On Oct 20, 2008, at 11:56 AM, Marc Guillemot wrote:
Attila Szegedi wrote:
On Oct 19, 2008, at 1:58 PM, Marc Guillemot wrote:
Attila Szegedi wrote:
...
Yeah, this could work without too much complexity -- just stash
somewhere the last identifier encountered within the current
statement
(null it on statement boundary), and use that as the function name.
thanks to confirm that it wouldn't be a technical problem.
What do you think of such a feature? Would you welcome it in the
code
base?
I would.
great!
I've made some modifications to the parser (locally) to capture the
name
and can save it in the FunctionNode.
What would be for you the best way to make this information
available in
the InterpretedFunction / Script once we have it in the AST?
Honestly, I have no idea. Also, Norris said few times on the list that
Steve Yegge is internally working on some massive AST rework, so I'd
be reluctant to touch the parsing/AST machinery right now 'cause
there's bound to be a bunch of CVS conflict resolution downstream. Of
course, if we'd commit this stuff now, then it'd be *theirs* conflict
resolution :-) For now, I'll give them a bit more benefit of the
patience. That's not to say that if you'd come up with a complete
patch, I wouldn't commit it -- I guess then they'll need to deal with
the conflict resolution and that's it.
Except if Steve maybe listens here and decides to throw in this
functionality into his new code too :-)
Attila.
--
weblog: http://constc.blogspot.com
twitter: http://twitter.com/szegedi
Cheers,
Marc.
--
Web: http://www.efficient-webtesting.com
Blog: http://mguillem.wordpress.com
_______________________________________________
dev-tech-js-engine-rhino mailing list
[email protected]
https://lists.mozilla.org/listinfo/dev-tech-js-engine-rhino
