Feng Zhichao: add cloudstack model for list virtual machines
Merge branch 'master', remote-tracking branch 'fog/master'
Feng Zhichao: add cloudstack server save function
Feng Zhichao: add get methods in cloudstack node
Feng Zhichao: add destroy function in cloudstack server model
Feng Zhichao: add key_pair model to cloudstack with crud functions
Feng, these changes limit the information available from the list virtual machines command and will break existing implementations because you are using parsers to alter the response.
I am all for adding models, but I am against adding parsers into the cloudstack space since the responses are already json formatted. I'd suggest making your models work without changing (and in the process limiting) the response format.
@mingjin I'm with @bdorry on this. If there is json available I would much prefer that we just use that over adding parsers. If you have any questions on that feel free to let us know though and we'd be happy to help out in terms of getting this fixed up. Thanks!
[shilei] add group for list node parser
use cloudstack json response to create server models.
remove parse for deploy virtual machine in cloudstack.
get correct node id from json response after deploying virtual instance
remove useless response parser for cloudstack.
Hi, @bdorry @geemus, we've updated the code according to your suggestion, could you help review them? Thanks
@mingjin - sorry for the delay, not sure how this fell through the cracks.
@bdorry - could you check it out and see how this fits in (I realize due to the duration of time passed that it may make less sense and/or not work now). Thanks!
@bdorry hey, any chance you could take a look at this? I know it has been a little while, but would be good to figure out if this is worth rebasing and trying to get it in. Thanks!
@dm1try - could you review this? Thanks!
@geemus hey, CRUD for server was already implemented, surprising :).
Maybe a key_pair model worth something, but here we have some deprecated things like a connection and we don't have any tests.
Original work is 2 years old, despite the recent reviews.
Closing since it is too stale in my view.