Arch Linux User Repository

Errors

Traceback (most recent call last):
  File "/usr/bin/conda-build", line 33, in <module>
    sys.exit(load_entry_point('conda-build==3.20.5', 'console_scripts', 'conda-build')())
  File "/usr/bin/conda-build", line 25, in importlib_load_entry_point
    return next(matches).load()
  File "/usr/lib/python3.9/importlib/metadata.py", line 77, in load
    module = import_module(match.group('module'))
  File "/usr/lib/python3.9/importlib/__init__.py", line 127, in import_module
    return _bootstrap._gcd_import(name[level:], package, level)
  File "<frozen importlib._bootstrap>", line 1030, in _gcd_import
  File "<frozen importlib._bootstrap>", line 1007, in _find_and_load
  File "<frozen importlib._bootstrap>", line 986, in _find_and_load_unlocked
  File "<frozen importlib._bootstrap>", line 680, in _load_unlocked
  File "<frozen importlib._bootstrap_external>", line 790, in exec_module
  File "<frozen importlib._bootstrap>", line 228, in _call_with_frames_removed
  File "/usr/lib/python3.9/site-packages/conda_build/cli/main_build.py", line 12, in <module>
    from glob2 import glob
ModuleNotFoundError: No module named 'glob2'

If you see something like this, the "error" is that AUR packages are not rebuilt when the system Python version is upgraded, which just happened again with 3.9. You need to manually upgrade each and every Python package from AUR, unfortunately. That is quite a caveat, because not everyone notices that something specific like Python is upgraded when they run pacman -Syu and see a list of 200 or more packages. This also happens with Node.js and what not.

Solution

So I have just bumped the package here as well as python-conda-build, which triggers updates with e.g. yay -Su. However, some dependencies may not have a new version now. What we could do for this case is to increase pkgrel for those other packages. It feels hacky, but at least it works.

Addendum

I am sorry for the inconvenience. Now the remaining problem is: Because of the bunch of dependencies and dependencies of dependencies, it is easy to miss one of them. =) Secondly, I do not own all the dependencies, so some will need manual maintenance either way, unfortunately. For example, you will need to reinstall python-glob2-git. Please let us know here if anyone of you runs into issues and we will try to help.

Notes

To rebuild, you need to pass the --rebuild option.

Example: yay -S python-glob2-git --rebuild

@specter119 Would you make /etc/conda/condarc as backup in PKGBUILD?

Errors

Traceback (most recent call last):
  File "/usr/bin/conda-build", line 33, in <module>
    sys.exit(load_entry_point('conda-build==3.20.5', 'console_scripts', 'conda-build')())
  File "/usr/bin/conda-build", line 25, in importlib_load_entry_point
    return next(matches).load()
  File "/usr/lib/python3.9/importlib/metadata.py", line 77, in load
    module = import_module(match.group('module'))
  File "/usr/lib/python3.9/importlib/__init__.py", line 127, in import_module
    return _bootstrap._gcd_import(name[level:], package, level)
  File "<frozen importlib._bootstrap>", line 1030, in _gcd_import
  File "<frozen importlib._bootstrap>", line 1007, in _find_and_load
  File "<frozen importlib._bootstrap>", line 986, in _find_and_load_unlocked
  File "<frozen importlib._bootstrap>", line 680, in _load_unlocked
  File "<frozen importlib._bootstrap_external>", line 790, in exec_module
  File "<frozen importlib._bootstrap>", line 228, in _call_with_frames_removed
  File "/usr/lib/python3.9/site-packages/conda_build/cli/main_build.py", line 12, in <module>
    from glob2 import glob
ModuleNotFoundError: No module named 'glob2'

If you see something like this, the "error" is that AUR packages are not rebuilt when the system Python version is upgraded, which just happened again with 3.9. You need to manually upgrade each and every Python package from AUR, unfortunately. That is quite a caveat, because not everyone notices that something specific like Python is upgraded when they run pacman -Syu and see a list of 200 or more packages. This also happens with Node.js and what not.

Solution

So I have just bumped the package here as well as python-conda-build, which triggers updates with e.g. yay -Su. However, some dependencies may not have a new version now. What we could do for this case is to increase pkgrel for those other packages. It feels hacky, but at least it works.

Addendum

I am sorry for the inconvenience. Now the remaining problem is: Because of the bunch of dependencies and dependencies of dependencies, it is easy to miss one of them. =) Secondly, I do not own all the dependencies, so some will need manual maintenance either way, unfortunately. For example, you will need to reinstall python-glob2-git. Please let us know here if anyone of you runs into issues and we will try to help.

Notes

To rebuild, you need to pass the --rebuild option.

Example: yay -S python-glob2-git --rebuild

@monson I will try to fix the error follow the python-conda in Fedora. It seems like fix all bugs in porting conda as a native program under Linux's package manager, and then update to 4.9.2, Thanks for your attention.

@specter119 python3.9 is out, and https://aur.archlinux.org/packages/python-conda/#comment-769183 happens again. would you update to 4.9.2 and make all the files in /etc/profile.d linked?

@monson will be fix in the next update in recent days.

Run /usr/bin/conda shell.zsh hook failed for

    FileNotFoundError: [Errno 2] No such file or directory: '/usr/lib/python3.8/site-packages/conda/shell/etc/profile.d/conda.sh'

ln -s /etc/profile.d/conda.sh /usr/lib/python3.8/site-packages/conda/shell/etc/profile.d/conda.sh fix it, would you add this is PKGBUILD?

@leuko You need create a environment first. The 'base' environment of python-conda is system python, which is not actually controlled by python-conda.

I get the following error when I issue conda install:

NoBaseEnvironmentError: This conda installation has no default base environment. Use
'conda create' to create new environments and 'conda activate' to
activate environments.

Is a preconfiguration needed or is this a packaging issue?

PS: My goal is to install conda on a multi-user environment for students that they can install personal conda packages.