August 2018
Intermediate to advanced
366 pages
10h 14m
English
You might be wondering why we configured logging to send its output to stderr, instead of the standard output. This allows us to separate the output of our software (which is written to stdout through the print statements) from the logging information. This is usually a good practice because the user of your tool might need to call the output of your tool without all the noise generated by logging messages, and doing so allows us to call our script with something such as the following:
$ python dosum.py 2>/dev/null 8 16 5 0
We'll only get back the results, without all the noise, because we redirected stderr to /dev/null, which on Unix systems leads to throwing away all that was written to stderr.