arai-a

talked with Yoric, and here's summary. regarding IDL versions (and format versions), there are several things to consider. here's the example based on async function' case: # How to extend...

this is troublesome for automation (batch encoding) with non-filter mode.

the issue happens only with non-filter mode, and the automation in m-c is using filter mode only. so not urgent for me.

do you mean they're not created if entropy encoding is used? or you've fixed this issue?

*CheckParameterNames* checks the order in step 5: https://binast.github.io/ecmascript-binary-ast/#sec-checkparameternames and *actualParams* has names in the syntactical order: https://tc39.github.io/ecma262/#sec-function-definitions-static-semantics-boundnames so, if the order doesn't match, it's the reference encoder's bug. I'll check...

moved the issue to binjs-ref. I'd suggest closing this issue.

"random accessible" means that the reader doesn't have to iterate over all statements to reach the function, but just by following the `NodeLink` with 1 step, right? What's the purpose...

Thanks, makes sense. > > Then, is it correct that lazyInnerFunctions should be an exhaustive list of all directly inner functions, with 1:1 match? > > Good question, I went...

> The current PR is not exhaustive in the sense that all actual inner functions be listed in `.innerFunctions`. It's fine to not list some inner functions ahead of time....

having another variant for IfStatement with empty-then-clause sounds interesting. it would be nice to figure out how common it is. Then, about variant byte, if we can (or should?) throw...