jsoncols.1.md

%jsoncols(1) user manual | version 1.3.5 f86e208 % R. S. Doiel % 2026-02-12

NAME

jsoncols

SYNOPSIS

jsoncols [OPTIONS] [EXPRESSION] [INPUT_FILENAME] [OUTPUT_FILENAME]

DESCRIPTION

jsoncols provides scripting flexibility for data extraction from JSON data returning the results in columns. This is helpful in flattening content extracted from JSON blobs. The default delimiter for each value extracted is a comma. This can be overridden with an option.

• EXPRESSION can be an empty string or dot notation for an object's path
• INPUT_FILENAME is the filename to read or a dash "-" if you want to explicitly read from stdin
  • if not provided then jsoncols reads from stdin
• OUTPUT_FILENAME is the filename to write or a dash "-" if you want to explicitly write to stdout
  • if not provided then jsoncols write to stdout

OPTIONS

-help : display help

-license : display license

-version : display version

-csv : output as CSV or other flat delimiter row

-d, -delimiter : set the delimiter for multi-field csv output

-i, -input : input filename

-nl, -newline : if true add a trailing newline for end of file (EOF)

-crlf : use CRLF for CSV writes, defaults to true on Windows

-o, -output : output filename

-p, -pretty : pretty print JSON output

-quiet : suppress error messages

-quote : quote strings and JSON notation

-r, -repl : run interactively

EXAMPLES

If myblob.json contained

    {"name": "Doe, Jane", "email":"jane.doe@example.org", "age": 42}

Getting just the name could be done with

    jsoncols -i myblob.json .name

This would yield

    "Doe, Jane"

Flipping .name and .age into pipe delimited columns is as easy as listing each field in the expression inside a space delimited string.

    jsoncols -i myblob.json -d\|  .name .age

This would yield

    Doe, Jane|42

You can also pipe JSON data in.

    cat myblob.json | jsoncols .name .email .age

Would yield

   "Doe, Jane","jane.doe@xample.org",42

jsoncols 1.3.5