CODEWITHSUNDEEP
ansiblejinja2
Maximiliano Felice
Maximiliano Felice

Reputation: 367

Filter Jinja list by key existence

I'm writing an Ansible Template in Jinja, which has the following pattern:

# Inventory
[Group1]
vm1 cluster=clusterName

[Group2]
vm2

Notice that the second group does not have the property cluster defined.

# Task Definition
vars:
  potential_seeds: "{{groups.all | map('extract', hostvars) | groupby('cluster') | list}}"

This, of course, results in an obvious error:

"the field 'vars' has an invalid value, which appears to include a variable that is undefined. The error was: 'dict object' has no attribute 'cluster'

I'm in need of filtering the list in a way that it only groups the dictionaries that have cluster defined.

Upvotes: 3

Views: 3833

Answers (1)

Konstantin Suvorov
Konstantin Suvorov

Reputation: 68239

You can use selectattr prior to groupby:

groups.all | 
  map('extract', hostvars) |
  selectattr('cluster','defined') |
  groupby('cluster') |
  list

This will select only elements with cluster property defined before grouping.

Upvotes: 9

Related Questions