Implement less restrictive env module merge #1107
Conversation
reidmv
force-pushed
the
module_conflicts-env-setting
branch
2 times, most recently
from
8edf6a2 to
bc955aa
Compare
By default, allow a module defined in an environment to override a copy defined in a Puppetfile, rather than hard-failing the deploy. This allows for a more frictionless path to using environment modules.
reidmv
force-pushed
the
module_conflicts-env-setting
branch
from
bc955aa to
50acb3f
Compare
reidmv
force-pushed
the
module_conflicts-env-setting
branch
from
a684ae6 to
d6264cd
Compare
Co-authored-by: Molly Waggett <mwaggett@alumni.reed.edu>
reidmv
force-pushed
the
module_conflicts-env-setting
branch
from
d6264cd to
8c097fe
Compare
|
@mwaggett Thanks so much for the detailed review! I think I've implemented all the suggestions. Ready for re-review. |
| case conflict_opt = @options[:module_conflicts] | ||
| when 'override_puppetfile' | ||
| logger.debug log_msg | ||
| when 'override_puppetfile_and_warn', nil |
There was a problem hiding this comment.
I didn't notice this before, but what's the nil here for?
There was a problem hiding this comment.
If the setting isn't specified, the option is nil. Default behavior. I don't think r10k has a nice way of using schemas or other systems for this kind of thing? Could be wrong. Any suggestions?
There was a problem hiding this comment.
ahhhh, gotcha. I think it's fine, I'm just a little out of practice at reading ruby haha 😅
I had skimmed and thought that the first case was the default, so wasn't sure what this was doing here in the middle, but I see now.
|
@dhollinger Would you mind taking another look, since there have been changes since your review? |
By default, allow a module defined in an environment to override a copy
defined in a Puppetfile, rather than hard-failing the deploy.
This allows for a more frictionless path to using environment modules.
This commit changes default behavior of an experimental feature.