search.py

"""
search.py — free web search via DuckDuckGo HTML scraping.
No API key, no signup, no cost. Slightly fragile (scraping), so kept
isolated here in case the markup changes and needs a patch.
"""

import requests
from bs4 import BeautifulSoup

HEADERS = {
    "User-Agent": (
        "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 "
        "(KHTML, like Gecko) Chrome/124.0.0.0 Safari/537.36"
    )
}

DDG_HTML_URL = "https://html.duckduckgo.com/html/"


def ddg_search(query: str, max_results: int = 8):
    """Returns a list of {title, url, snippet} dicts."""
    try:
        resp = requests.post(
            DDG_HTML_URL,
            data={"q": query},
            headers=HEADERS,
            timeout=10,
        )
        resp.raise_for_status()
    except requests.RequestException as e:
        return {"error": f"search request failed: {e}"}

    soup = BeautifulSoup(resp.text, "html.parser")
    results = []

    for result in soup.select("div.result"):
        link_el = result.select_one("a.result__a")
        snippet_el = result.select_one("a.result__snippet") or result.select_one(
            "div.result__snippet"
        )
        if not link_el:
            continue

        title = link_el.get_text(strip=True)
        url = link_el.get("href", "")
        snippet = snippet_el.get_text(strip=True) if snippet_el else ""

        if title and url:
            results.append({"title": title, "url": url, "snippet": snippet})

        if len(results) >= max_results:
            break

    return results