🔎 async-search-scraper

Query a dozen search engines from a single async Python call — and get back a unified, deduplicated list of results.

• 🧠 One unified async API over 12 engines (Bing, Brave, Yahoo, Startpage, AOL, Ask, Tor's Torch, …).
• ⚡ Drop-in CLI for one-shot searches with print / html / csv / json output.
• 🧰 Built-in pagination, deduplication, result filtering, and HTTP/SOCKS proxy support.

---

🚀 Quick start

pip install -r requirements.txt

import asyncio
from search_engines import Bing

async def main():
    async with Bing() as engine:
        results = await engine.search("my query")
        print(results.links())

asyncio.run(main())

That's it — results.links() gives you a flat list[str] of URLs; results itself iterates over {host, link, title, text} dicts.

---

🧭 Supported engines

Engine	Status	Notes
Bing	✅ Working
Brave	✅ Working	Uses the official Brave Search API; set BRAVE_API_KEY or pass api_key=....
Yahoo	✅ Working	Handles the GDPR consent redirect automatically.
Startpage	✅ Working
AOL	✅ Working
Torch	🧅 Tor only	Requires a running TOR proxy (socks5://127.0.0.1:9050).
Ask	❌ Deprecated
Google	❌ Deprecated
DuckDuckGo	❌ Deprecated
Dogpile	❌ Deprecated
Mojeek	❌ Deprecated
Qwant	❌ Deprecated

---

📚 Usage as a library

Single engine

import asyncio
from search_engines import Bing

async def main():
    async with Bing() as engine:
        results = await engine.search("my query", pages=2)
        for item in results:
            print(item["title"], "→", item["link"])

asyncio.run(main())

Multiple engines, deduplicated

import asyncio
from search_engines import Bing, Yahoo, Startpage

async def main():
    all_links = set()
    for cls in (Bing, Yahoo, Startpage):
        async with cls() as engine:
            engine.ignore_duplicate_urls = True
            results = await engine.search("python programming", pages=1)
            all_links.update(results.links())
    print(f"{len(all_links)} unique URLs")

asyncio.run(main())

Brave (API key)

Note

Brave uses the official Search API. Get a free key at api.search.brave.com and either export BRAVE_API_KEY or pass api_key=... to the constructor.

import asyncio, os
from search_engines import Brave

async def main():
    async with Brave(api_key=os.environ["BRAVE_API_KEY"]) as engine:
        results = await engine.search("my query")
        print(results.links())

asyncio.run(main())

---

🖥️ Usage as a CLI

python search_engines_cli.py -e bing,yahoo -q "my query" -o json,print

Flag	Purpose	Default
-q	Search query (required)	—
-e	Engine(s), comma-separated, or all	google
-p	Number of pages to fetch	20
-o	Output: any combination of print, html, csv, json	print
-n	Output filename (without extension)	search_results/output
-f	Filter results by url / title / text / host	none
-i	Drop duplicate URLs across engines	off
-proxy	HTTP/SOCKS proxy URL (protocol://ip:port)	none

Typical output:

Searching Bing
page: 1        links: 10
page: 2        links: 20
1  https://www.python.org/                          Welcome to Python.org
2  https://en.wikipedia.org/wiki/Python_(...)       Python (programming language) - Wikipedia
...

---

⚙️ Configuration

• BRAVE_API_KEY — environment variable read by the Brave engine when api_key= isn't passed explicitly.
• Proxy — any URL supported by aiohttp_socks: http://, https://, socks4://, socks5://. Pass via Bing(proxy="socks5://127.0.0.1:9050") or -proxy on the CLI.
• Tunables — TIMEOUT, USER_AGENT, default page count, and output directory live in search_engines/config.py.

---

🧩 Extending — adding a new engine

Drop a new class into search_engines/engines/, subclass SearchEngine, override _selectors, _first_page, _next_page, then register it in search_engines/engines/__init__.py's search_engines_dict. Mimic bing.py for a pure-HTML engine, or brave.py / ask.py for JSON-driven ones.

---

📦 Requirements

Python 3.9–3.13, plus the pinned dependencies in requirements.txt (aiohttp, aiohttp_socks, beautifulsoup4, curl_cffi, requests).

pip install -r requirements.txt
python setup.py install   # optional, to install the package itself

---

📄 License

MIT — see LICENSE.

Originally created by Tasos M. Adamopoulos — see Search-Engines-Scraper for the upstream repo.