Convert results from digitalocean objects to dicts and lists

[nmaludy]

Fix for #2

This parses the custom digitalocean objects and converts them into dicts and lists so StackStorm can understand them.

[nmaludy]

Before This Fix

[root@stackstorm ~]# st2 run digitalocean.get_all_droplets
.
id: 5955c366a814c00bdd89a618
status: succeeded
parameters: None
result: 
  exit_code: 0
  result: None
  stderr: ''
  stdout: ''

After This Fix

[root@stackstorm st2]# st2 run digitalocean.get_all_droplets
.
id: 5955cc86a814c00bdd89a63c
status: succeeded
parameters: None
result: 
  exit_code: 0
  result:
  - action_ids: []
    backup_ids: []
    backups: false
    created_at: '2017-04-18T20:11:18Z'
    disk: 20
    end_point: https://api.digitalocean.com/v2/
    features: []
    id: 12345
    image:
      created_at: '2017-03-03T16:10:57Z'
      distribution: Ubuntu
      id: 1234
      min_disk_size: 20
      name: LAMP on 14.04
      public: true
      regions:
      - nyc1
      - sfo1
      - nyc2
      - ams2
      - sgp1
      - lon1
      - nyc3
      - ams3
      - fra1
      - tor1
      - sfo2
      - blr1
      size_gigabytes: 0.5
      slug: lamp
      type: snapshot
    ip_address: x.x.x.x
    ip_v6_address: null
    ipv6: false
    kernel: null
    locked: false
    memory: 512
    monitoring: null
    name: sendy
    networks:
      v4:
      - gateway: x.x.x.x
        ip_address: x.x.x.x
        netmask: 255.255.255.0
        type: public
      v6: []
    next_backup_window: null
    private_ip_address: null
    private_networking: false
    region:
      available: true
      features:
      - private_networking
      - backups
      - ipv6
      - metadata
      - install_agent
      name: New York 2
      sizes:
      - 512mb
      - 1gb
      - 2gb
      - 4gb
      - 8gb
      - 16gb
      - 32gb
      - 48gb
      - 64gb
      slug: nyc2
    size:
      available: true
      disk: 20
      memory: 512
      price_hourly: 0.00744
      price_monthly: 5.0
      regions:
      - ams2
      - ams3
      - blr1
      - fra1
      - lon1
      - nyc1
      - nyc2
      - nyc3
      - sfo1
      - sfo2
      - sgp1
      - tor1
      slug: 512mb
      transfer: 1.0
      vcpus: 1
    size_slug: 512mb
    snapshot_ids: []
    ssh_keys: []
    status: active
    tags:
    - emailer
    - spamtester
    token: ******
    user_data: null
    vcpus: 1
    volume_ids: []
    volumes: []
  stderr: ''
  stdout: ''

[Kami]

LGTM,

But some test cases for function which converts those custom objects would be nice :)

[Kami]

I think this would also throw if an object contains an attribute which is datetime class instance or instance of another class which is not JSON serializable, right?

[nmaludy]

Yes, that's correct.

Any other thoughts or idiomatic ways of converting custom python objects into dicts/lists?

[Kami]

Nothing simple I can think of.

This should work for now, as long as the class doesn't contain attributes which are not JSON serializable.

Otherwise the best way would probably be to manually build up the dict and exclude those attributes (we already do this in some other places like aws pack, etc.).

[nmaludy]

@Kami I took some time and wrote unit tests for the functionality i added. When doing this i noticed there were no unit tests at all for this pack so i went ahead and did a first pass at some tests. Unfortunately i'm not too familiar with this pack or with Digital Ocean so any comments/suggestions are appreciated.

[Kami]

Great, thanks for working on the tests - will have a look shortly :)

[Kami]

You can also add this to contributors field (array) in pack.yaml :)

[Kami]

Merged - thanks.

Will also add you as a contributor to this repo :)