What's the main reason ripgrep is fast?

According to the README, ripgrep is fast for several interconnected reasons [README.md:L183-L210]:

1. Rust's regex engine: It uses finite automata, SIMD, and aggressive literal optimizations. PCRE2 support is available via the -P/--pcre2 flag for cases where you need additional features.

2. UTF-8 decoding integration: Rust's regex library maintains performance while providing full Unicode support by building UTF-8 decoding directly into its deterministic finite automaton engine.

3. Intelligent search strategy selection: ripgrep supports both memory-mapped file searching (better for single files) and incremental searching with an intermediate buffer (better for large directories), choosing the best strategy automatically.

4. Efficient ignore pattern matching: It applies .gitignore patterns using a RegexSet, allowing a single file path to be matched against multiple glob patterns simultaneously.

5. Parallel directory traversal: It uses a lock-free parallel recursive directory iterator via crossbeam and ignore libraries.

The foundational advantage is the finite automaton-based regex engine (as opposed to backtracking), which provides worst-case linear time complexity [FAQ.md:L450-L451].

FOLLOW_UPS:

• How does ripgrep choose between memory maps and buffers?
• What's the difference between the default regex engine and PCRE2?
• How does ripgrep handle binary file detection?