bug-bison
[Top][All Lists]
Advanced

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

Re: bison'd files don't compile under Microsoft VC++


From: David Gluss
Subject: Re: bison'd files don't compile under Microsoft VC++
Date: Wed, 13 Feb 2002 19:29:48 -0800

In general, you're so right.  But in practice, in this case,
you're talking about the compiler that probably 90% or
more of the developers are using.  Hey, I don't know if it's
90, or 99, or whatever...I just know it's not anything like
75.  This is also not the "compiler of the week".  It's been
like this for quite a long while (years).

Although #ifdef is considered harmful, I don't know any
actual products built without it. I don't know any java products
that don't use cpp. But I'm getting off topic.  My point is
that the code in its current state does not work for most
people.  You can fix that with a few lines of #ifdef...
if you choose to.

If I can't convince you that this is actually a good idea, can
you recommend alternative parser generators that are portable?

DG
----- Original Message ----- 
From: "Paul Eggert" <address@hidden>
To: <address@hidden>
Cc: <address@hidden>
Sent: Wednesday, February 13, 2002 5:40 PM
Subject: Re: bison'd files don't compile under Microsoft VC++


> > From: "David Gluss" <address@hidden>
> > Date: Wed, 13 Feb 2002 17:13:49 -0800
> > 
> > Considering how trivial the patch is, and how pervasive the
> > Microsoft C compiler is, I'm surprised that this seems such a hard
> > problem.
> 
> The general problem is hard because there are a lot of C compilers out
> there, the code is supposed to run on them all (and ideally on C++
> compilers too), and we have no real way to test them.
> 
> We should not make a change just for Microsoft's compiler-of-the-week.
> If we take that approach, we'll need to make a change "just for Sun",
> another change "just for IBM", another change "just for older
> Microsoft compilers", etc., and the code will turn into even worse
> spaghetti than it is.
> 
> For more on this subject, please see the classic paper "ifdef
> Considered Harmful" by Spencer and Collyer.
> 
> _______________________________________________
> Bug-bison mailing list
> address@hidden
> http://mail.gnu.org/mailman/listinfo/bug-bison
> 




reply via email to

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