Question

I am trying to run a comparison between the output of a command called by subprocess.check_output(). Since I am running this on Windows I am also getting \r\n‘s in the output (which is a good thing).

Now I want to compare that output from that command with a text file. This fails because the open() does not preserve the \r‘s. Here is what I got so far:

try:
    output = subprocess.check_output(paramList, universal_newlines=False,
                                     stderr=subprocess.STDOUT)
except subprocess.CalledProcessError as err:
    output = err.output
    errorCode = err.returncode

with open(EMCMD_INCORRECT_PARAMS, 'r') as fd_usage:
    usageLines = fd_usage.read()
    usage = True if usageLines == output else False

Any suggestions to preserve the \r‘s? Thanks!

Answer 1

In python 3, open takes an newlines argument to control the behaviour of the newline translator (see the docs). A values of '' for this argument disables the newline translation.

---

If you can’t use python 3, file objects opened by open have a newline attribute that contains the original type of newlines, so you can use this to get the original content :

with open(EMCMD_INCORRECT_PARAMS, 'r') as fd_usage:
    usageLines = fd_usage.read().replace('\n', fd_usage.newline)

Binary mode might also be a solution if you don’t like the idea of post-processing what read returns to get the original content of the file.