gd-crawler-mcp 0.1.2__tar.gz

gd_crawler_mcp-0.1.2/PKG-INFO

@@ -0,0 +1,11 @@
+Metadata-Version: 2.5
+Name: gd-crawler-mcp
+Version: 0.1.2
+Summary: 广东信息价爬虫 MCP：抓网页、提链接、联网搜索、PDF提取
+Requires-Python: >=3.10
+Requires-Dist: beautifulsoup4>=4.12
+Requires-Dist: duckduckgo-search>=6.0
+Requires-Dist: httpx>=0.27
+Requires-Dist: lxml>=5.0
+Requires-Dist: mcp[cli]<2,>=1.0
+Requires-Dist: pdfplumber>=0.11

gd_crawler_mcp-0.1.2/gd_crawler_mcp/__main__.py

@@ -0,0 +1,223 @@
+# -*- coding: utf-8 -*-
+"""广东信息价爬虫 MCP - stdio 模式，部署在阿里云百炼 FC 上。
+
+提供工具：
+  - fetch_webpage: 抓网页提取正文 Markdown
+  - extract_links: 提取网页中所有超链接
+  - search_web: 联网搜索（DuckDuckGo）
+  - fetch_pdf: 下载 PDF 提取文本
+"""
+from __future__ import annotations
+
+import asyncio
+import io
+import logging
+import re
+from urllib.parse import urljoin, urlsplit
+
+import httpx
+from bs4 import BeautifulSoup
+try:
+    # mcp 2.x: FastMCP 更名为 MCPServer
+    from mcp.server.mcpserver import MCPServer as _ServerCls
+except ModuleNotFoundError:
+    # mcp 1.x
+    from mcp.server.fastmcp import FastMCP as _ServerCls
+
+logging.basicConfig(level=logging.INFO, format="%(asctime)s %(levelname)s: %(message)s")
+logger = logging.getLogger("gd-crawler-mcp")
+
+mcp = _ServerCls(
+    "gd-crawler-mcp",
+    instructions=(
+        "广东信息价爬虫工具集：抓取网页正文、提取链接、联网搜索、下载PDF提取文本。"
+        "适用于查询广东省各市工程造价信息价、抓取政府网站通知公告和PDF附件。"
+    ),
+)
+
+HEADERS = {
+    "User-Agent": (
+        "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 "
+        "(KHTML, like Gecko) Chrome/131.0.0.0 Safari/537.36"
+    ),
+    "Accept": "text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8",
+    "Accept-Language": "zh-CN,zh;q=0.9,en;q=0.8",
+}
+TIMEOUT = 25.0
+MAX_CHARS = 80_000
+
+
+def _normalize_url(url: str) -> str:
+    url = url.strip()
+    if not re.match(r"^https?://", url, re.I):
+        url = "http://" + url
+    return url
+
+
+def _html_to_markdown(html: str, base_url: str = "") -> str:
+    soup = BeautifulSoup(html, "lxml")
+    for tag in soup(["script", "style", "noscript", "template"]):
+        tag.decompose()
+
+    parts = []
+    for el in soup.descendants:
+        name = getattr(el, "name", None)
+        if name is None:
+            txt = str(el).strip()
+            if txt:
+                parts.append(txt)
+        elif name == "a":
+            href = el.get("href", "")
+            txt = el.get_text(" ", strip=True)
+            if txt and href:
+                if base_url:
+                    href = urljoin(base_url, href)
+                parts.append(f"[{txt}]({href})")
+        elif name in ("h1", "h2", "h3", "h4", "h5", "h6"):
+            level = int(name[1])
+            txt = el.get_text(" ", strip=True)
+            if txt:
+                parts.append("\n" + "#" * level + " " + txt + "\n")
+        elif name in ("p", "div", "li", "tr", "blockquote"):
+            txt = el.get_text(" ", strip=True)
+            if txt:
+                parts.append(txt)
+        elif name == "td":
+            txt = el.get_text(" ", strip=True)
+            if txt:
+                parts.append(txt + " | ")
+
+    text = "\n".join(parts)
+    text = re.sub(r"\n{3,}", "\n\n", text)
+    return text[:MAX_CHARS]
+
+
+@mcp.tool()
+async def fetch_webpage(url: str, timeout: float = TIMEOUT) -> str:
+    """抓取一个网页，返回正文 Markdown。
+
+    Args:
+        url: 网页地址
+        timeout: 超时秒数
+    """
+    url = _normalize_url(url)
+    try:
+        async with httpx.AsyncClient(
+            follow_redirects=True, timeout=httpx.Timeout(timeout, connect=10.0),
+            headers=HEADERS, verify=False,
+        ) as client:
+            resp = await client.get(url)
+            resp.raise_for_status()
+            html = resp.text
+            md = _html_to_markdown(html, base_url=str(resp.url))
+            return f"# {str(resp.url)}\n\n{md}"
+    except Exception as e:
+        return f"抓取失败: {type(e).__name__}: {e}"
+
+
+@mcp.tool()
+async def extract_links(url: str, timeout: float = TIMEOUT) -> str:
+    """提取网页中所有超链接。
+
+    Args:
+        url: 网页地址
+        timeout: 超时秒数
+    """
+    url = _normalize_url(url)
+    try:
+        async with httpx.AsyncClient(
+            follow_redirects=True, timeout=httpx.Timeout(timeout, connect=10.0),
+            headers=HEADERS, verify=False,
+        ) as client:
+            resp = await client.get(url)
+            resp.raise_for_status()
+            soup = BeautifulSoup(resp.text, "lxml")
+            links = []
+            for a in soup.find_all("a", href=True):
+                href = urljoin(str(resp.url), a["href"])
+                txt = a.get_text(" ", strip=True)
+                if href.startswith("http"):
+                    links.append(f"[{txt}]({href})" if txt else href)
+            return "\n".join(links[:200])
+    except Exception as e:
+        return f"提取链接失败: {type(e).__name__}: {e}"
+
+
+@mcp.tool()
+async def search_web(query: str, max_results: int = 10) -> str:
+    """联网搜索（DuckDuckGo），返回结果标题+链接+摘要。
+
+    Args:
+        query: 搜索关键词
+        max_results: 最大返回条数
+    """
+    try:
+        from ddgs import DDGS
+    except ImportError:
+        try:
+            from duckduckgo_search import DDGS
+        except ImportError:
+            return "搜索依赖未安装"
+
+    results = []
+    try:
+        with DDGS() as ddgs:
+            for r in ddgs.text(query, max_results=max_results, region="cn-zh"):
+                results.append(
+                    f"## {r.get('title', '')}\n{r.get('body', '')}\n{r.get('href', r.get('link', ''))}\n"
+                )
+    except Exception as e:
+        return f"搜索失败: {type(e).__name__}: {e}"
+    return "\n---\n".join(results) if results else "无搜索结果"
+
+
+@mcp.tool()
+async def fetch_pdf(url: str, max_chars: int = 50000, timeout: float = 60.0) -> str:
+    """下载一个 PDF 文件并提取文本内容。
+
+    Args:
+        url: PDF 文件地址
+        max_chars: 最多返回字符数
+        timeout: 下载超时秒数
+    """
+    url = _normalize_url(url)
+    try:
+        async with httpx.AsyncClient(
+            follow_redirects=True, timeout=httpx.Timeout(timeout, connect=10.0),
+            headers=HEADERS, verify=False,
+        ) as client:
+            resp = await client.get(url)
+            resp.raise_for_status()
+            pdf_bytes = resp.content
+
+        try:
+            import pdfplumber
+            text_parts = []
+            with pdfplumber.open(io.BytesIO(pdf_bytes)) as pdf:
+                for i, page in enumerate(pdf.pages[:50]):
+                    t = page.extract_text() or ""
+                    text_parts.append(f"--- 第{i+1}页 ---\n{t}")
+            text = "\n".join(text_parts)
+        except ImportError:
+            try:
+                import fitz
+                doc = fitz.open(stream=pdf_bytes, filetype="pdf")
+                text = "\n".join(
+                    f"--- 第{i+1}页 ---\n{page.get_text()}"
+                    for i, page in enumerate(doc)
+                )
+                doc.close()
+            except ImportError:
+                return "PDF解析库未安装"
+
+        return f"# PDF: {url}\n\n{text[:max_chars]}"
+    except Exception as e:
+        return f"PDF抓取失败: {type(e).__name__}: {e}"
+
+
+def main():
+    mcp.run(transport="stdio")
+
+
+if __name__ == "__main__":
+    main()

gd_crawler_mcp-0.1.2/inline_b64.txt

@@ -0,0 +1 @@
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

gd_crawler_mcp-0.1.2/inline_server.py

@@ -0,0 +1,101 @@
+from mcp.server.fastmcp import FastMCP
+import httpx, re, io
+from bs4 import BeautifulSoup
+from urllib.parse import urljoin
+
+mcp = FastMCP("gd-crawler")
+
+H = {"User-Agent": "Mozilla/5.0 (Windows NT 10.0; Win64; x64) Chrome/131.0.0.0 Safari/537.36",
+     "Accept-Language": "zh-CN,zh;q=0.9"}
+
+def _norm(u):
+    u=u.strip()
+    if not re.match(r'^https?://',u,re.I): u='http://'+u
+    return u
+
+@mcp.tool()
+async def fetch_webpage(url: str) -> str:
+    """抓网页正文，返回 Markdown 格式文本"""
+    url=_norm(url)
+    try:
+        async with httpx.AsyncClient(follow_redirects=True,timeout=25,verify=False,headers=H) as c:
+            r=await c.get(url); r.raise_for_status()
+            soup=BeautifulSoup(r.text,'lxml')
+            for t in soup(['script','style','noscript']): t.decompose()
+            out=[]
+            for el in soup.descendants:
+                n=getattr(el,'name',None)
+                if n is None:
+                    x=str(el).strip()
+                    if x: out.append(x)
+                elif n=='a':
+                    h=el.get('href',''); x=el.get_text(' ',strip=True)
+                    if x and h: out.append(f'[{x}]({urljoin(str(r.url),h)})')
+                elif n in ('h1','h2','h3'):
+                    x=el.get_text(' ',strip=True)
+                    if x: out.append('\n'+'#'*int(n[1])+' '+x+'\n')
+                elif n in ('p','div','li','tr'):
+                    x=el.get_text(' ',strip=True)
+                    if x: out.append(x)
+            return f'# {r.url}\n\n'+re.sub(r'\n{3,}','\n\n','\n'.join(out))[:80000]
+    except Exception as e:
+        return f'抓取失败: {e}'
+
+@mcp.tool()
+async def extract_links(url: str) -> str:
+    """提取网页中所有超链接"""
+    url=_norm(url)
+    try:
+        async with httpx.AsyncClient(follow_redirects=True,timeout=25,verify=False,headers=H) as c:
+            r=await c.get(url); r.raise_for_status()
+            soup=BeautifulSoup(r.text,'lxml')
+            links=[]
+            for a in soup.find_all('a',href=True):
+                h=urljoin(str(r.url),a['href']); x=a.get_text(' ',strip=True)
+                if h.startswith('http'):
+                    links.append(f'[{x}]({h})' if x else h)
+            return '\n'.join(links[:200])
+    except Exception as e:
+        return f'提取失败: {e}'
+
+@mcp.tool()
+async def search_web(query: str, max_results: int = 10) -> str:
+    """联网搜索（DuckDuckGo），返回标题+摘要+链接"""
+    try:
+        from ddgs import DDGS
+    except ImportError:
+        from duckduckgo_search import DDGS
+    res=[]
+    try:
+        with DDGS() as d:
+            for r in d.text(query,max_results=max_results,region='cn-zh'):
+                res.append(f"## {r.get('title','')}\n{r.get('body','')}\n{r.get('href','')}\n")
+    except Exception as e:
+        return f'搜索失败: {e}'
+    return '\n---\n'.join(res) if res else '无结果'
+
+@mcp.tool()
+async def fetch_pdf(url: str, max_chars: int = 50000) -> str:
+    """下载PDF并提取文本"""
+    url=_norm(url)
+    try:
+        async with httpx.AsyncClient(follow_redirects=True,timeout=60,verify=False,headers=H) as c:
+            r=await c.get(url); r.raise_for_status()
+            data=r.content
+        try:
+            import pdfplumber
+            parts=[]
+            with pdfplumber.open(io.BytesIO(data)) as pdf:
+                for i,p in enumerate(pdf.pages[:50]):
+                    parts.append(f'--- 第{i+1}页 ---\n'+(p.extract_text() or ''))
+            text='\n'.join(parts)
+        except ImportError:
+            import fitz
+            doc=fitz.open(stream=data,filetype='pdf')
+            text='\n'.join(f'--- 第{i+1}页 ---\n{p.get_text()}' for i,p in enumerate(doc))
+            doc.close()
+        return f'# PDF: {url}\n\n'+text[:max_chars]
+    except Exception as e:
+        return f'PDF失败: {e}'
+
+mcp.run(transport='stdio')

gd_crawler_mcp-0.1.2/pyproject.toml

@@ -0,0 +1,23 @@
+[build-system]
+requires = ["hatchling"]
+build-backend = "hatchling.build"
+
+[project]
+name = "gd-crawler-mcp"
+version = "0.1.2"
+description = "广东信息价爬虫 MCP：抓网页、提链接、联网搜索、PDF提取"
+requires-python = ">=3.10"
+dependencies = [
+    "mcp[cli]>=1.0,<2",
+    "httpx>=0.27",
+    "beautifulsoup4>=4.12",
+    "lxml>=5.0",
+    "duckduckgo-search>=6.0",
+    "pdfplumber>=0.11",
+]
+
+[project.scripts]
+gd-crawler-mcp = "gd_crawler_mcp.__main__:main"
+
+[tool.hatch.build.targets.wheel]
+packages = ["gd_crawler_mcp"]