Add support for typecasting in mapping generation on cli init #1534
+96
−21
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
This issue has been raised before (#949, #1017), but when generating a new subgraph for a contract that has events with
address[]
inputs, the generated mappings won't compile properly.This is due to an issue where arrays of derived types (
Address[]
in this case) are not assignable to arrays of base types (Bytes[]
here).I wasn't sure how to fix this on a lower level as I'm not familiar with AssemblyScript, so I went and implemented a fix where the mapping assignments generated for events on
init
are conditionally wrapped withchangetype<typeB>(...parameter.of.typeA...)
iftypeA
is not able to be implicitly cast totypeB
.This got a bit messy and I had to extend some functions since we also must add imports for the types we are typecasting as (e.g.
Bytes
), and these must be calculated before we iterate events inside the template string.I tested this locally and the generated mappings compile, but I'm not too sure about events with tuples, since they were a special case to begin with. A solution that doesn't do typecasting, but improves the base types or better uses AssemblyScript's existing type system is probably superior to my approach.