Jinja2 map() Filter

Reference: jinja

Applies a filter on a sequence of objects or looks up an attribute. This is useful when dealing with lists of objects but you are really only interested in a certain value of it.

The basic usage is mapping on an attribute. Imagine you have a list of users but you are only interested in a list of usernames:

Users on this page:

{{ users|map(attribute='username')|join(', ') }}

Alternatively you can let it invoke a filter by passing the name of the filter and the arguments afterwards. A good example would be applying a text conversion filter on a sequence:

Users on this page: {{ titles|map('lower')|join(', ') }}

Examples

Reference: OzNetNerd.com

The code below passes the names dictionary to the map filter on four separate occasions to demonstrate the ways in which it can be used:

- hosts: localhost connection: local gather_facts: no vars: names: - first: Paul last: Thompson - first: Rod last: Johnson tasks: - debug: msg={{ names | map(attribute='first') | list }} - debug: msg={{ names | map(attribute='last') | list }} - debug: msg={{ names | map('upper') | list }} - debug: msg={{ names | map(attribute='last') | map('upper') | list }}

Output #1

In this run we see that the map(attribute='first') filter prints the “first” values of the Key Value pairs:

TASK [debug] *******************************************************************
ok: [localhost] => {
    "msg": [
        "Paul", 
        "Rod"
    ]
}

Output #2

Similar to the previous run, we see that the map(attribute='last') filter prints the “last” values of the Key Value pairs:

TASK [debug] *******************************************************************
ok: [localhost] => {
    "msg": [
        "Thompson", 
        "Johnson"
    ]
}

Output #3

This example differs from the previous two in that we’re passing the upper Jinja filter to map as opposed to an attribute:

TASK [debug] *******************************************************************
ok: [localhost] => {
    "msg": [
        "{U'LAST': U'THOMPSON', U'FIRST': U'PAUL'}", 
        "{U'LAST': U'JOHNSON', U'FIRST': U'ROD'}"
    ]
}

Output #4

This example demonstrates how we can chain two map filters. Note that one is being passed an attribute while the other is being passed the upper filter:

TASK [debug] *******************************************************************
ok: [localhost] => {
    "msg": [
        "THOMPSON", 
        "JOHNSON"
    ]
}

Cisco Example

The following Playbook uses map() to extract interface information from Cisco devices:

- hosts: ""
  connection: local
  gather_facts: no

  vars:
    provider:
      host: "{{ ansible_host }}"
      username: ""
      password: ""
      platform: ""

  tasks:

  - name: ntc_show
    ntc_show_command:
      provider: ""
      connection: ""
      command: ""
      secret: ""
    register: results

  - name: DEBUG - print original results
    debug: msg=

  - name: DEBUG - print extracted results
    debug: msg={{ results.response | map(attribute='intf') | list }}

Running this Playbook results in the following output: ```

$ ansible-playbook interface_map.yml -i hosts.ini --extra-vars "hosts=gns3 username=cisco password=cisco connection=ssh platform=cisco_ios secret=cisco command='show ip interface brief'"

PLAY [gns3] ********************************************************************

TASK [ntc_show] ****************************************************************
ok: [R2]
ok: [R1]

TASK [DEBUG - print original results] ******************************************
ok: [R1] => {
    "msg": {
        "changed": false,
        "response": [
            {
                "intf": "FastEthernet0/0",
                "ipaddr": "192.168.0.210",
                "proto": "up",
                "status": "up"
            },
            {
                "intf": "FastEthernet0/1",
                "ipaddr": "unassigned",
                "proto": "down",
                "status": "administratively down"
            }
        ],
        "response_list": []
    }
}
ok: [R2] => {
    "msg": {
        "changed": false,
        "response": [
            {
                "intf": "FastEthernet0/0",
                "ipaddr": "192.168.0.197",
                "proto": "up",
                "status": "up"
            },
            {
                "intf": "FastEthernet0/1",
                "ipaddr": "unassigned",
                "proto": "down",
                "status": "administratively down"
            }
        ],
        "response_list": []
    }
}

TASK [DEBUG - print extracted results] *****************************************
ok: [R1] => {
    "msg": [
        "FastEthernet0/0",
        "FastEthernet0/1"
    ]
}
ok: [R2] => {
    "msg": [
        "FastEthernet0/0",
        "FastEthernet0/1"
    ]
}

PLAY RECAP *********************************************************************
R1                         : ok=3    changed=0    unreachable=0    failed=0
R2                         : ok=3    changed=0    unreachable=0    failed=0

Source

These Playbooks are available in the map() section of my Github repo.

---

As always, if you have any questions or have a topic that you would like me to discuss, please feel free to post a comment at the bottom of this blog entry, e-mail at will@oznetnerd.com, or drop me a message on Reddit (OzNetNerd).

Note: The opinions expressed in this blog are my own and not those of my employer.