Re: preparing for a new gtk-doc release

[Bastien Nocera <hadess hadess net>]

On Thu, 2019-02-14 at 21:15 +0100, Stefan Sauer wrote:
> On 02/13/2019 10:31 PM, Bastien Nocera wrote:
> > On Wed, 2019-02-13 at 21:58 +0100, Stefan Sauer wrote:
> > > hi,
> > >
> > > did anyone try the new 'no-xslt' toolchain yet? In configure.ac
> > > all
> > > you need to do is:
> > >
> > > -GTK_DOC_CHECK([1.14])
> > > +GTK_DOC_CHECK([1.29],[--flavour no-xslt])
> > The m4 file shows the macro only taking one argument:
> > https://gitlab.gnome.org/GNOME/gtk-doc/blob/master/buildsystems/autotools/gtk-doc.m4#L31
> >
> > GTK_DOC_CHECK([minimum-gtk-doc-version])
> >
> > Or is this not the canonical version of the m4 macro?
>
> It is actually gtkdocize that grabs the extra arg.
> https://gitlab.gnome.org/GNOME/gtk-doc/blob/master/buildsystems/autotools/gtkdocize.in
>
> this is indeed not beautiful.
> > > for meson/cmake I have no idea and would gladly take patches. The
> > Which one of the tools does this "--flavour" argument end up being
> > passed to?
> >
> > Which arguments get passed to which tool, for meson, is documented
> > here:
> > http://mesonbuild.com/Gnome-module.html#gnomegtkdoc
> It not new args, we're calling different tools instead.
>
> meld buildsystems/autotools/gtk-doc.{,no-xslt.}make
>
> Sounds like this needs changes in meson itself? Would be awesome if
> someone could take case of this - the diff of the makefile should be
> straight forward. This would give me more time to work on gtkdoc
> itself.

I filed this meson bug:
https://github.com/mesonbuild/meson/issues/4932

Cheers