UGH #6
UGH #6
Conversation
|
this was previously seen in chefdepartie https://github.com/dalehamel/chefdepartie/blob/master/lib/chefdepartie/databag.rb#L10-L40 |
|
I'm fine with this for now but let's push to get this fixed upstream. |
| @@ -0,0 +1,33 @@ | |||
| # because https://github.com/chef/chef-zero/pull/147/files will never get merged | |||
There was a problem hiding this comment.
Why do you think that PR will never get merged?
|
This makes me 🐼, and I understand why it's required. I agree with @dwradcliffe - let's push to get chef/chef-zero#147 merged. |
|
actually, this won't work with test kitchen because chef-zero isn't running on the host.... hm... |
|
i think we should deal with this on a case by case basis, and convert symbolic references to string references for hash keys... gives us an excuse to fix it when things break. Then again, what about when this breaks silently? Or when an upstream cookbook references data bags assuming they will be a mash? The only alternative i can think of is to include the monkey patch in our cookbooks, assuming that it will work in that context.... but that's horrific. |
@dwradcliffe @andrewjamesbrown
tl;dr chef-zero returns non-mash data for data bags. because it's being braindead and losing the data bag type. It treats the data bag data thus as a plain hash instead of a mash.
Our code depends on it being a mash.
this is dirty, i know, but because of chef/chef-zero#147 this is necessary.