[bug#50627] [PATCH 0/2] Make wayland-protocols dependency native-input.

[Liliana Marie Prikler]

Hi

Am Freitag, den 17.09.2021, 05:35 +0300 schrieb muradm:
> Liliana Marie Prikler <liliana.prikler@gmail.com> writes:
> 
> > Hi,
> > 
> > Am Donnerstag, den 16.09.2021, 22:23 +0300 schrieb muradm:
> > > wayland-protocols is not runtime dependency and only build time
> > > dependency for applications that directly using wayland.
> > Guix does not distinguish between "build time" and run time
> > dependencies.
> True, here issue could be related to miss wording, but same 
> wording is used in the manual as well, so do I.
I'll respond to that in your quote below.

> > > Initially I tought that making wayland-protocols a 
> > > native-inputs dependency as it should, it would reduce number of
> > > dependants on it. But it turns out other way around. With this
> > > patchset we are fixing gtk+ to not advertise it as dependency in
> > > its .pc files, and moving wayland-protocols to native-inputs
> > > where it should be.
> > That's not what native-inputs are used for.  native-inputs 
> > provide binaries that the host/build machine needs to run in order
> > to compile a package.  It doesn't seem to be the case that wayland-
> > protocols is such a package, is it?
> wayland-protocols is different package. It does not include any 
> binaries only protocol specifications (some xml files), which are
> used for code generation. We could consider them as a kind of
> autoconf/bison like inputs, but tightly scoped for wayland needs,
> although they are not so and not binaries.
And what kind of code is generated from them?  I would assume it's
target code.  And since wayland-protocols is no tool to process those
XML files, but the files themselves, I'd hazard a guess that it should
rather be built for the target.  While currently this appears to make
no difference, there might well be a time in which those files differ
for some two architectures, which then would cause problems in cross-
compiling contexts were it a native input.

> > > Patch provided for gtk+ also merged with upstream.
> > > 
> > > Patchset prepared from core-updates-frozen. While it seems that
> > > it will impact many other packages, actually this patch reduces
> > > number of packages that touches wayland-protocols and probably
> > > avoids it at runtime.
> > But it still impacts a large number of packages in ways that 
> > could
> > potentially break and haven't been tested, right?
> Technically, this package does not change anything in terms of 
> binary producing. wayland-protocols remains to be an input as it was 
> before. I.e. wayland compositor, wayland application, wayland using 
> library, application which uses wayland using library, binary output
> is not impacted. If binary output is the same, is there any thing
> else to test?
In that case I'd hazard a guess that it's fine, but the phrase
"wayland-protocols remains to be an input" is perhaps a bit weird given
the change to native-input.

> There are two things which are being changed. First as you 
> pointing out is the way Guix treats it, i.e. reducing closure, etc.
> Second is propagation of inputs. Currently (without this patch),
> since it is listed in propagated-inputs (and also advertised in .pc
> files), wayland-protocols as requirement, needlessly, getting pushed
> down then hierarchy.
We ought to move it from propagated-inputs to inputs and either (if we
can) ignore pkg-config or patch the pkg-config files.  W.r.t. pkg-
config I do wonder whether Requires.private needs propagation, though,
it normally should be just Requires.

> Let's take 4 cases that we have here (I do not pretend to be 
> complete, of course, there are might be more levels/combinations,
> just attempting to illustrate current case in simple words/terms):
> 
> 1. wayland compositor (weston, wlroots/sway, etc.)
> 2. wayland client application (grim, mpv, etc. applications 
> directly interacting with wayland interfaces)
> 3. wayland client library (qt or gtk+ in this case, also directly
> interacts with wayland to abstract it for user applications)
> 4. user application of wayland client library (in this case some 
> gtk+ based application)
> 
> For 1 and 2, both types should have to specify wayland in inputs 
> (or propagated-inputs), and wayland-protocols in native-inputs.
Why?

> One of purposes to have layer 3, is to abstract from 1 and 2.
> i.e. when I write gtk application, as user I should not be aware 
> of where/how this application is going to run, via xorg or wayland. 
> Then why I should be aware of wayland/wayland-protocols and make
> sure that it is provided as build input for my application?
IIUC you don't need to be aware when gtk propagates the input?  It's
similar to how you still need an Xorg server to test your GTK
application.

> More over, if I will have some other unrelated package that 
> depends on my gtk application (item 4 above), i still will see
> wayland-protocols among my inputs.
> 
> Currently, thanks to Guix, it is getting resolved by having it 
> listed in propagated-inputs.
> 
> For the long run, it was also fixed in gtk, so that 
> wayland-protocols is not going to be advertised in gtk's .pc files
> any more (https://gitlab.gnome.org/GNOME/gtk/-/merge_requests/3960
> and https://gitlab.gnome.org/GNOME/gtk/-/merge_requests/3961).
Which is fine in and of its own, but not the right thing w.r.t. native-
inputs.

> I suppose that, initially wayland-protocols was listed in
> propagated-inputs for this same reason, because gtk was 
> advertising it in .pc files.
Probably.

> > While reducing closure size is generally a good thing, I think we
> > do need to be careful whenever "build time vs. run time" and native
> > vs. non-native are confused.
> I'm using terminology as per documentation :) may be it should be
> reworded in some other way to avoid confusion. 8.2.1 package 
> reference:
> 
>           ‘native-inputs’ is typically used to list tools needed 
>           at build time, but not at run time...
You're quoting the manual out-of-context and (accidentally) misuse the
word tool.

>           When cross-compiling, dependencies listed in ‘inputs’ are
>           built for the _target_ architecture; conversely, 
>           dependencies listed in ‘native-inputs’ are built for the 
>           architecture of the _build_ machine.
This is the distinction to make here.  "Typically used to list tools" 
here means that the package provides a tool (i.e. a binary) that you
invoke at some point of your recipe.  This can be a compiler like GCC,
a tool to create Makefiles like automake, or an X server to launch
tests in.  The only thing in that regard when talking about wayland
would be the wayland-scanner tool provided by the wayland package.  

Notice the contrast to what you said before with wayland being an input
and wayland-protocols being a native one.  If you need wayland-scanner
for you build, it should be a native-input (as well as an input,
probably).  If this does become a problem later on, a bin/lib split for
wayland might make sense.

Regards