Get fancy CLI output

Unlock the power of the MAAS CLI with jq

The MAAS CLI emits JSON data, which while information-rich, can become cumbersome to sift through, especially when dealing with lists of machines. The jq utility emerges as a lifeline, offering robust capabilities for filtering and formatting JSON data directly on the command line.

In this tutorial, we’ll journey through essential jq functionalities—key selection, array manipulation, and interplay with other CLI tools to transform raw MAAS JSON into neat, analysable tabular output.

Extracting key fields from JSON

Single key selection

To focus on the hostname field for each machine, run the following command:

maas machines read | jq '.[].hostname'

Here, jq navigates through each machine in the array (.[]) and picks the hostname field.

Multiple key selection

To fetch multiple keys, such as hostname and status_name, use:

maas machines read | jq '.[].hostname, .[].status_name'

This will produce output resembling:

[
	"vm-1",
    "Deployed"
]
[
    "vm-2", 
    "Ready"
]

Converting JSON to tabular output

Basic tabular conversion

Utilise the @tsv filter to transform JSON arrays into tab-separated values:

maas machines read | jq -r '.[].hostname, .[].status_name | @tsv'

Use -r to output raw text (devoid of quotes).

Aligned columns

For better readability, pipe the output to column -t:

maas machines read | jq -r '.[].hostname, .[].status_name | @tsv' | column -t

vm-1       Deployed
vm-2       Ready

Adding column headers to the output

Basic headers

Prepend a literal array to jq to introduce column headings:

maas machines read | jq -r '["HOSTNAME", "STATUS"], (.[] | [.hostname, .status_name]) | @tsv' | column -t

Divider row

Create a separating line between headers and data:

maas machines read | jq -r '["HOSTNAME", "STATUS"] | (.[], map(length*"-")), (.[] | [.hostname, .status_name]) | @tsv' | column -t

Sorting and filtering the output

Sorting rows

To sort the output, append sort -k 1 to the pipeline:

... | sort -k 1 

Status filtering

To sieve out machines by their status, use select():

... | jq 'select(.status_name == "Ready")'

This opens doors for more complex text processing through chaining CLI tools.

Small but powerful

This tutorial has armed you with practical skills to exploit jq for extracting, shaping, and enhancing the JSON output from the MAAS CLI. By coupling jq with other command-line utilities, you gain a potent toolkit for dissecting and analysing API outputs.

Pro-tip: If you need more jq skills, try working through the jq manual, which is freely available online.