Arch Linux User Repository

I'm still not sure what the use-case for virtualfish without fish would be, so I just tried using it in bash with this result.

bash $ vf
VirtualFish isn't compatible with bash, only Fish 3.1.0+.

The upstream author also says that it requires fish>=3.1 to work, so I will keep the fish dependency in the package until someone comes up with a convincing reason to remove it.

What I did is moving fish to a runtime dependency. Because what @ruahcra said is true. The package doesn't need fish to build, it only needs it to run. I hope this solves your problem.

I ended up just making virtualfish-git and removed the fish dependency check in its PKGBUILD

Since fish isn't needed to build this package then you can just remove the check:

https://wiki.archlinux.org/index.php/PKGBUILD#depends An array of packages that must be installed for the software to build and run.

I think their use of epoch is mistaken. Looks like it has had it ever since the first commit of their PKGBUILD, but their use doesn't appear to be valid according to https://wiki.archlinux.org/index.php/PKGBUILD#epoch

EDIT: Actually I guess it is (or was) valid since the versioning scheme appears to have changed in 2012, but I wonder if it can be removed now...

fish-git seems to have an epoch in the version. I have no idea how to handle this. :( I'll try to find out later today after work.

Somehow fish>=3.1 isn't playing nicely with fish-git:

>git clone ssh://aur@aur.archlinux.org/virtualfish.git
>cd virtualfish/                                                
>makepkg PKGBUILD
==> Making package: virtualfish 2.5.0-1
==> Checking runtime dependencies...
==> Missing dependencies:
  -> fish>=3.1
==> Checking buildtime dependencies...
==> ERROR: Could not resolve all dependencies.
>pacman -Qo fish
/usr/bin/fish is owned by fish-git 2:3.1.2.r1526.gf73ee3011-1

Removing the xdg dependency seems sensible to me. If you want to be fully compliant with the XDG spec, the replacement line would be

INSTALL_DIR = os.path.join(os.getenv('XDG_CONFIG_HOME', os.path.expanduser('~/.config')), 'fish', 'conf.d')

I completely removed the damn package now. It's useless anyway. it basically installs and imports a package just to get a path that is fixed anyway. Feels like the javascript debacle all over again. Should work now.

Virtualfish' Python script doesn't seem to be able to pickup the vendored xdg.py

>vf ls
Traceback (most recent call last):
  File "/usr/bin/vf", line 6, in <module>
    from pkg_resources import load_entry_point
  File "/usr/lib/python3.8/site-packages/pkg_resources/__init__.py", line 3259, in <module>
    def _initialize_master_working_set():
  File "/usr/lib/python3.8/site-packages/pkg_resources/__init__.py", line 3242, in _call_aside
    f(*args, **kwargs)
  File "/usr/lib/python3.8/site-packages/pkg_resources/__init__.py", line 3271, in _initialize_master_working_set
    working_set = WorkingSet._build_master()
  File "/usr/lib/python3.8/site-packages/pkg_resources/__init__.py", line 584, in _build_master
    ws.require(__requires__)
  File "/usr/lib/python3.8/site-packages/pkg_resources/__init__.py", line 901, in require
    needed = self.resolve(parse_requirements(requirements))
  File "/usr/lib/python3.8/site-packages/pkg_resources/__init__.py", line 787, in resolve
    raise DistributionNotFound(req, requirers)
pkg_resources.DistributionNotFound: The 'xdg<5.0,>=4.0' distribution was not found and is required by virtualfish

>ll /usr/lib/python3.8/site-packages/virtualfish/loader/xdg.py
-rw-r--r-- 1 root root 3.7K Apr  7 10:43 /usr/lib/python3.8/site-packages/virtualfish/loader/xdg.py

I added my 2 cents to the pull request upstream. Good idea btw.

It's amazing how much trouble this one line of code is causing down the road. If the package still complains about checksums try deleting the cached package. I had to do that because yay doesn't seem to recognize new files with unchanged filename.