Alex Beregszaszi

> We should use a pragma to activate the standard library behaviour on a per-file basis. That is how I started initially (see the commit history), but the the symbol...

Oops, forgot that struct getters return the members, and not the struct itself. That is super unintuitive.

``` Error: Overriding public state variable return types differ. --> struct-interface.sol:10:5: | 10 | S public override s; | ^^^^^^^^^^^^^^^^^^^ ``` Could also consider displaying the return types in this...

> In loops with many iterations, or many array index accesses, or nested loops this might really matter. I don't like paying gas for unneeded checking, which is most of...

> Is there a simple way available today to do an unchecked read? Even if it involves an inline-assembly sload. Short loops (n < 10) are not uncommon, and sload...

@nventuro it is not planned atm. Also created a new issue to discuss range based loop specifically (#10162), given this issue is a different discussion.

Also there's this comment from @chriseth on the previous discussion https://github.com/ethereum/solidity/issues/9054#issuecomment-641194378: > @nventuro are you talking about the check in `++i`? In the new code generator, the compiler should be...

Should we consider C++-style literals and use them in expressions, such as `"stringtobehashed"_keccak256` instead of the built in function?

> I think we need to support all possible kinds of bytecode's. > > e.g. ir, irOptimized, legacy and legacyOptimized. Maybe also wast, ewasm and ewasm.wast? I guess that is...

I think that is the question where we'll find all the subtle details and complications. I'd say the practical decision right now is that storage in synchronised at the point...