I think the main hurdle with xtext is that it uses ANTLR3, and the new
grammar is for ANTLR4.
That said, it might not be too hard to refactor the grammar, and beyond
that, we're not using ANTLR for the tag parsing at all, so we can
continue to use what we've come up with so far.
I think that to address the "ignore output" problem, we just need to add
a bit of logic that either adds what's "inside" to a token that gets
ignored, or flip the token stream on and off at the start/end of the
Should we use Andrew's go at it as the starting place? We can create a
repo under the cfeclipse group if we'd like. Whatever is clever!
Another option that should be considered is just building the "document
model" in cfml.parsing, and then having tools use that as a basis for
doing stuff. Right now cfml.parsing is just sorta walking a tree and
doing stuff, but if we had a model that we populated, we could pass it
around as an abstract kind of deal that anything could leverage.