[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: Interactive parsing with Bison
From: |
Frans Englich |
Subject: |
Re: Interactive parsing with Bison |
Date: |
Wed, 28 Jun 2006 19:32:31 +0000 |
User-agent: |
KMail/1.8.50 |
On Wednesday 28 June 2006 19:06, Akim Demaille wrote:
> Le 28 juin 06 à 19:24, Richard Stallman a écrit :
[...]
> a question to which you already answered no, years ago (but the
> answer, arguably, depends on the moment the question is asked).
[...]
> Similarly, another time related question: can we drop the
> (hardly used, barely supported by maintained tools) VCG
> graph format in favor of the (wide spread) dot format?
> One or two years ago, you said that the release of Graphviz
> as free-software was too recent to give enough confidence
> that that move would be smart.
I jump in here with something I always have wondered:
By lurking on the various bison lists I think you have a *massive* struggle
with tons of backwards-compatibility issues. This can be old Bison behavior,
yacc compatibility, and so on.
Why don't you decide to separate development into two "branches"?
The first would simply be Bison as it is now. Conservative, backwards
compatibility, and so on.
The second would be a "cleaned" and "modern" version of Bison. It could
feature a rename(where deemed useful) of all keywords to provide a consistent
interface, consistent error/warning handling, drop all legacy code,
consistent/practical extension mechanisms, use modern language features, and
so on. It would require users to do source changes in order to use
Bison "3.0", but I think they would be few and I also think many would be
willing to do it.
In that way Bison would be able to develop in an unlimited way(and a way
attractive to developers), while the "old school" would still be there for
those who prefers that.
I know too little about Bison to be able to tell whether it's a good idea(I
don't even know if it sounds naive), but it feels like it is something that's
badly needed.
Cheers,
Frans
- Re: Interactive parsing with Bison, (continued)
- Re: Interactive parsing with Bison, Satya, 2006/06/23
- Re: Interactive parsing with Bison, Bob Rossi, 2006/06/23
- Re: Interactive parsing with Bison, Hans Aberg, 2006/06/23
- Re: Interactive parsing with Bison, Akim Demaille, 2006/06/27
- Re: Interactive parsing with Bison, Frans Englich, 2006/06/27
- Re: Interactive parsing with Bison, Satya, 2006/06/27
- Re: Interactive parsing with Bison, Akim Demaille, 2006/06/27
- Re: Interactive parsing with Bison, Akim Demaille, 2006/06/27
- Re: Interactive parsing with Bison, Richard Stallman, 2006/06/28
- Re: Interactive parsing with Bison, Akim Demaille, 2006/06/28
- Re: Interactive parsing with Bison,
Frans Englich <=
- Re: Interactive parsing with Bison, Satya, 2006/06/28
- Re: Interactive parsing with Bison, Frans Englich, 2006/06/28
- Re: Interactive parsing with Bison, Hans Aberg, 2006/06/29
- Re: Interactive parsing with Bison, Frans Englich, 2006/06/29
- Re: Interactive parsing with Bison, Hans Aberg, 2006/06/29
- Re: Interactive parsing with Bison, Richard Stallman, 2006/06/29
- Re: Interactive parsing with Bison, Paul Eggert, 2006/06/30
- Re: Interactive parsing with Bison, Richard Stallman, 2006/06/30
Re: Interactive parsing with Bison, Richard Stallman, 2006/06/24