thordata-sdk 1.8.1

The Official Python SDK for Thordata - AI Data Infrastructure & Proxy Network.

Thordata Python SDK

The Official Python Client for Thordata APIs

Proxy Network • SERP API • Web Unlocker • Web Scraper API

---

📖 Introduction

The Thordata Python SDK v1.8.0 is a production-ready wrapper for Thordata's AI data infrastructure. It is architected for high reliability, strict type safety, and maximum performance.

Why v1.6.0?

• 🛡️ Bulletproof Networking: Custom core handles HTTP, HTTPS, and SOCKS5h (Remote DNS) tunneling, solving common SSL/TLS handshake issues in complex network environments.
• ⚡ Async First: First-class asyncio support with aiohttp for high-concurrency scraping (1000+ RPS).
• 🧩 100% API Coverage: Every endpoint documented by Thordata (including Hourly Usage, Server Monitor, and Task Management) is implemented.
• 🤖 Type Safe: Fully typed (mypy strict) for excellent IDE autocompletion and error checking.

---

📦 Installation

pip install thordata-sdk

---

🔐 Configuration

Set environment variables to avoid hardcoding credentials. Full reference: copy .env.example to .env and fill in values.

# [Scraping APIs]
export THORDATA_SCRAPER_TOKEN="your_scraper_token"

# [Management APIs]
export THORDATA_PUBLIC_TOKEN="your_public_token"
export THORDATA_PUBLIC_KEY="your_public_key"

# [Proxy: Residential / Unlimited / Datacenter / Mobile / ISP]
export THORDATA_RESIDENTIAL_USERNAME="your_username"
export THORDATA_RESIDENTIAL_PASSWORD="your_password"
# Optional: Unlimited (high-bandwidth) if your plan has separate credentials
# export THORDATA_UNLIMITED_USERNAME="..."
# export THORDATA_UNLIMITED_PASSWORD="..."

# Optional: Upstream proxy when behind firewall (e.g. Clash Verge port 7897)
# export THORDATA_UPSTREAM_PROXY="http://127.0.0.1:7897"

Default proxy port is 9999 (residential); other products use different ports (see .env.example).

---

🚀 Quick Start

1. SERP Search (Google/Bing)

from thordata import ThordataClient, Engine

client = ThordataClient() 

# Search Google
results = client.serp_search(
    query="latest AI trends",
    engine=Engine.GOOGLE,
    num=10,
    location="United States"
)

for item in results.get("organic", []):
    print(f"{item['title']} - {item['link']}")

2. Universal Scrape (Web Unlocker)

Automatically handles JS rendering, CAPTCHAs, and fingerprinting.

html = client.universal_scrape(
    url="https://example.com",
    js_render=True,
    country="us",
    wait_for=".content-loaded"  # Smart waiting
)

3. High-Performance Proxy Tunneling

Use Thordata's residential IPs directly with requests (Sync) or aiohttp (Async). The SDK handles the complex authentication and rotation logic.

from thordata import ProxyConfig, ProxyProduct

# Config is optional if env vars are set
proxy = ProxyConfig(
    product=ProxyProduct.RESIDENTIAL,
    country="jp",
    session_duration=10  # Sticky IP for 10 mins
)

# The client automatically routes this through Thordata's network
response = client.get("https://httpbin.org/ip", proxy_config=proxy)
print(response.json())

---

⚙️ Advanced Usage

Async High-Concurrency

Perfect for building high-throughput AI agents.

import asyncio
from thordata import AsyncThordataClient

async def main():
    async with AsyncThordataClient() as client:
        # Fire off 10 requests in parallel
        tasks = [client.serp_search(f"query {i}") for i in range(10)]
        results = await asyncio.gather(*tasks)
        print(f"Completed {len(results)} searches")

asyncio.run(main())

Task Management (Batch Scraping)

Handle large-scale scraping jobs asynchronously.

# 1. Create a task
task_id = client.create_scraper_task(
    file_name="daily_scrape",
    spider_id="universal",
    spider_name="universal",
    parameters={"url": "https://example.com"}
)

# 2. Poll for completion (Helper method)
status = client.wait_for_task(task_id, max_wait=600)

# 3. Download results
if status == "finished":
    data_url = client.get_task_result(task_id)
    print(f"Download: {data_url}")

Web Scraper Tools (120+ Pre-built Tools)

Use pre-built tools for popular platforms. See Tool Coverage Matrix for full list.

from thordata import ThordataClient
from thordata.tools import Amazon, GoogleMaps, YouTube, TikTok, eBay, Walmart

client = ThordataClient()

# Amazon Product by ASIN
task_id = client.run_tool(Amazon.ProductByAsin(asin="B0BZYCJK89"))

# Google Maps by Place ID
task_id = client.run_tool(GoogleMaps.DetailsByPlaceId(place_id="ChIJPTacEpBQwokRKwIlDXelxkA"))

# YouTube Video Download
from thordata import CommonSettings
settings = CommonSettings(resolution="<=360p", video_codec="vp9")
task_id = client.run_tool(YouTube.VideoDownload(
    url="https://www.youtube.com/watch?v=jNQXAC9IVRw",
    common_settings=settings
))

# Wait and get results
status = client.wait_for_task(task_id, max_wait=300)
if status == "ready":
    download_url = client.get_task_result(task_id)
    print(f"Results: {download_url}")

Available Platforms:

• E-Commerce: Amazon, eBay, Walmart
• Social Media: TikTok, Instagram, Facebook, Twitter/X, Reddit, LinkedIn
• Search: Google Maps, Google Shopping, Google Play
• Video: YouTube (download, info, subtitles)
• Code: GitHub
• Professional: Indeed, Glassdoor, Crunchbase
• Travel/Real Estate: Booking, Airbnb, Zillow

See examples/tools/ for more examples.

---

🛠️ Management APIs

Manage your infrastructure programmatically.

# Check Balance
balance = client.get_traffic_balance()

# Manage Whitelist
client.add_whitelist_ip("1.2.3.4")

# Create Sub-users
client.create_proxy_user("new_user", "pass123", traffic_limit=500)

# Monitor Unlimited Proxies
monitor = client.unlimited.get_server_monitor(
    ins_id="ins-123", 
    region="us", 
    start_time=1700000000, 
    end_time=1700003600
)

---

🧪 Development & Testing

• Full env reference: Copy .env.example to .env and fill in credentials.
• Unit tests (no network): pytest or python -m coverage run -m pytest -p no:cov tests && python -m coverage report -m
• Integration tests (live API/proxy): Set THORDATA_INTEGRATION=true in .env; optional THORDATA_UPSTREAM_PROXY (e.g. Clash) if behind a firewall. See CONTRIBUTING.md.

---

📄 License

MIT License. See LICENSE for details.