tools/wikipedia.py

import os
import requests

# Wikipedia needs the local proxy to be accessible from China
_PROXY = os.getenv("LOCAL_PROXY", "http://127.0.0.1:7890")
_wiki_session = requests.Session()
_wiki_session.proxies = {"http": _PROXY, "https": _PROXY}
_wiki_session.trust_env = False
_wiki_session.headers.update({
    "User-Agent": "Mozilla/5.0 (compatible; ResearchAgent/1.0; educational use)"
})


def wikipedia_search(query: str) -> str:
    """Search Wikipedia for encyclopedic information."""
    # Use Wikipedia REST API directly
    search_url = "https://en.wikipedia.org/w/api.php"
    try:
        # Step 1: search for page title
        search_resp = _wiki_session.get(search_url, timeout=15, params={
            "action": "query",
            "list": "search",
            "srsearch": query,
            "srlimit": 3,
            "format": "json",
        })
        search_resp.raise_for_status()
        results = search_resp.json().get("query", {}).get("search", [])
        if not results:
            return "No Wikipedia articles found."

        # Step 2: fetch content of top result
        title = results[0]["title"]
        content_resp = _wiki_session.get(search_url, timeout=15, params={
            "action": "query",
            "titles": title,
            "prop": "extracts",
            "exintro": False,
            "explaintext": True,
            "format": "json",
        })
        content_resp.raise_for_status()
        pages = content_resp.json().get("query", {}).get("pages", {})
        page = next(iter(pages.values()))
        text = page.get("extract", "")
        if not text:
            return f"Wikipedia article '{title}' has no extractable content."
        url = f"https://en.wikipedia.org/wiki/{title.replace(' ', '_')}"
        return f"Wikipedia: {title}\nURL: {url}\n\n{text[:3000]}"
    except Exception as e:
        return f"Wikipedia search failed: {e}"