Conversation
16d7b24 to
ed3e4fa
Compare
|
This PR seems to fix some of the immediate form validation problems I was having (even though the issue you were mentioning this PR in (#113) is about There seems to be one failing test though I am not sure why this is happening though, as right now I didn't dive deep enough into phoenix form handling yet. Maybe it is happening because recent changes in phoenix html/form handling? Furthermore, you mentioned in #113 that there were issues with the sort params as well, specifically with the approach taken in this PR. Did you find some time to have another look at this yet @nduitz ? |
|
Hey there, One thing I noticed yesterday which caused me a lot of headaches was that |
Never mind my comment about |
This fixes two things:
a) It picks up the correct action (not from the changeset but from the parent form)
b) It fixes handling parameters for cardinality :many
If you remove the first commit you will see the test is failing and it will produce the incorrect form.
Instead of generating two forms per polymorphic type it will generate only one form for the
devicetype