[Kimchi-devel] [PATCH][Kimchi 3/5] Update REST API

[Rodrigo Trujillo]

Hi Ramon,
yes, it is possible to use ENUM (with a little trick )... please, 
analyze and try this schema:

********************************************************************
     "boot": {
       "description": "Boot order and bootmenu configuration",
       "type": "object",
       "error": "KCHVM0052E",
       "properties": {
         "order": {
           "description": "Boot order",
           "type": "array",
           "error": "KCHVM0053E",
           "items": [
             {"type": "string",
              "enum": ["hd", "cdrom", "network", "fd"]
             },
             {"type": "string",
              "enum": ["hd", "cdrom", "network", "fd"]
             },
             {"type": "string",
              "enum": ["hd", "cdrom", "network", "fd"]
             },
             {"type": "string",
              "enum": ["hd", "cdrom", "network", "fd"]
             }
           ],
           "minItems": 1,
           "additionalItems": false,
           "uniqueItems": true
         }
       },
       "additionalProperties": false,
       "required": ["order"]
     }

********************************************************************

In my tests out of Kimchi it worked , but it accepts "boot" empty 
("minProperties: 1" did not work).
So you have to treat this in the code.

Let me know.

Rodrigo Trujillo


On 06/08/2016 02:46 PM, Ramon Medeiros wrote:
>
>
> On 06/07/2016 10:02 AM, Rodrigo Trujillo wrote:
>>
>>
>> On 06/06/2016 06:13 PM, Ramon Medeiros wrote:
>>> Signed-off-by: Ramon Medeiros <ramonn at linux.vnet.ibm.com>
>>> ---
>>>   API.json | 12 ++++++++++++
>>>   i18n.py  |  2 ++
>>>   2 files changed, 14 insertions(+)
>>>
>>> diff --git a/API.json b/API.json
>>> index 8d4f725..d0e6831 100644
>>> --- a/API.json
>>> +++ b/API.json
>>> @@ -294,6 +294,18 @@
>>>                       "minLength": 1,
>>>                       "error": "KCHVM0011E"
>>>                   },
>>> +                "boot": {
>>> +                    "description": "Boot order and bootmenu 
>>> configuration",
>>> +                    "type": "object",
>>> +                    "error": "KCHVM0052E",
>>> +                    "properties": {
>>> +                        "order": {
>>> +                            "description": "Boot order",
>>> +                            "type": "array",
>>> +                            "error": "KCHVM0053E"
>> Here you can use ENUM and restrict the options accepted. See the 
>> graphics schema a
> I can't use enum, it returns error all time because this patch is to 
> add the boot order, not the first device.
>
> Enum seems to be usable to fixed options, like at VNC. I was not able 
> to use in this case
>> nd:
>> http://spacetelescope.github.io/understanding-json-schema/reference/array.html 
>>
>>
>>> +                        }
>>> +                    }
>>> +                },
>>>                   "users": {
>>>                       "description": "Array of users who have 
>>> permission to the VM",
>>>                       "type": "array",
>>> diff --git a/i18n.py b/i18n.py
>>> index 2d8390f..ec060ae 100644
>>> --- a/i18n.py
>>> +++ b/i18n.py
>>> @@ -103,6 +103,8 @@ messages = {
>>>       "KCHVM0049E": _("Cannot power off %(name)s. Virtual machine is 
>>> shut off."),
>>>       "KCHVM0050E": _("Cannot shutdown %(name)s. Virtual machine is 
>>> shut off."),
>>>       "KCHVM0051E": _("Cannot reset %(name)s. Virtual machine is 
>>> already shut off."),
>>> +    "KCHVM0052E": _("Boot configuration must have order list"),
>>> +    "KCHVM0053E": _("Boot order must be a list of the devices"),
>> I think the messages must be clearer. Like:
>> "Boot configuration must receive a list of devices in the boot order"
>> "Boot order must be a list of devices among ['network', 'hd', ...]"
>>
>>>
>>>       "KCHVM0055E": _("Migrate to localhost %(host)s is not allowed."),
>>>       "KCHVM0056E": _("To migrate a virtual machine to the remote 
>>> host %(host)s the user %(user)s must have password-less login to the 
>>> remote host."),
>>
>> _______________________________________________
>> Kimchi-devel mailing list
>> Kimchi-devel at ovirt.org
>> http://lists.ovirt.org/mailman/listinfo/kimchi-devel
>>
>