jq is ~2 times faster in no-filter mode

[Pooya Moradi]

I ran a test on a 3.8MiB json file with hyperfine.

hyperfine --export-markdown a.md -r 3 \
    './query-json '.'  channels.json' \
    'jq '.'  channels.json'

Command	Mean [ms]	Min [ms]	Max [ms]	Relative
./query-json . channels.json	485.8 ± 1.4	484.4	487.2	1.93 ± 0.01
jq . channels.json	251.2 ± 0.4	250.9	251.7	1.00

Summary
  'jq .  channels.json' ran
    1.93 ± 0.01 times faster than './query-json .  channels.json'

$ jq --version && ./query-json --version
jq-1.5-1-a5b5cbe
0.5.13

[David Sancho]

Hey @pvonmoradi

Thanks for opening the issue, even there's no issue here.

The benchmarks that I made with big JSON files are around 90M and there the performance is similar. Would be nice to have this benchmark as part of the repo, please feel free to add it and I will keep an eye on it.

Thanks again.