First version of string literals handling #176
Conversation
| , SFunction <$> (pKeyword "function" *> identifier) <*> (parens (commaSep coreArg)) <*> (symbol "->" *> coreType) | ||
| <*> (symbol "{" *> many coreStmt <* symbol "}") | ||
| , SAssembly <$> (pKeyword "assembly" *> yulBlock) | ||
| , SRevert <$> (pKeyword "revert" *> stringLiteral) |
There was a problem hiding this comment.
Don't know if we need revert builtin in core, but I renamed it to __revert__, because otherwise you cannot define a function called revert
| Just e -> pure (e, []) | ||
| Nothing -> pure (Core.EVar (unwrapId x), []) | ||
| -- special handling of revert | ||
| emitExp (Call _ (Id "revert" _) [Lit(StrLit s)]) = pure(Core.EUnit, [Core.SRevert s]) |
There was a problem hiding this comment.
Is this supposed to be commented?
There was a problem hiding this comment.
Yes, I have disabled special handling of revert.
Another option might be to keep special handling of revert and transform string literals on the Core <-> Yul stage.
But the current typechecker chokes on type literals, so I decided to do it in the preliminar desugaring stage, at least for now.
|
For comnparison, I will try another approach, keeping string literals throughout the frontend and desugaring them in the backend. |
|
I think this can be closed as outdated? |
|
Mostly obsolete now, but parts can be reused in #264 |
2 participants
Quick and dirty string literal handling - literals longer than 32 chars are truncated