exhibit: 1.2.0 -> 1.4.1

[lucasew]

Closes: #369543

• exhibit: 1.2.0 -> 1.4.1

Things done

• Built on platform(s)
  • x86_64-linux
  • aarch64-linux
  • x86_64-darwin
  • aarch64-darwin
• For non-Linux: Is sandboxing enabled in nix.conf? (See Nix manual)
  • sandbox = relaxed
  • sandbox = true
• Tested, as applicable:
  • NixOS test(s) (look inside nixos/tests)
  • and/or package tests
  • or, for functions and "core" functionality, tests in lib/tests or pkgs/test
  • made sure NixOS tests are linked to the relevant packages
• Tested compilation of all packages that depend on this change using nix-shell -p nixpkgs-review --run "nixpkgs-review rev HEAD". Note: all changes have to be committed, also see nixpkgs-review usage
• Tested basic functionality of all binary files (usually in ./result/bin/)
• 25.05 Release Notes (or backporting 24.11 and 25.05 Release notes)
  • (Package updates) Added a release notes entry if the change is major or breaking
  • (Module updates) Added a release notes entry if the change is significant
  • (Module addition) Added a release notes entry if adding a new NixOS module
• Fits CONTRIBUTING.md.

---

Add a 👍 reaction to pull requests you find important.

[ddogfoodd]

nixpkgs-review result

Generated using nixpkgs-review.

Command: nixpkgs-review pr 369684

---

x86_64-linux

❌ 1 package failed to build:

• exhibit

[ddogfoodd]

got this build error on x86_64-linux:

error: builder for '/nix/store/0dff57hx7lsylzw52psnln83b803d6pi-exhibit-1.4.1.drv' failed with exit code 1;
       last 25 log lines:
       > Source dir: /build/source
       > Build dir: /build/source/build
       > Build type: native build
       > Project name: exhibit
       > Project version: 1.4.1
       > Host machine cpu family: x86_64
       > Host machine cpu: x86_64
       > Found pkg-config: YES (/nix/store/0d4m43yp69lrm8imxbqgl9zxjwwz52jw-pkg-config-wrapper-0.29.2/bin/pkg-config) 0.29.2
       > Build-time dependency gio-2.0 found: YES 2.82.1
       > Program /nix/store/y8fxzs8srzd6d74a85zbpssjr8wkj9q4-glib-2.82.1-dev/bin/glib-compile-resources found: YES (/nix/store/y8fxzs8srzd6d74a85zbpssjr8wkj9q4-glib-2.82.1-dev/bin/glib-compile-resources)
       > Program msgfmt found: YES (/nix/store/9a46kza0vbb4vdsdmpzbdc7f31y7hs0i-gettext-0.22.5/bin/msgfmt)
       > Program desktop-file-validate found: YES (/nix/store/1a68w7414qcrvh3dkmyk434jckrzg41s-desktop-file-utils-0.28/bin/desktop-file-validate)
       > Program appstreamcli found: NO
       > Program glib-compile-schemas found: YES (/nix/store/y8fxzs8srzd6d74a85zbpssjr8wkj9q4-glib-2.82.1-dev/bin/glib-compile-schemas)
       > Program python3 found: YES (/nix/store/c9m6yd8fg1flz2j5r4bif1ib5j20a0cy-python3-3.12.8/bin/python3)
       > Configuring exhibit using configuration
       > Program msginit found: YES (/nix/store/9a46kza0vbb4vdsdmpzbdc7f31y7hs0i-gettext-0.22.5/bin/msginit)
       > Program msgmerge found: YES (/nix/store/9a46kza0vbb4vdsdmpzbdc7f31y7hs0i-gettext-0.22.5/bin/msgmerge)
       > Program xgettext found: YES (/nix/store/9a46kza0vbb4vdsdmpzbdc7f31y7hs0i-gettext-0.22.5/bin/xgettext)
       > Could not find file LINGUAS in /build/source/help
       > Program itstool found: NO
       >
       > help/meson.build:1:6: ERROR: Program 'itstool' not found or not executable
       >
       > A full log can be found at /build/source/build/meson-logs/meson-log.txt
       For full logs, run 'nix log /nix/store/0dff57hx7lsylzw52psnln83b803d6pi-exhibit-1.4.1.drv'.
error: 1 dependencies of derivation '/nix/store/4diz0g8kw7vkac646y4pzgq8bk3b7z4k-review-shell.drv' failed to build

[lucasew]

nixpkgs-review result

Generated using nixpkgs-review.

Command: nixpkgs-review pr 369684

---

x86_64-linux

✅ 1 package built:

• exhibit

[ddogfoodd]

builds fine now but still seems to miss some dependency.
When executing I get:

Traceback (most recent call last):
  File "/nix/store/l122xj8im5zgxjcdw21ifb67n3f0ljav-exhibit-1.4.1/bin/.exhibit-wrapped", line 46, in <module>
    from exhibit import main
  File "/nix/store/l122xj8im5zgxjcdw21ifb67n3f0ljav-exhibit-1.4.1/share/exhibit/exhibit/main.py", line 26, in <module>
    from .window import Viewer3dWindow
  File "/nix/store/l122xj8im5zgxjcdw21ifb67n3f0ljav-exhibit-1.4.1/share/exhibit/exhibit/window.py", line 29, in <module>
    from wand.image import Image
ModuleNotFoundError: No module named 'wand'

[Atemu]

Drafting until the issue above is resolved.

[ddogfoodd]

new error when running exhibit:

Traceback (most recent call last):
  File "/nix/store/xxvi52va23nasb7f6rxh4cr866d5rj9j-exhibit-1.4.1/bin/.exhibit-wrapped", line 46, in <module>
    from exhibit import main
  File "/nix/store/xxvi52va23nasb7f6rxh4cr866d5rj9j-exhibit-1.4.1/share/exhibit/exhibit/main.py", line 30, in <module>
    info = f3d.Engine.get_lib_info()
           ^^^^^^^^^^^^^^^^^^^^^^^
AttributeError: type object 'f3d.pyf3d.Engine' has no attribute 'get_lib_info'

[lucasew]

new error when running exhibit:

Traceback (most recent call last):
  File "/nix/store/xxvi52va23nasb7f6rxh4cr866d5rj9j-exhibit-1.4.1/bin/.exhibit-wrapped", line 46, in <module>
    from exhibit import main
  File "/nix/store/xxvi52va23nasb7f6rxh4cr866d5rj9j-exhibit-1.4.1/share/exhibit/exhibit/main.py", line 30, in <module>
    info = f3d.Engine.get_lib_info()
           ^^^^^^^^^^^^^^^^^^^^^^^
AttributeError: type object 'f3d.pyf3d.Engine' has no attribute 'get_lib_info'

I know. I asked upstream for a release because of that.

[lucasew]

It works now but there is a detail.

The app reads the data location using the environment variable XDG_DATA_HOME, which by standard [1] has the default value of $HOME/.local/share, which is missing by default and the program has no fallback, which I added by patching it.

The problem, which doesn't happen in Flatpak, is that it doesn't create a subfolder inside there for that, so it assumes that, in this example, ~/.local/share is the root data folder, which causes a mess. And this is not a problem in the flatpak version because flatpak itself does the scoping and stuff not letting the mess happen.

So there is basically the following strategies:

• Let the package the way it's now in this PR, patching a fallback in to the missing variable and allow the mess to happen.
• Introduce a breaking change upstream that creates a scope folder to store stuff in.
• Introduce the subfolder only locally, not upstreaming the change.
• Introduce upstream the fallback value with the subfolder, so behavior on flatpak wouldn't change but in alternative packages there can be this issue.

I, myself, for these utility graphical packages, am prefering to use the flatpak versions because they are directly maintained upstream so I am inclined to basically just give up on this package and suggest people to just use the flatpak version.

What are your thoughs on this?

[1] https://specifications.freedesktop.org/basedir-spec/latest/

[Atemu]

We should not disregard a user's XDG_DATA_HOME preference and hard code it to some path.

If the upstream code is this simply/dumb, please just patch it to do the right thing instead (create subdir under XDG_DATA_HOME and use that).

[lucasew]

I am not disregarding, I am adding a fallback.

If the variable is defined it will use the original value.

[Atemu]

Ah, I read it wrong.

I'd prefer a patch or at least replace-fail.

[lucasew]

Patch is too hardcoded and replace-fail would become a build failure if eventually upstream fixes that.

[Atemu]

Yes that's the point.

[lucasew]

And bot bumps fail silently

[lucasew]

Basically, r-ryantm only notifies you when everything went right.

[Atemu]

We want to know when the patch no longer applies and needs to be updated. Making the build fail is the only reliable way to achieve that.