Open Akumzy opened 2 years ago
Sorry for the late response. I was quite bussy doing other works.
Thanks for the suggestion. It is true that extending is much cleaner, and actually that is what I was doing in the first place. However, I found that extending the existing rules can effects existing rules behavior. (I am trying to remember good example.)
So override the result of existing rules is more predictable.
And for your usecase, since we have beforeAddingTypeModifiers
and afterAddingTypeModifiers
, you can override scalar conversion without messing adding type modifiers
. (with beforeAddingTypeModifiers
).
If you are still curious or I didn't get your point, please provide some detailed example.
I agree that current design is quiet messy. We'll fix this. (Suggestions welcomed.)
Hi all, Couple of days now I have been extending this package to support generating a valid
where
argument that maps to Prismawhere
filter My fork but I have bumped into issues with the test based on how things currently work.For example the by adding
CustomRules
you've automatically overridden the package definedmodifiers
which are very important too.I'm suggesting that
CustomRules
should only extend instead of overriding.