[antlr-interest] Re: [antlr-dev] ANTLR 3 Tool behavior

Matt Benson gudnabrsam at yahoo.com
Fri May 26 11:59:22 PDT 2006



--- Terence Parr <parrt at cs.usfca.edu> wrote:

> 
> On May 26, 2006, at 9:59 AM, Matt Benson wrote:
> 
> > This may be best related to the Ant task thread,
> but I
> > started a new one.  I wanted to throw out some
> > thoughts regarding the way the Tool currently
> works,
> > and see if anything came of it... it seems that
> ANTLR
> > is basically designed such that your grammar files
> are
> > in your source tree.  Obviously there are the
> output
> > directory switches, but there is still weirdness
> if,
> > e.g. you have two grammar files, Parser/Lexer and
> > TreeParser, in a directory separate from your
> source
> > tree, and want the source for all three generated
> > elsewhere.  My perspective here may very well be
> naive
> > so any thoughts anyone has on this subject are
> > appreciated.
> 
> Well, you can process one by one pulling from other
> dir or writing to  
> other dir.  You can also do *.g from command line

*.g?  In v3?  This doesn't work for me in 2.x; should
it?

Beyond that, for the scenario I gave above I would
typically make my desired output directory my current
directory and process the files one-by-one.  But since
Java doesn't have a CD (actually you can modify
"user.dir" but that's not such a good idea) it makes
it hard to do this in a single VM.

-Matt


> etc...  Should be  
> able to handle things..
> Ter
> 
> 


__________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around 
http://mail.yahoo.com 


More information about the antlr-interest mailing list