Closed realbigsean closed 8 months ago
awesome! will review tomorrow
fixed a little bug here, but now this is working in lighthouse https://github.com/sigp/lighthouse/pull/5158
Also a limitation right now is we don't have a way to nest inner types that have a subset of variants of the outer type. I think vice-versa should work though (haven't tested it). So nesting an execution payload in a block in lighthouse wouldn't benefit from this for example.
Also a limitation right now is we don't have a way to nest inner types that have a subset of variants of the outer type. I think vice-versa should work though (haven't tested it). So nesting an execution payload in a block in lighthouse wouldn't benefit from this for example.
This actually wasn't so bad to get working, I added it here: 76b67a19fdea613b5f7050e43644544f115aa517
I pushed a bugfix for an issue I noticed with using the field_index
to modify the variant_fields
. If there are earlier fields that aren't on a variant (due to only
), then the field index from the whole struct will not be the location of the field in the variant_fields
. I added a test that fails without my change, and passes with it, but I could do with a second pair of eyes on it
https://github.com/sigp/superstruct/pull/34/commits/52eba9fb1df8bc09628c615dc6117aea01bf1e3a
I also think we want to use the last part of the path, e.g. types::ExecutionPayload
we want ExecutionPayload
not types
: https://github.com/sigp/superstruct/pull/34/commits/30f768f5206858f090b89a93ba02e4038ce0307f
Resolves https://github.com/sigp/superstruct/issues/30