在subprocess.Popen命令中使用变量

Question

Right now I have a test file.dat that I run hexdump on and put the output into a hexdump.dat file.

subprocess.Popen(['hexdump file.dat > hexdump.dat' ], shell=True)  

As a side note, I have seen suggestions to not use shell=True but I essentially get the error OSError: [Errno 2] No such file or directory .

So, I would like to be able to pass in a variable or an array, files, instead of the hardcoded "file.dat". "files" could be a user input or an array/list generated from a previous subprocess section.

I have tried a user input case:

from subprocess import Popen, PIPE, STDOUT 
files = raw_input('File Name: ')                                                                                                 
p = subprocess.Popen(['hexdump files > hexdump.dat' ], stdout=PIPE, stderr=STDOUT)                                         
out,err = p.communicate(input=files)        

Also with:

p = subprocess.Popen(['hexdump',  inputs, ' > hexdump.dat' ], stdout=PIPE, stderr=STDOUT)                                          

Thanks for the help, I know I'm not appropriately understanding the structure needed here so some "handholdy" answers would be appreciated.

Answer 1

Warning: Passing shell=True can be a security hazard if combined with untrusted input. See the warning under Frequently Used Arguments for details.

Something like:

with open('hexdump.dat', 'wb') as f:
    p = subprocess.Popen(['hexdump', 'file.dat'], stdout=f)
    p.wait()

You should read up on Popen and what the shell argument does, and make your decision.

Answer 2

You need shell=True because otherwise it'll look for an executable with that name. shell=True tells the method to use the shell to execute the command so > and friends become what you originally intended them to be (redirection).

The following code you posted:

from subprocess import Popen, PIPE, STDOUT 
files = raw_input('File Name: ')                                                                                                 
p = subprocess.Popen(['hexdump files > hexdump.dat' ], stdout=PIPE, stderr=STDOUT)                                         
out,err = p.communicate(input=files)  

will not work because you're just passing files to hexdump , and if a file with the name files doesn't exist you'll get an error (and if it does exist, it's still probably not what you wanted.)

What you want is to build the string you're executing:

file = "input.dat"
p = subprocess.Popen("hexdump " + file + " > hexdump.dat", shell=True)

Answer 3

Instead of redirecting with > , you can redirect using the stdout param. As for the list of files, you can just append the list of files to an array containing hexdump, ie

myfiles = ['file1','file2']
with open('hexdump.dat', 'w') as output:
    proc = subprocess.Popen(['hexdump'] + myfiles, stdout=output)

Answer 4

I found the simplest way for me to do shell redirection with python and variables is as follows:

subprocess.check_output('svnadmin load %s < %s' % (repo, fname), shell=True)

It can handle very large files.

Answer 5

First off, in regards to file not found, you may need to specify the current working directory.

subprocess.Popen(['hexdump file.dat > hexdump.dat' ], shell=True, cwd='/bar/foo') 

In regards to passing in an array as an argument, that is typically something like:

args = [ 'hexdump', ] + inputs
subprocess.Popen( args, cwd='/foo/bar' )