Re: [Mesa-dev] Mesa (master): glsl: do not use deprecated bison-keyword

Erik Faye-Lund <[email protected]>
Newsgroups gmane.comp.parsers.bison.general,gmane.comp.video.mesa3d.devel
Message-ID <[email protected]>
On Wed, 2019-05-22 at 18:06 +0200, Akim Demaille wrote:
> Hi Erik,
> 
> > Le 22 mai 2019 à 08:54, Erik Faye-Lund <
> > [email protected]> a écrit :
> > 
> > I ended up reverting the change [from-%error-verbose to %define
> > parse.error verbose] , and I can't find an obcious way to
> > support both old and new versions with the same source file.
> 
> Well, you just disable warnings about deprecated features.
> -Wno-deprecated.
> 
> > The Android SDK also ships a pre-3.x version of Bison, so you're
> > not
> > the only one affected. Also, Chocolatey doesn't provide a new
> > enough
> > version of Bison for Windows either.
> 
> I would be interested in knowing why some distros lag.  Do you
> happen to know why for some of them?
> 

The only clear examples I know of are the Android SDK and Chocolatey.
As I said before, we mighit not have to use the bison version shipped
in the Android SDK, so perhaps that's fixable. And for Chocolatey, it
turns out they have a winflexbison pacakge and a winflexbison3 package,
and the latter has a more recent Bison version.

So looking at this again, it looks like it might be solveable, at least
in the cases where this is visible to me.

Brian: do you think VMWare cuold upgrade the Bison version used in the
forseeable future?

> > The problem is that Bison doesn't seem to have any mechanism for
> > doing
> > statements like these conditionally.
> 
> We try to keep your sources nice and clean, and using %if or
> things like this is anything but nice.  So far these old
> directives are not a maintenance burden, so there is no plan
> to drop support for them.  But still, at least for consistency,
> I promote the newest forms.
> 

Thanks for that clarification, makes sense.

> > The only way around that that I
> > can see is to pre-process the source files somehow. But especially
> > with
> > three different build systems, that's not really a venue I find
> > particularly attractive. If someone can think of a neat solution, I
> > would certainly love to hear about it :)
> 
> Don't do that!
> 

Yeah, I'm not planning on going down that road, as it seems needlessly
convoluted.

> > For now I guess we can just live with the warning. But whenever the
> > Bison devs decide to remove support for %error-verbose, I fear that
> > we'll either have to drop support for older versions or live with
> > pre-
> > processing.
> 
> This should not happen.  What should happen though, is that distros
> stop shipping old versions.
> 

I fully agree.

> [generating the parsers in the tarballs]
> > I don't really think this would work for us in Mesa. We no longer
> > generate the distribution using autotools since we switched to
> > Meson,
> > we just do git archive these days.
> 
> Well, I think that's a pity.  The distinction between end user tools
> and maintainer tools is a useful one.  As exemplified by the issue
> at hand.
> 

I don't think this would have solved the issue for VCS users anyway,
which is a large portion of our users these days, so I'm not sure it
would have helped all that much.


_______________________________________________
[email protected] https://lists.gnu.org/mailman/listinfo/help-bison
lmpx.com only provides a reader for public news (NNTP) servers. It is not affiliated with the servers or forums shown here and is not responsible for the content of articles, which is written by their respective authors.