问题
Possible Duplicate:
Python subprocess wildcard usage
Using the Python 2.6 subprocess module, I need to run a command on a src.rpm file that I am building with a previous subprocess call.
Unfortunately, I am working with spec files that are not consistent, so I only have a vague idea of what the filename of the src.rpm should look like (for instance, I know the name of the package and the extension in something named "{package}-{version}.src.rpm" but not the version).
I do know, however, that I will only have one src.rpm file in the directory that I am looking, so I can call mock with a command like
mock {options} *.src.rpm
and have it work in shell, but subprocess doesn't seem to want to accept the expansion. I've tried using (shell=True) as an argument to subprocess.call() but even if it worked I would rather avoid it.
How do I get something like
subprocess.call("mock *.src.rpm".split())
to run?
回答1:
Use the glob package:
import subprocess
from glob import glob
subprocess.call(["mock"] + glob("*.src.rpm"))
回答2:
The wildcard * has to be interpreted by the SHELL. When you run subprocess.call, by default it doesn't load a shell, but you can give it shell=True as an argument:
subprocess.call("mock *.src.rpm".split(), shell=True)
来源：https://stackoverflow.com/questions/14482135/shell-expansion-in-python-subprocess