LGTM with an extremely small suggestion that you are welcome to ignore.
It is interesting and convenient that so many of the commands have
(seemingly) reliably parsable output. The branch has nice and
interesting tests. Thank you!
[later] Oh, it's YAML, yeah? I guess that's OK. I think it would be
nice to add a comment to that effect, but others might disagree. Just a
suggestion.
LGTM with an extremely small suggestion that you are welcome to ignore.
It is interesting and convenient that so many of the commands have
(seemingly) reliably parsable output. The branch has nice and
interesting tests. Thank you!
https:/ /codereview. appspot. com/14441074/ diff/5001/ quickstart/ app.py
File quickstart/app.py (right):
https:/ /codereview. appspot. com/14441074/ diff/5001/ quickstart/ app.py# newcode49 app.py: 49: retcode, output, error = utils.call('juju',
quickstart/
'status', '-e', env_name)
I'm OK with this for expediency, but wouldn't the API (Kapil's library)
be better in a revision?
[later] Oh, it's YAML, yeah? I guess that's OK. I think it would be
nice to add a comment to that effect, but others might disagree. Just a
suggestion.
https:/ /codereview. appspot. com/14441074/ diff/5001/ quickstart/ app.py# newcode70 app.py: 70: 'juju', 'api-endpoints', '-e', env_name,
quickstart/
'--format', 'json')
yay, --format json
https:/ /codereview. appspot. com/14441074/ diff/5001/ quickstart/ utils.py
File quickstart/utils.py (right):
https:/ /codereview. appspot. com/14441074/ diff/5001/ quickstart/ utils.py# newcode69 utils.py: 69: # Switch to using "juju switch --raw" in order
quickstart/
to avoid the fragility
oh, cool.
https:/ /codereview. appspot. com/14441074/