I would like to monitor the output of a command and save parts of its output as it prints using a filter. Using grep would be a way to filter the output but I can only get it to show the matches, but it's not what I want.
In my case I am running curl on a list of URLs and would like to see its results, while only saving the results that has a status code of 000 or 200 to a text file.
Use tee /dev/stderr to duplicate output.
for example:
seq 9 | tee /dev/stderr | grep "[42]" > result.txt
seq 9 outputs 1 to 9, tee /dev/stderr duplicate output to stderr, grep "[42]" to filter output and save to file result.txt
or, using process substitution
seq 9 | tee >(grep "[42]" > result.txt)
We can treat >(grep "[42]" > result.txt) as a file. So tee >(grep "[42]" > result.txt) will duplicate output to stdout and save to this special file. But the file is a process, which is grep "[42]" > result.txt.
