4 ms·

How does it compare with serde, which AFAIK uses the same approach

• necubi 2y ago
  I'm not very familiar with the internals of serde_json, but it's largely solving a different problem. Serde_json is deserializing individual records (rows), and typically using compile-time code generation. (Alternatively, it can also deserialize into fully-dynamic structures like serde_json::Value). Arrow-json, by contrast, is doing columnar deserialization against a schema that's only known at runtime. To me, the interesting aspects of the design are how it does that performantly.