singular .handler property?

[dherman]

Back in 2012 I suggested a backwards-incompatible tweak to the TryStatement API to have a handler property instead of handlers, and it looks like Esprima's change didn't match it. So now SpiderMonkey and Esprima do not behave compatibly for TryStatements.

The change would have been a more reasonable API but is it way too late for implementations to change to match it?

If we can't change implementations, then the spec should have handlers instead of handler and I'll file a SpiderMonkey bug to match that behavior.

[sebmck]

Related: jquery/esprima#187 and https://code.google.com/p/esprima/issues/detail?id=433

[dherman]

Thanks! From those conversations it looks like it's unlikely anyone is willing to break compatibility on this, but it'd be nice to hear confirmation before closing the question definitively.

[mikesherov]

Esprima is going to change to handler in our next major build.

[dherman]

Great news!

[mikesherov]

@sebmck acorn already does handler?

[sebmck]

@mikesherov Yep

[mikesherov]

Sounds like we can close this and say handler it stays! @dherman close if you're satisfied. Perhaps we should add an errata section or add handlers to deprecated if it isn't already?

[mikesherov]

Ah actually, someone needs to PR this change. @dherman I choose you!

[ariya]

@dherman Are we keeping the empty guardedHandlers for now?

[dherman]

@ariya I filed this in issue #24.