we recently wrote our own json parser for etl style operations on huge json streams. the idea was that you are still able to define your structs/classes and an

we recently wrote our own json parser for etl style operations on huge json streams. the idea was that you are still able to define your structs/classes and annotated members are auto-parsed at full speed. some sse4 optimizations where used and performance should be on par or faster than rapidJSON (could not benchmark yet). http://code.dlang.org/packages/asdf http://code.dlang.org/packages/asdf as a simple tool you could try je, which is similar to a simple version of jq but to extract json objects as a csv/tsv table. its about 10x faster than jq. https://github.com/tamediadigital/je https://github.com/tamediadigital/je