Re: [updated PATCH] %language declaration

[Joel E. Denny]

On Thu, 7 Dec 2006, Paolo Bonzini wrote:

> http://www.mail-archive.com/address@hidden/msg01308.html
> http://www.mail-archive.com/address@hidden/msg01325.html
> http://www.mail-archive.com/address@hidden/msg01336.html

Thanks.

> If I understood the first two comments well, Akim doesn't want
> language-dependent code to pollute the Bison parser.  And I tend to agree.

Ok, but I don't get the connection.  Which of these do you think better 
summarizes it?

1. Bison should never support a language with a different basic syntax 
than C/C++ since we don't want Bison to have to parse such a language in 
actions.

2. Bison might one day support such a language but Bison should parse 
actions for them horribly incorrectly.  Thus, we'll refuse to have 
%language now to ensure that no one tries to parse them correctly then.

Obviously it's not #2.  If it's #1, what's the harm in supporting 
%language now?  That doesn't prevent us from opposing specific languages 
later.  Then again, if we're not willing to parse any other syntax, I 
guess that means we won't be adding very many other languages.  Is there 
another way of looking at this that I'm missing?

Sorry to force you to play devil's advocate, but I feel like I'm missing 
the point.

> The comments in Akim's last message are addressed by this revision of 
> the patch.  The %skeleton mechanism is not touched at all.

I agree with this point.