bellard
Revert "Add `JS_HasException()` (#265)"
This patch is not correct because "null" is a valid exception. Currently, the only way to test an exception is to test if JS_EXCEPTION is returned. The content of "current_exception" is undefined otherwise.
Would it be reasonable to make the contents of current_exception defined at all times so one can actually test for pending exceptions?
null is a valid exception, so testing for JS_NULL is not correct even if current_exception is defined all the time (try throw null). We may initialize it to JS_UNINITIALIZED instead.
That's an awfully good point! Initialising it to JS_UNINITIALIZED instead and letting that mean "no pending exception" sounds like a plan, I can put together a PR.
That's an awfully good point! Initialising it to
JS_UNINITIALIZEDinstead and letting that mean "no pending exception" sounds like a plan, I can put together a PR.
Using JS_UNINITIALIZED is indeed a clever way to allow for any valid value to be thrown.
@chqrlie Isn't
JS_UNDEFINEDit's own thing?
Yes, I was confused in my earlier comment. JS_UNDEFINED and JS_UNINITIALIZED are different types, JS_UNINITALIZED is used for special let semantics and is a perfect solution for the issue at stake here.