Re: symbol_type::token () removed?

Hello Akim,

I am trying with 3.5.92 and I see now:

/Users/kumbhar/nmodl/src/parser/diffeq_context.cpp:81:75: error: no member
named
      'type_get' in 'nmodl::parser::DiffeqParser::by_kind'
        if (token_type ==
DiffeqParser::by_type(DiffeqParser::token::END).type_get()) {
                          ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ^

The full example is here
<https://github.com/BlueBrain/nmodl/blob/584af62be9cd6f9017f63e2d9cb732e1da58d420/src/parser/diffeq_context.cpp#L81>.
Did* I* miss to point this out in my last email?

Next time if you push the fix to the git repo, I can build bison from there
and try it.

Regards,
Pramod

On Sun, May 3, 2020 at 9:19 AM Akim Demaille <[email protected]> wrote:

>
> > Le 3 mai 2020 à 09:10, pramod kumbhar <[email protected]> a
> écrit :
> >
> > Hello Akim,
> >
> > Thank you very much for quick response and addressing this already!
>
> Sure :)
>
> > By the way, I was also confused when brew started shipping 3.5.91. I was
> trying to find out if 3.5.91 is stable release or not and came across this
> and this. By looking at the release tarballs here, it's not obvious which
> one is beta vs which one is stable. Typically when we package new version
> inside Spack, we often look at the release tarballs and add new version
> (even automatically find one). I agree it's not good idea but I wouldn't be
> surprised if such tarballs are perceived as stable releases. If this is
> beta version, may be tarball name can indicate that?
>
> Heck, the mistake is mine...  I uploaded it as a stable release, instead
> of a beta :(
>
> So we should better hurry up and wrap 3.6 quite soon :(
>
> Thanks!