Python sys.stdin and subprocess stdout difference

Question

I currently have a larger problem at hand so I started by trying to match the smaller differences to see if that'll fix my issue.

In my program, I was using a pipe and iterating through that input through sys.stdin. I noticed it's type is <class '_io.TextIOWrapper'> . I'm trying to avoid using a pipe and replaced my code to use subprocess.run() instead and noticed that the result has the type <class 'str'> instead.

This could be a really stupid question but I'm wondering why they're different and if I can get the subprocess stdout to be the same type as sys.stdin.

Using Python 3.7.5

Answer 1

You are comparing apples and oranges. The contents of sys.stdin are str instances (though you can configure it to return bytes , somewhat painstakingly; subprocess on Python 3.x returns bytes unless you specify text=True or similar to decode those bytes into a str ).

Python 3.7.2 (default, Jan 29 2019, 13:41:02) 
[Clang 10.0.0 (clang-1000.10.44.4)] on darwin
Type "help", "copyright", "credits" or "license" for more information.
>>> import sys
>>> type(sys.stdin)
<class '_io.TextIOWrapper'>
>>> line = sys.stdin.readline()
fnord
>>> type(line)
<class 'str'>
>>> import subprocess
>>> s = subprocess.run(['true'], capture_output=True, text=True)
>>> type(s)
<class 'subprocess.CompletedProcess'>
>>> type(s.stdout)
<class 'str'>

Answer 2

subprocess.run returns a CompletedProcess instance, with the collected stdout/stderr captured all at once.

If you want a stream, create a Popen instance, which will have stdout and stderr attributes that act like sys.stdin , and a stdin attribute that acts like sys.stdout . But beware pipe buffering/deadlock problems if you do anything fancy.