LiveJQ

livejq is JSON parser like jq but is designed to work in continuous input without crashing on invalid JSON. With json filtering.

It uses livejq.toml file to specify filter rules.

User Case

When you have a program that is printing logs which may have other formats in between like text along with JSON, and you want to parse JSON for better readability. You can use livejq to parse JSON without crashing on other formats.

Or when you want to apply filters when paring json.

Demo

livejq.mp4

Install

Install using cargo

cargo install livejq

or you can find binaries in the Release page

Usage

./your_program | livejq

Filter

To apply filtering, you need to create livejq.toml file in the project root.

It contains labels. labels are filter labels which you can apply with -f / --filter flag.

Example config file:

when not label is created, default is used. For each label, you can only give allow or disallow, not both.

#livejq.toml

allow = ["name"] # default

[network] # -f network
allow = ["net-failed"]

[memory] # -f memory
allow = ["memory-info"]

[not-console] # -f not-console
disallow = ["console"]

for seeing the schema format, check schema.toml file in this repository.

Usage:

# If no flag is given, it will use default 
# i.e allow = ["name"]
# it will only allow json who have "name" key
node main.js | livejq

# you can combine different labels together
node main.js | livejq -f network memory

# or using example_data.txt from this repository
cat example_data.txt | livejq --filter not-console

Here | is for piping output of my_program into livejq as input.