help-bison
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: unmaintained package?


From: Bob Rossi
Subject: Re: unmaintained package?
Date: Tue, 20 Dec 2005 20:19:36 -0500
User-agent: Mutt/1.5.9i

On Wed, Dec 21, 2005 at 01:15:38AM +0100, Odd Arild Olsen wrote:
> 
> I would be pleased to see my patch included into Bison. I used it for a 
> project to implement a state machine in a process control program. I'm not 
> aware of any problem, but am not m4-skilled enough to do the integration with 
> Bison. 

Just out of curiosity, did you ever try to get this included into bison?
or were you just satisfied using the patch?

I did find this in the archives, 
  http://lists.gnu.org/archive/html/help-bison/2004-03/msg00001.html

I think it would be an excellent feature, if bison supported an
asynchronous parse algorithm. It would benefit the entire community. In
my short searches, I've already found dozens of people looking for
exactly this feature. I've even found people looking for this feature
from bison as far as 10 years ago!

What would be the steps I could take to get this integrated into bison?
Akim already told me I need a testsuite, documentation, ChangeLog, and a 
FSF disclaimer.

As far as the testsuite, I noticed bison doesn't use dejagnu like
GDB/GCC does, so is there any documentation into how to change the
testsuite? and what would be an appropriate testsuite change?
I'm guessing you would like to run the new skeleton file through the
current set of tests that the normal skeleton file get's put through.
Does that sound correct?

Odd, you don't have the ChangeLog, but I would be more than happy to 
create that if you don't want to. Also, I would be happy to use the
documentation in README-push as a starting point to modify bison's
bison.texinfo.

Odd, have you registered yet with the FSF? I noticed that the push.c
file already had the FSF notice. I'm assuming your OK with submitting
the copyright over to the FSF. Is that true?

Could the above be a starting point into getting this functionality into
bison?

Finally, if this ends up being an acceptable patch, would there only be
1 skeleton released, as we described? That is, the yyparse () function
would call the yyaparse () function over and over. Or would there simply
have to be 2 skeleton's released, so that if you want asynchronous
functionality out of bison you have to do --skeleton= option?

Thanks,
Bob Rossi




reply via email to

[Prev in Thread] Current Thread [Next in Thread]