Unable to upgrade R: Error: Failed to configure R: tk must be installed with +x11.
Kenneth Wolcott
kennethwolcott at gmail.com
Sun Dec 22 23:46:48 UTC 2024
Thanks, KenC, for your explanation.
I guess what I need now is to implement your suggestion, but I'm wary that
I'll screw things up royally in so doing.
I don't think I have any particular need to have the +quartz varient of any
of my ports.
It would be nice to have a SQL (or TCL, or whatever) script that would list
all ports that require the +x11 varient and the same of the +quartz
varient, and the ones that will work with either.
Then I could see what port(s) I'll have to "discard" by choosing one way or
the other.
After that is the problem of implementing the decision and trying to
prevent/minimize the chaos resulting from making errors in that process.
I'm not much good at this type of filtering with regards to varients; I
guess there's no better time that now to learn.
Thanks,
Ken Wolcott
On Sun, Dec 22, 2024 at 1:34 PM Ken Cunningham <
ken.cunningham.webuse at gmail.com> wrote:
> Generally speaking, for almost all ports, they are allowed to install as
> either x11 or quartz.
>
> So you should choose which flavour you want, x11 or quartz, and go with
> that.
>
> x11 is more compatible-- because it is (was, anyway, until wayland)
> broadly tested on linux, most or all such ports can install as x11.
>
> quartz is more Mac-like in UI and sometimes in other features, but it is
> somewhat less compatible as not many developers test things as quartz , and
> so certain ports won't build as +quartz, or will be missing features.
>
> To be noted, macports defaults to +x11, and so the macports buildbots are
> set up to build +x11, and so if you go with +quartz, you'll be building
> many ports, and sometimes that can lead to build errors and frustrations.
>
> Once you're decided which flavour you want, if it is +quartz, you should
> probably put that in your /opt/local/etc/variants.conf
>
> And then either way, x11 or quartz, you should probably deactivate all
> your existing ports and install the ones you want -- they will be called in
> with the right flavour. (Well, maybe not _all_ your ports, but at least
> deactivate any ports that have +x11 or +quartz variants installed.)
>
> The issue you are having here is that you have installed some ports with
> the +quartz variant (tk at least) and some ports with the +x11 variant (R),
> so you either need to reinstall R with the +quartz variant or tk with the
> +x11 variant.
>
> And it looks like at some point in the past you had tk installed with the
> +x11 variant, as it's in your list of installed ports, currently
> deactivated.
>
> I personally prefer +quartz variants, so I have my system set up that way,
> and so I build things, and sometimes things won't install, but I live with
> that because I like +quartz.
>
> Ken
>
>
>
> > On Dec 22, 2024, at 12:13 PM, Kenneth Wolcott <kennethwolcott at gmail.com>
> wrote:
> >
> > How to proceed?
> >
> > port installed | grep ' R '
> > R @4.4.1_2+aqua+builtin_lapack+cairo+gcc14+openmp+tcltk+x11
> > R @4.4.1_3+aqua+builtin_lapack+cairo+gcc14+openmp+tcltk+x11
> > R @4.4.1_4+aqua+builtin_lapack+cairo+gcc14+openmp+tcltk+x11
> > R @4.4.2_0+aqua+builtin_lapack+cairo+gcc14+openmp+tcltk+x11 (active)
> >
> > port installed | grep ' tk '
> > tk @8.6.14_0+x11
> > tk @8.6.15_0+quartz (active)
> >
> > Does this mean that I need to have a tk based on x11 to be active?
> >
> > sudo port deactivate tk
> > Note: It is not recommended to uninstall/deactivate a port that has
> dependents as it breaks the dependents.
> > The following ports will break:
> > tkdiff @4.2_0
> > tklib @0.7_0
> > R @4.4.2_0
> > py312-tkinter @3.12.8_0
> > py313-tkinter @3.13.1_0
> > maxima @5.47.0_9
> > Continue? [y/N]: n
> > ---> Cleaning tk
> >
> > Thanks,
> > Ken Wolcott
> > <port_R_failed.log>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.macports.org/pipermail/macports-users/attachments/20241222/25b63c4d/attachment.htm>
More information about the macports-users
mailing list