Dear Rohit,

a key thing I miss in Cloudmonkey is a way to reference and use
previous outputs. (But maybe I am missing something and this is
already possible, through 'shell' commands or something?)

Very often with CloudStack I need to find what exists for a type of
resource, reference one of these by its uuid (or a string value for a
name or description...), and put the uuid reference into a new
Cloudmonkey command.

It would be useful to have a *modest and lightweight* way of doing
this. There is no point to implement a comprehensive way of doing
this, which duplicates what is already possible with cloudmonkey
commands inside bash shell, or API calls in Python or etc.

I imagine say '%' refers to the last output as a json object, '%%' the
second last, '%%%', ... and so on for a few levels back (it is
limited, maybe 5). I can extract data out of the json using list
position numbers or keys (in Python style).

Then I could type something like:

>list zones

>list virtualmachines zoneid=%[4]['id']

>list virtualmachines id=%[3]['id']

>list networks id=%['nic'][0]['networkid']

and maybe a way to print out data:

> print %['cidr']

So it is lightweight. It is not for tasks that I want to turn into a
script but something quick and for immediate use.

Thanks, Phillip

---------- Forwarded message ----------
From: Rohit Yadav <[email protected]>
To: "[email protected]" <[email protected]>
Cc:
Date: Thu, 4 Dec 2014 17:15:14 +0530
Subject: Any more CloudMonkey improvements you want to see?
Hi all,

We want to work on the next release of CloudMonkey do you have any
issues that we should fix, or any new feature you want? Thanks.

--
Regards,
Rohit Yadav
Software Architect, ShapeBlue
M. +91 8826230892 | [email protected]
Blog: bhaisaab.org | Twitter: @_bhaisaab

Reply via email to