[antlr-interest] Re: [antlr-dev] ANT build cruft?

[Matt Benson]

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

> 
> On May 17, 2006, at 1:19 AM, Jim Idle wrote:
> 
> >
> >
> >
> > On 5/16/06 8:25 PM, "yawl.20884287 at bloglines.com"
> > <yawl.20884287 at bloglines.com> wrote:
> >
> >> Hi Terence,
> >> Almost all projects I know put them in the main
> directory.
> >>
> >
> > Though, the ones I use with netbeans seem to be
> able to live  
> > anywhere in
> > fact so long as you set the root paths and so on
> and use variables  
> > for the
> > locations.
> >
> > This reminds me though that the current ANT task
> for V2 is not quite
> > compatible with the one required for V3. I started
> to look at it,  
> > but there
> > are few comments and it seems that you have to
> know more about ANT  
> > than I
> > want to. So, it seems we need to find a volunteer
> to update the V2  
> > ANT task
> > for V3?
> 
> I guess you're right!  Ewwww...i dont' want to get
> too close to it  
> lest the "stank" wear off me ;)
> 
> Ter
> 

I can't commit to writing the task... I may do it
anyway, I just can't officially commit to it as I'm
extremely busy lately.  What I would like to see:

- ANTLR hosts the antlr3 task.  It is just sensible
for non-core tasks to be hosted by the related product
or an interested third party (Scott? ;).  It might
look cool packaged as an antlib, so that a user would
commonly declare an "antlr" namespace and invoke

<antlr:tool> or whatever the v3 idiom is.

- The new task addresses the shortcomings of the
existing apache-hosted task.  I don't remember what
they are actually; having gotten frustrated with that
task many years ago I have actually used <java
class="antlr.Tool"> or the commandline since.  Can any
current users offer input here (except for that silly
"ANTLR task should create its own output dir" stuff).

- Oh, yeah, the current task has to parse the grammar
files to figure out the destination file(s) for a
given grammar.  Having this info determined by some
kind of "official" ANTLR utility code would be nice. 
I imagine all codegen authors have to determine this
info somehow; could ANTLR3 make this info available
via an API call (does it already)?

It'd be nice to make the new task as slick as
possible.  If I had enough info to start with I could
probably slap something together pretty quickly.

-Matt

> _______________________________________________
> antlr-dev mailing list
> antlr-dev at antlr.org
> http://www.antlr.org/mailman/listinfo/antlr-dev
> 


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