Config plist validation about imagr

[clburlison]

I'm not sure if Github sends a message when a previous issue is edited. I've updated the original to update the status of this issue as validation is now present.

[grahamgilbert]

Yes, you can have more than one password key, but only the bottom most one will be used. Let's say we have loaded the configuration plist into the variable, config. What we're effectively doing is setting config['password'] and then setting config['password'] again. I set the password twice in my config plist and then ran this bit of python:

import FoundationPlist

config = FoundationPlist.readPlist(plist)

for key in config:
    print key

And I got

password
workflows

[clburlison]

Alright, that doesn't sound like a show stopper to me. Unless someone else can find something else that breaks the validation process I believe the new script solves this issue.

[gregneagle]

"Yes, you can have more than one password key, but only the bottom most one will be used."

It's probably safer to say that which one will be actually used is undefined. It appears that the CoreFoundation plist parser currently behaves as you've described (processing keys and values in order, so that a later key with the same name as a previous key would cause the key's value to be replaced), but no-one should count on this behavior.