jello

Filter JSON and JSON Lines data using Python expressions

TLDR

Filter JSON with Python

$ echo '[1,2,3]' | jello '[x*2 for x in _]'

Extract field

$ cat [data.json] | jello '_.name'

Filter array

$ cat [data.json] | jello '[x for x in _ if x["age"] > 30]'

Get nested value

$ cat [data.json] | jello '_.users[0].email'

Print JSON schema in grep-able format

$ cat [data.json] | jello -s

Output as lines (for bash array assignment)

$ cat [data.json] | jello -l '_.items'

Load input from a file instead of stdin

$ jello -f [data.json] '_.name'

jello filters JSON and JSON Lines data using Python expressions. The input data is deserialized and made available as the underscore (_) variable. Any valid Python expression can be used to transform, filter, or extract data.The tool combines jq-like filtering with Python's full expression power, including list comprehensions, dictionary methods, string operations, and standard library functions. It handles both JSON objects and JSON Lines (newline-delimited JSON) input.

PARAMETERS

EXPRESSION

Python expression (input is _).

-p, --pretty

Pretty print output.

-l, --lines

Output JSON lines.

-r, --raw

Raw string output.

-c, --compact

Compact JSON output (no pretty printing).

-C

Force color output even when using pipes.

-m, --mono

Monochrome output (no colors).

-n, --nulls

Print selected null values.

-s, --schema

Print the JSON schema in grep-able format.

-t, --types

Print type annotations in schema view.

-e

Empty data: initialize _ as None instead of reading input.

-i

Initialize the data as a list for JSON Lines processing.

-f FILE

Load input data from a JSON file.

-q FILE

Load query from a file.

-h, --help

Display help information.

INSTALL

sudo dnf install jello

brew install jello

nix profile install nixpkgs#jello

CAVEATS

Requires Python 3.6 or later. Slower than jq for simple tasks due to Python startup overhead. The underscore (_) variable name is a convention and cannot be changed. Importing external modules is not supported by default.

HISTORY

jello was created by Kelly Brazil as a Python-based alternative to jq for users more familiar with Python syntax.