JSON parser uses lots of memory and takes twice the time of jq

[tooptoop4]

for 100mb file (10k records)

this takes 2mins and uses 14gb memory:
mlr --ijson count <file>

while jq length command takes 1min and uses 5gb memory

[tooptoop4]

i found https://github.com/TeskaLabs/cysimdjson to be best for just counting records and https://github.com/python-rapidjson/python-rapidjson for accessing all the fields