I have a python script that resides on a remote server, under version control, and I'd like to execute it from my local shell.
I know that curl https://remote.path/script.py | python will work (as confirmed here) when there are no additional parameters.
The problem is, I can't figure out how to pass in additional command line arguments, e.g. python script.py arg1 arg2 arg3?
I recognize this is may not be the most secure practice, but the script is pretty benign.
If you check a manual page you will see that the python command takes either a script or the character -. The - argument instead of a script is used to tell the Python command that the script should be read from standard input. So using that all other arguments are know to be arguments to the script.
Used like
$ curl https://remote.path/script.py | python - arg1 arg2 arg3
man python would have answered your question:
python [ -B ] [ -d ] [ -E ] [ -h ] [ -i ] [ -m module-name ]
[ -O ] [ -OO ] [ -R ] [ -Q argument ] [ -s ] [ -S ] [ -t ] [ -u
]
[ -v ] [ -V ] [ -W argument ] [ -x ] [ -3 ] [ -? ]
[ -c command | script | - ] [ arguments ]
Say:
curl https://remote.path/script.py | python - arg1 arg2 arg3
Example:
$ cat s
import sys
print sys.argv[1:]
$ cat s | python - arg1 arg2 arg3
['arg1', 'arg2', 'arg3']
Very easily (and not recommended) you can just catch them in python using sys.argv
In file import sys
print sys.argv[1]
In terminal
python myfile.py foo
foo
If you do
print sys.argv[1:]
As another response suggested you'll get
["foo"]
argparseoroptparse?