run_resfinder.py aborts on git error

Running run_resfinder.py fails with this exception:

Traceback (most recent call last):
  File "/usr/src/ext/resfinder/run_resfinder.py", line 34, in get_git_tag
    subprocess.check_output(
  File "/opt/conda/lib/python3.9/subprocess.py", line 424, in check_output
    return run(*popenargs, stdout=PIPE, timeout=timeout, check=True,
  File "/opt/conda/lib/python3.9/subprocess.py", line 528, in run
    raise CalledProcessError(retcode, process.args,
subprocess.CalledProcessError: Command '['git', 'describe', '--tags', '--abbrev=0']' returned non-zero exit status 128.

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "/usr/src/ext/resfinder/run_resfinder.py", line 181, in <module>
    parser.add_argument("-v", "--version", action="version", version=get_git_tag(),
  File "/usr/src/ext/resfinder/run_resfinder.py", line 45, in get_git_tag
    raise Exception(str(exc_info.output))
Exception: b'fatal: not a git repository (or any of the parent directories): .git\n'

This is apparently due to it calling git describe, and failing because it is not in a git repository at runtime.

Scripts shouldn't assume they are in a git repository when invoked. This is the case in the development environment, but nowhere else. Once a python module is installed (pip, setup.py), or deployed in a Docker container or in a Conda env, the .git subdirectory won’t be there.

I’d suggest sticking to the standard way to do this: set a variable __version__ = x.y.z in the module’s __init.py__, so it is available globally, and remember to update this when you release (you could even automate this in a commit hook).

PS: note that ResFinder’s .git history is nearly 200MB(!), so you really don’t want that in a Docker container.

Comments (4)