Tdx Professional Data Source

Why This Skill Exists

When building quantitative trading systems or financial analysis tools, obtaining reliable stock market data is the first major hurdle. The pytdx library wraps Tongdaxin's data protocol into Python interfaces, but using it directly presents several challenges: difficulty selecting stable servers (performance varies significantly), complex API parameters (market codes, K-line periods, etc.), and lack of systematic documentation.

This skill addresses these pain points by providing a comprehensive reference system and intelligent connection strategy, allowing you to focus on data analysis rather than wrestling with data acquisition details.

Core Capabilities and Key Steps

Dual-Engine API Architecture

The skill encapsulates Tongdaxin's two core interfaces, covering virtually all financial instruments:

Market codes follow Tongdaxin standards: 0 for Shenzhen Stock Exchange, 1 for Shanghai Stock Exchange. K-line periods range from 0 (5-minute) to 11 (yearly).

Intelligent Server Selection Mechanism

A core highlight is the automated server optimization strategy:

1. Multi-dimensional scoring: Combines preset weights, historical success rates, response times, and freshness
2. Automatic failover: Servers with 3+ consecutive failures are automatically downgraded
3. Seamless switching: Default 2 automatic retries without manual intervention

Pre-configured optimal servers (e.g., 115.238.56.198:7709, weight 1.0) are empirically validated for out-of-the-box reliability.

Standard Workflow

A typical usage scenario:

# 1. Connect (with auto resource management)
with TdxHq_API() as api:
    # 2. Select market and stock code
    market = 1  # Shanghai
    code = "600519"

    # 3. Fetch K-line data
    kline = api.get_kline_data(
        market=market,
        code=code,
        category=9,  # Daily K-line with volume
        start=0
    )

    # 4. Convert to DataFrame for analysis
    df = kline.to_df()

The skill includes references/api-reference.md (detailed parameter documentation) and references/usage-examples.md (scenario-based code samples).

Limitations and Considerations

Data source limitations: pytdx retrieves data via Tongdaxin protocol, essentially scraping public market data. It does not include Level 2 depth data or real-time tick-by-tick trades. For Level 2 data, you'll need alternative sources.

Network dependency: As data is fetched in real-time, network instability directly impacts data retrieval. Use in stable network environments and implement timeout/retry mechanisms.

Extended market specifics: Futures and options require TdxExHq_API and dedicated server addresses, different from A-share standard servers. Pay attention to API type selection.

Historical data indexing: The start parameter begins from 0 (counting backwards from latest). To fetch data for specific dates, calculate offset values accordingly.

In summary, this skill suits scenarios requiring batch stock market data retrieval for quantitative analysis. It solves data acquisition engineering challenges, but analysis and strategy development still require your domain expertise.