Update app.py
Browse files
app.py
CHANGED
|
@@ -1,163 +1,150 @@
|
|
| 1 |
-
|
| 2 |
-
|
| 3 |
-
|
| 4 |
-
|
| 5 |
-
|
| 6 |
-
|
| 7 |
-
|
| 8 |
-
|
| 9 |
-
|
| 10 |
-
|
| 11 |
-
|
| 12 |
-
|
| 13 |
-
|
| 14 |
-
|
| 15 |
-
|
| 16 |
-
|
| 17 |
-
|
| 18 |
-
|
| 19 |
-
|
| 20 |
-
|
| 21 |
-
|
| 22 |
-
|
| 23 |
-
|
| 24 |
-
|
| 25 |
-
|
| 26 |
-
|
| 27 |
-
|
| 28 |
-
|
| 29 |
-
|
| 30 |
-
|
| 31 |
-
|
| 32 |
-
|
| 33 |
-
|
| 34 |
-
|
| 35 |
-
|
| 36 |
-
|
| 37 |
-
|
| 38 |
-
|
| 39 |
-
|
| 40 |
-
|
| 41 |
-
|
| 42 |
-
|
| 43 |
-
|
| 44 |
-
|
| 45 |
-
|
| 46 |
-
|
| 47 |
-
|
| 48 |
-
|
| 49 |
-
|
| 50 |
-
|
| 51 |
-
|
| 52 |
-
|
| 53 |
-
|
| 54 |
-
|
| 55 |
-
|
| 56 |
-
|
| 57 |
-
|
| 58 |
-
|
| 59 |
-
|
| 60 |
-
|
| 61 |
-
|
| 62 |
-
|
| 63 |
-
|
| 64 |
-
|
| 65 |
-
|
| 66 |
-
|
| 67 |
-
|
| 68 |
-
|
| 69 |
-
|
| 70 |
-
|
| 71 |
-
|
| 72 |
-
|
| 73 |
-
|
| 74 |
-
|
| 75 |
-
|
| 76 |
-
|
| 77 |
-
|
| 78 |
-
|
| 79 |
-
|
| 80 |
-
|
| 81 |
-
|
| 82 |
-
|
| 83 |
-
|
| 84 |
-
|
| 85 |
-
|
| 86 |
-
|
| 87 |
-
|
| 88 |
-
|
| 89 |
-
|
| 90 |
-
|
| 91 |
-
|
| 92 |
-
|
| 93 |
-
|
| 94 |
-
|
| 95 |
-
|
| 96 |
-
|
| 97 |
-
|
| 98 |
-
|
| 99 |
-
|
| 100 |
-
|
| 101 |
-
|
| 102 |
-
|
| 103 |
-
|
| 104 |
-
|
| 105 |
-
|
| 106 |
-
|
| 107 |
-
|
| 108 |
-
|
| 109 |
-
|
| 110 |
-
|
| 111 |
-
|
| 112 |
-
|
| 113 |
-
|
| 114 |
-
|
| 115 |
-
|
| 116 |
-
|
| 117 |
-
|
| 118 |
-
|
| 119 |
-
|
| 120 |
-
|
| 121 |
-
|
| 122 |
-
|
| 123 |
-
|
| 124 |
-
|
| 125 |
-
|
| 126 |
-
|
| 127 |
-
|
| 128 |
-
|
| 129 |
-
|
| 130 |
-
|
| 131 |
-
|
| 132 |
-
|
| 133 |
-
|
| 134 |
-
|
| 135 |
-
|
| 136 |
-
|
| 137 |
-
|
| 138 |
-
|
| 139 |
-
|
| 140 |
-
|
| 141 |
-
|
| 142 |
-
|
| 143 |
-
|
| 144 |
-
|
| 145 |
-
|
| 146 |
-
|
| 147 |
-
|
| 148 |
-
|
| 149 |
-
|
| 150 |
-
|
| 151 |
-
).then(
|
| 152 |
-
lambda: gr.update(visible=True),
|
| 153 |
-
outputs=[copy_status]
|
| 154 |
-
).then(
|
| 155 |
-
lambda: gr.update(visible=False),
|
| 156 |
-
inputs=None,
|
| 157 |
-
outputs=[copy_status],
|
| 158 |
-
_js="(x) => new Promise((resolve) => setTimeout(() => resolve(), 2000))"
|
| 159 |
-
)
|
| 160 |
-
|
| 161 |
-
if __name__ == "__main__":
|
| 162 |
-
# 启动 Gradio 应用
|
| 163 |
-
demo.launch(mcp_server=True)
|
|
|
|
| 1 |
+
import gradio as gr
|
| 2 |
+
import httpx
|
| 3 |
+
import readabilipy
|
| 4 |
+
import markdownify
|
| 5 |
+
import asyncio
|
| 6 |
+
|
| 7 |
+
# 使用与原始服务器中用户发起请求时相同的 User-Agent
|
| 8 |
+
# 这有助于模拟常规浏览器行为，减少被网站屏蔽的概率
|
| 9 |
+
USER_AGENT = "ModelContextProtocol/1.0 (User-Specified; +https://github.com/modelcontextprotocol/servers)"
|
| 10 |
+
|
| 11 |
+
async def fetch(url: str, force_raw: bool) -> str:
|
| 12 |
+
"""
|
| 13 |
+
异步获取并处理给定 URL 的内容。
|
| 14 |
+
|
| 15 |
+
Args:
|
| 16 |
+
url: 要抓取的网页地址。
|
| 17 |
+
force_raw: 如果为 True，则返回原始 HTML/文本，不进行 Markdown 转换。
|
| 18 |
+
|
| 19 |
+
Returns:
|
| 20 |
+
处理后的内容字符串。
|
| 21 |
+
"""
|
| 22 |
+
if not url or not url.strip():
|
| 23 |
+
return "错误：URL 不能为空。"
|
| 24 |
+
|
| 25 |
+
# 为了方便用户，如果 URL 缺少协议头，则自动添加 https://
|
| 26 |
+
if not url.startswith(('http://', 'https://')):
|
| 27 |
+
url = 'https://' + url
|
| 28 |
+
|
| 29 |
+
try:
|
| 30 |
+
# 使用 httpx 库发起异步 HTTP 请求，设置 User-Agent、允许重定向并设置30秒超时
|
| 31 |
+
async with httpx.AsyncClient(headers={"User-Agent": USER_AGENT}, follow_redirects=True, timeout=30.0) as client:
|
| 32 |
+
print(f"正在请求: {url}...")
|
| 33 |
+
response = await client.get(url)
|
| 34 |
+
|
| 35 |
+
# 检查请求是否成功，如果不成功（例如404或500错误），则会抛出异常
|
| 36 |
+
response.raise_for_status()
|
| 37 |
+
print(f"收到响应，状态码: {response.status_code}")
|
| 38 |
+
|
| 39 |
+
page_raw = response.text
|
| 40 |
+
content_type = response.headers.get("content-type", "").lower()
|
| 41 |
+
|
| 42 |
+
# 判断内容是否为 HTML
|
| 43 |
+
# 通过检查 Content-Type 或内容开头是否包含 <html> 标签来判断
|
| 44 |
+
is_page_html = "text/html" in content_type or page_raw.strip().lower().startswith("<html")
|
| 45 |
+
|
| 46 |
+
if is_page_html and not force_raw:
|
| 47 |
+
# 1. 使用 readabilipy 从原始 HTML 中提取核心文章内容
|
| 48 |
+
# 这可以有效去除广告、导航栏、页脚等无关元素
|
| 49 |
+
article = readabilipy.simple_json.simple_json_from_html_string(page_raw, use_readability=True)
|
| 50 |
+
|
| 51 |
+
if not article or not article.get("content"):
|
| 52 |
+
return f"## 无法解析页面\n\n未能成功提取页面主要内容。以下是原始 HTML：\n\n```html\n{page_raw}\n```"
|
| 53 |
+
|
| 54 |
+
# 2. 将清理后的 HTML 转换为 Markdown 格式，使其更易于阅读
|
| 55 |
+
html_content = article["content"]
|
| 56 |
+
markdown_content = markdownify.markdownify(html_content, heading_style=markdownify.ATX)
|
| 57 |
+
|
| 58 |
+
title = article.get('title', '无标题')
|
| 59 |
+
return f"# {title}\n\n{markdown_content}"
|
| 60 |
+
else:
|
| 61 |
+
# 如果内容不是 HTML 或用户选择了"原始模式"，则直接返回原始内容
|
| 62 |
+
# 使用代码块包裹，以便更好地显示
|
| 63 |
+
prefix = f"已获取原始网页内容 (Content-Type: {content_type}):\n\n"
|
| 64 |
+
lang = "html" if is_page_html else "text"
|
| 65 |
+
return f"{prefix}```{lang}\n{page_raw}\n```"
|
| 66 |
+
|
| 67 |
+
except httpx.RequestError as e:
|
| 68 |
+
return f"访问 URL 时发生网络错误: {url}\n\n详细信息: {e}"
|
| 69 |
+
except Exception as e:
|
| 70 |
+
return f"发生未知错误: {e}"
|
| 71 |
+
|
| 72 |
+
def copy_to_clipboard(content: str) -> str:
|
| 73 |
+
"""
|
| 74 |
+
将内容复制到剪贴板
|
| 75 |
+
"""
|
| 76 |
+
try:
|
| 77 |
+
import pyperclip
|
| 78 |
+
pyperclip.copy(content)
|
| 79 |
+
return "✅ 内容已复制到剪贴板！"
|
| 80 |
+
except ImportError:
|
| 81 |
+
return "❌ 无法复制：请安装 pyperclip 库 (pip install pyperclip)"
|
| 82 |
+
except Exception as e:
|
| 83 |
+
return f"❌ 复制失败：{str(e)}"
|
| 84 |
+
|
| 85 |
+
# 使用 gr.Blocks() 创建 Gradio 界面，可以更自由地布局
|
| 86 |
+
with gr.Blocks(theme=gr.themes.Soft(), title="网页内容提取工具") as demo:
|
| 87 |
+
gr.Markdown("# 网页内容提取工具 (Fetch)")
|
| 88 |
+
gr.Markdown(
|
| 89 |
+
"输入一个网址，此工具可以提取其主要内容并将其转换为干净的 Markdown 格式。"
|
| 90 |
+
"非常适合用于阅读文章，可以去除广告和导航等干扰元素。"
|
| 91 |
+
)
|
| 92 |
+
|
| 93 |
+
with gr.Row():
|
| 94 |
+
url_input = gr.Textbox(
|
| 95 |
+
label="输入网址",
|
| 96 |
+
placeholder="例如: en.wikipedia.org/wiki/Python_(programming_language)",
|
| 97 |
+
scale=4 # 使输入框更宽
|
| 98 |
+
)
|
| 99 |
+
raw_checkbox = gr.Checkbox(
|
| 100 |
+
label="获取原始 HTML",
|
| 101 |
+
info="如果勾选，将返回未经处理的完整网页 HTML 代码。",
|
| 102 |
+
scale=1
|
| 103 |
+
)
|
| 104 |
+
|
| 105 |
+
with gr.Row():
|
| 106 |
+
submit_btn = gr.Button("提取内容", variant="primary")
|
| 107 |
+
copy_btn = gr.Button("📋 复制内容", variant="secondary")
|
| 108 |
+
|
| 109 |
+
output_markdown = gr.Markdown(label="提取结果")
|
| 110 |
+
copy_status = gr.Textbox(label="复制状态", interactive=False, visible=False)
|
| 111 |
+
|
| 112 |
+
# 提供一些示例，方便用户快速体验
|
| 113 |
+
gr.Examples(
|
| 114 |
+
examples=[
|
| 115 |
+
["https://modelcontextprotocol.io/", False],
|
| 116 |
+
["https://www.gradio.app/guides/quickstart", False],
|
| 117 |
+
["https://www.anthropic.com/news/claude-3-5-sonnet", False],
|
| 118 |
+
],
|
| 119 |
+
inputs=[url_input, raw_checkbox],
|
| 120 |
+
outputs=output_markdown,
|
| 121 |
+
fn=fetch,
|
| 122 |
+
cache_examples=False, # 禁用缓存，确保每次都获取最新网页内容
|
| 123 |
+
)
|
| 124 |
+
|
| 125 |
+
# 将按钮的点击事件与核心处理函数绑定
|
| 126 |
+
submit_btn.click(
|
| 127 |
+
fn=fetch,
|
| 128 |
+
inputs=[url_input, raw_checkbox],
|
| 129 |
+
outputs=output_markdown,
|
| 130 |
+
api_name="fetch_content" # 为 API 模式命名
|
| 131 |
+
)
|
| 132 |
+
|
| 133 |
+
# 复制按钮点击事件
|
| 134 |
+
copy_btn.click(
|
| 135 |
+
fn=copy_to_clipboard,
|
| 136 |
+
inputs=[output_markdown],
|
| 137 |
+
outputs=[copy_status]
|
| 138 |
+
).then(
|
| 139 |
+
lambda: gr.update(visible=True),
|
| 140 |
+
outputs=[copy_status]
|
| 141 |
+
).then(
|
| 142 |
+
lambda: gr.update(visible=False),
|
| 143 |
+
inputs=None,
|
| 144 |
+
outputs=[copy_status],
|
| 145 |
+
_js="(x) => new Promise((resolve) => setTimeout(() => resolve(), 2000))"
|
| 146 |
+
)
|
| 147 |
+
|
| 148 |
+
if __name__ == "__main__":
|
| 149 |
+
# 启动 Gradio 应用
|
| 150 |
+
demo.launch(mcp_server=True)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|