Include has_one attributes correctly when calling to_params #262
Conversation
…arams Include has_one attributes correctly when calling to_params
|
Great job. Would you mind merge this code into your own fork? I would like to point to your fork/master. |
|
👍 |
|
@remiprev Besides this now causing conflicts, what was the reason this was never merged? This is causing pain for us and I'd really like to see this on |
|
@dturnerTS Ah, didn't see that and didn't realise that @remiprev was looking for a new maintainer. Thanks for the heads-up! |
|
@dturnerTS i'm working to get Her down to 0 PRs. given there is already has_many serialization, i think it makes sense to do the same for has_one associations. few things before we merge:
|
…s_in_to_params * upstream/master: (54 commits) Provide a more ActiveRecord-like interface by aliasing new_record? to new? Do not check if the record is persisted to get the default value. Add a failing test for a belongs_to association with nil foreign key in a non-persisted record. Fix belongs_to fetch Fix find through an association Fix parsing associations (Parse#extract_array) Add specs for testing associations with active_model_serializers use `request_path` in specs instead of internal `build_request_path` Fixes remi#357 specify the code block as ruby Escape path variables Parse has_one association data before initializing objects acquire a mutex when defining attribute methods use ActiveModel convention for defining attribute methods define all attribute methods inside generated_attribute_methods add specs for when attribute methods predefined make possible to spawn_model from superclass Update release notes for info on security fix Bump version to 0.7.6 Loosen activemodel, activesupport dependencies to < 4.3 to accommodate rails security fixes ...
…t the case where associations[:has_one] or associations[:has_many] is nil
|
@hubert Thanks for looking into this again. |
|
@dturnerTS thanks for taking the time to go through this again. can you rebase this instead of merging it? |
|
I'm not a big fan or rebasing. Can you explain why you think its preferable to the master merge here? |
5 participants
Follow on to remiprev#206 to add support for has_one in to_params.
Also fixes remiprev#258