Explanations¶
Explain how things work under the hood and why design decisions were made.
What are Explanations?¶
Explanations provide in-depth understanding of the library's internal mechanics, design philosophy, and algorithms. Unlike tutorials (which teach you how to do something) or how-to guides (which provide step-by-step solutions), explanations focus on the "why" and "how it works."
Documentation Types¶
| Type | Purpose | Example |
|---|---|---|
| Tutorials | Learning-oriented, step-by-step lessons | "Make your first parallel request" |
| How-to Guides | Problem-solving, practical solutions | "Handle rate limits for an API" |
| Reference | Technical details, API documentation | FastRequests class parameters |
| Explanations | Understanding internal design | "Why token bucket for rate limiting?" |
When to Consult Explanations¶
Consult explanations when you want to understand:
- Design decisions - Why certain algorithms or patterns were chosen
- Internal mechanics - How features work under the hood
- Trade-offs - Benefits and limitations of different approaches
- Performance characteristics - How different backends or settings affect performance
- Algorithm details - Mathematical formulas or implementation details
Explanation Topics¶
Architecture¶
Architecture - Design philosophy, component overview, and how parts interact
Backends¶
Backends - HTTP client library comparison, auto-detection, and when to use each
Rate Limiting¶
Rate Limiting - Token bucket algorithm, burst handling, and concurrency control
Retry Strategy¶
Retry Strategy - Exponential backoff, jitter, and thundering herd prevention
Proxy Rotation¶
Proxy Rotation - IP rotation, proxy validation, and health tracking
Related Documentation¶
- Tutorials - Start here for hands-on learning
- How-to Guides - Practical solutions to common problems
- Reference - Complete API documentation