validating the grammar

# Michael Dyck (13 years ago)

Allen Wirfs-Brock wrote:

Another concern is that the entire grammar really needs to be revalidated after major changes. The last time the grammar was validated was for ES3 (1999) and we don't currently have the tools or people in place to validate it again (any volunteers out there??).

What exactly do you mean by "validate the grammar"?

In the past, you mentioned "the committee's desire to have a generally LR(1) grammar", so I'm guessing "validation" = "confirmation that the grammar is [generally] LR(1)". Is it anything beyond that? Are the results of the ES3 grammar-validation archived somewhere? (Or are the results simply "Yes"?)

# Brendan Eich (13 years ago)

Michael Dyck wrote:

Allen Wirfs-Brock wrote:

Another concern is that the entire grammar really needs to be revalidated after major changes. The last time the grammar was validated was for ES3 (1999) and we don't currently have the tools or people in place to validate it again (any volunteers out there??).

What exactly do you mean by "validate the grammar"?

See lxr.mozilla.org/js/source/js2/semantics -- this is Waldemar's ES3-to-original-JS2/ES4-era grammar and semantics checking code, written in Mac Common Lisp.

In the past, you mentioned "the committee's desire to have a generally LR(1) grammar", so I'm guessing "validation" = "confirmation that the grammar is [generally] LR(1)". Is it anything beyond that? Are the results of the ES3 grammar-validation archived somewhere? (Or are the results simply "Yes"?)

I recall Waldemar saying his code checks for unambiguous LR(1) after applying the lookahead restrictions and handling restricted productions ([no LineTerminator here] rules for ASI). He said it checks other properties of the grammar, but I can't recall details right now. Cc'ing him.