL`、`HIDE_NHENTAI_BASE_URL`
+
+### 2.0.5
+
+- feat: 添加隐藏 ascii2d 网页反向代理地址的配置及功能 by [@iona-s](https://github.com/iona-s) in [#186](https://github.com/lgc-NB2Dev/YetAnotherPicSearch/pull/186)
+
 ### 2.0.4
 
 - 兼容 HTTPX 0.28

YetAnotherPicSearch/__init__.py

@@ -11,7 +11,7 @@
 
 load_search_func()
 
-__version__ = "2.0.5"
+__version__ = "2.0.6"
 __plugin_meta__ = PluginMetadata(
     name="YetAnotherPicSearch",
     description="基于 NoneBot2 及 PicImageSearch 的另一个 NoneBot 搜图插件",

YetAnotherPicSearch/config.py

@@ -16,6 +16,8 @@ class ConfigModel(BaseModel):
     exhentai_cookies: Optional[str] = None
     nhentai_useragent: Optional[str] = None
     nhentai_cookies: Optional[str] = None
+    nhentai_base_url: Annotated[str, HttpUrl] = "https://nhentai.net"
+    hide_nhentai_base_url: bool = True
 
     saucenao_low_acc: int = 60
     auto_use_ascii2d: bool = True

YetAnotherPicSearch/data_source/nhentai.py

@@ -23,13 +23,17 @@
 NHENTAI_COOKIES = parse_cookies(config.nhentai_cookies)
 
 
+def get_nh_display_base():
+    return config.nhentai_base_url if config.hide_nhentai_base_url else "https://nhentai.net"
+
+
 async def update_nhentai_info(item: NHentaiItem) -> None:
     async with AsyncClient(
         headers=NHENTAI_HEADERS,
         cookies=NHENTAI_COOKIES,
         proxy=config.proxy,
     ) as session:
-        resp = await session.get(item.url)
+        resp = await session.get(item.with_base_url(config.nhentai_base_url))
         uft8_parser = HTMLParser(encoding="utf-8")
         data = PyQuery(fromstring(resp.text, parser=uft8_parser))
         item.origin = data
@@ -44,15 +48,16 @@ async def update_nhentai_info(item: NHentaiItem) -> None:
 
 async def nhentai_title_search(title: str) -> list[UniMessage]:
     query = preprocess_search_query(title)
-    url = "https://nhentai.net/search/"
-    params = {"q": query}
     async with AsyncClient(
         headers=NHENTAI_HEADERS,
         cookies=NHENTAI_COOKIES,
         proxy=config.proxy,
     ) as session:
-        resp = await session.get(url, params=params)
-        if res := NHentaiResponse(resp.text, str(resp.url)):
+        resp = await session.get(
+            f"{config.nhentai_base_url}/search/",
+            params={"q": query},
+        )
+        if res := NHentaiResponse(resp.text, str(resp.url).replace(config.nhentai_base_url, "", 1)):
             # 只保留标题和搜索关键词相关度较高的结果，并排序，以此来提高准确度
             if res.raw:
                 res.raw = filter_results_with_ratio(res, title)
@@ -62,7 +67,11 @@ async def nhentai_title_search(title: str) -> list[UniMessage]:
 
 
 async def search_result_filter(res: NHentaiResponse) -> list[UniMessage]:
-    url = await shorten_url(res.url)
+    display_base = get_nh_display_base()
+    url = await shorten_url(
+        res.with_base_url(display_base),
+        force_shorten=True,
+    )
     if not res.raw:
         return [UniMessage.text(f"NHentai 搜索结果为空\n搜索页面：{url}")]
 
@@ -85,7 +94,7 @@ async def search_result_filter(res: NHentaiResponse) -> list[UniMessage]:
         selected_res.title,
         f"类型：{selected_res.type}",
         f"日期：{date}",
-        f"来源：{selected_res.url}",
+        f"来源：{selected_res.with_base_url(display_base)}",
         f"搜索页面：{url}",
     ]
     return [combine_message(res_list)]

YetAnotherPicSearch/nhentai_model.py

@@ -9,17 +9,23 @@ def __init__(self, data: PyQuery):
         self.origin: PyQuery = data  # 原始数据
         self.title: str = cast("str", data.find(".caption").text())
         cover = data.find(".cover")
-        self.url: str = f"https://nhentai.net{cover.attr('href')}"
+        self.href: str = cast("str", cover.attr("href"))
         self.thumbnail: str = cast("str", cover.find("img").attr("data-src"))
         self.type: str = ""
         self.date: str = ""
         self.tags: list[str] = []
 
+    def with_base_url(self, base: str) -> str:
+        return f"{base}/{self.href}"
+
 
 class NHentaiResponse:
-    def __init__(self, resp_text: str, resp_url: str):
+    def __init__(self, resp_text: str, resp_href: str):
         self.origin: str = resp_text  # 原始数据
         uft8_parser = HTMLParser(encoding="utf-8")
         data = PyQuery(fromstring(self.origin, parser=uft8_parser))
         self.raw: list[NHentaiItem] = [NHentaiItem(i) for i in data.find(".gallery").items()]
-        self.url: str = resp_url
+        self.href: str = resp_href
+
+    def with_base_url(self, base: str) -> str:
+        return f"{base}/{self.href}"

YetAnotherPicSearch/utils.py

@@ -157,7 +157,7 @@ def confuse_url(url: str) -> str:
     )
 
 
-async def shorten_url(url: str) -> str:
+async def shorten_url(url: str, force_shorten: bool = False) -> str:
     pid_search = re.compile(
         r"(?:pixiv.+(?:illust_id=|artworks/)|/img-original/img/(?:\d+/){6})(\d+)",
     )
@@ -172,7 +172,7 @@ async def shorten_url(url: str) -> str:
     if host == "danbooru.donmai.us":
         return confuse_url(url.replace("/post/show/", "/posts/"))
 
-    if host in {
+    if force_shorten or host in {
         "e-hentai.org",
         "exhentai.org",
         "graph.baidu.com",