[antlr-interest] vocabulary (token) import and export

[jsrs701]

Hi gang,

I am confused.  I recently decided to bust up my mammoth .g file into 
three different files, one each for lexer, parser, and tree walker.

Well, I'm running into vocab hell from this.

What I want to do is have the lexer define some tokens that it uses 
and exports.

Then I want to have the parser define some tokens that IT uses and 
exports.

Then I want the tree walker to use all of those.

Unfortunately, it looks like by the time the parser is compiled, it 
has only exported the tokens that it uses, and left out all the 
tokens that only the lexer uses.  This is unfortunate, since then all 
the C++ files are compiled, and the Lexer.cpp file can't find a bunch 
of symbols that it needs!

Ack!

So anyone else have this problem?  What's the solution?  By the way, 
take it as a given that I cannot compile the C++ files right after 
compiling the .g file.

Thanks,
JSRS


 

Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/