Parsing gigabytes of JSON per second (shared object)

 JSON is everywhere on the Internet. Servers spend a *lot* of time parsing it.
 We need a fresh approach. The simdjson library uses commonly available SIMD
 instructions and microparallel algorithms to parse JSON 4x faster than
 RapidJSON and 25x faster than JSON for Modern C++.
 .
  * Fast: Over 4x faster than commonly used production-grade JSON parsers.
  * Record Breaking Features: Minify JSON at 6 GB/s, validate UTF-8 at 13 GB/s,
    NDJSON at 3.5 GB/s.
  * Easy: First-class, easy to use and carefully documented APIs.
  * Strict: Full JSON and UTF-8 validation, lossless parsing.
    Performance with no compromises.
  * Automatic: Selects a CPU-tailored parser at runtime.
    No configuration needed.
  * Reliable: From memory allocation to error handling, simdjson's design
    avoids surprises.
  * Peer Reviewed: Our research appears in venues like VLDB Journal,
    Software: Practice and Experience.
 .
 This package ships the shared object.