Keywords trip up the parser in unexpected ways

[danielxmoore]

Keywords trip up the parser in unexpected ways

[edemaine]

Actually, looking at true and false shows the intended solution here: append the negative lookahead assertion NonIdContinue.

Somewhat related, I notice that {true: false} parses fine, so perhaps you can adapt the explicit object literal code to the implicit object literal code.

[STRd6]

Thanks for the bug report, fixed in 0.5.8!

[lorefnon]

Not necessarily recommending we use them, but conventional tokenizers like moo handle this kind of thing well enough that you don't have to think about them when defining parsers at all.

[edemaine]

This should be easy to fix with some negative assertions (like (?!\w), same as what moo does it seems) or \bs. I think I've seen a few places in the grammar where I've wondered whether this is an issue; thanks for specific bug reports!

By the way, you can just write /\bnull\b/ and use $0 as usual. \b doesn't match anything.

[lorefnon]

Thank you