Parsing CSV with Ansible - csv

Question

I need to parse a CSV with Ansible. The issue I'm facing is that I don't have fixed columns, they are variables.
For example: I need to extract network informations and to use them for configure vmware guests networks. In this case I can have a CSV with these:
(other columns) Nic1_Vlan Nic1_IP Nic1_MASK Nic1_Scope (other columns)
Otherwise I can have this:
(other columns) Nic1_Vlan Nic1_IP Nic1_MASK Nic1_Scope Nic2_Vlan Nic2_IP Nic2_MASK Nic2_Scope (other columns )
I can have 1 nic only or 3 in one shot.
Do you have any hint about this scenario? Would you use jinja2 or do you think is better to use a json/yaml variable files?
Thank you for any reply!!
Regards

Answer 1

For example the play below
- hosts: localhost
vars:
rec_len: 4
delimeter: ' '
tasks:
- set_fact:
nics: "{{ (lookup('file', 'data.csv')|trim).split(delimeter) }}"
- set_fact:
len: "{{ (nics|length/rec_len)|int }}"
- set_fact:
nics_list: "{{ nics_list|default([]) + [[
nics[(item*rec_len)|int],
nics[(item*rec_len+1)|int],
nics[(item*rec_len+2)|int],
nics[(item*rec_len+3)|int] ]] }}"
loop: "{{ range(0, len|int)|list }}"
- debug:
msg: "{{ item }}"
loop: "{{ nics_list }}"
gives
ok: [localhost] => (item=[u'Nic1_Vlan', u'Nic1_IP', u'Nic1_MASK', u'Nic1_Scope']) => {
"msg": [
"Nic1_Vlan",
"Nic1_IP",
"Nic1_MASK",
"Nic1_Scope"
]
}
ok: [localhost] => (item=[u'Nic2_Vlan', u'Nic2_IP', u'Nic2_MASK', u'Nic2_Scope']) => {
"msg": [
"Nic2_Vlan",
"Nic2_IP",
"Nic2_MASK",
"Nic2_Scope"
]
}
Is this what you're looking for?