Pipe results of bash script to a text file

Question

I'm having an issue with pipping the results from the below script to a file. When I run the below script, nothing is written to the filecheck_output file.

#!/bin/bash

cd /var/www/html/images/
results="$(find projects -name "*.*" | sort -n)"

echo "${results}" > filecheck_output

The name of the script is filecheck. When I remove the > filecheck_output section from the end of the script and run ./filecheck > filecheck_output from the command line, the script runs and outputs the results into the filecheck_output file without any issues.

Why will the output only be redirected when I run the command from the command prompt and not from in the script?

sorry if this is obvious, but are you looking in /var/www/html/images/ , or your /home dir or /usr/local/bin or wherever filecheck is being stored. It should be in /var/www/html/images/. AND be sure you're not getting a permissions error message when running as a script. Good luck. — shellter
– shellter, Commented Oct 11, 2016 at 12:45

twentyonemore · Accepted Answer · 2016-10-12 05:37:07Z

1

Like others have mentioned, you'll want to put the full path, or relative path of filecheck_ouput into your script.

#!/bin/bash

cd /var/www/html/images/
results="$(find projects -name "*.*" | sort -n)"

echo "${results}" > /home/fedorqui/filecheck_output

answered Oct 12, 2016 at 5:37

twentyonemore

292 bronze badges

Sign up to request clarification or add additional context in comments.

Comments

tripleee · Accepted Answer · 2016-10-12 05:57:43Z

Your terminology is off. You want to redirect output to a file. To pipe something means to take the output from one command and feed it to another, like you are doing in find | sort.

Capturing the results in a temporary variable is simply wasteful, so you are really looking just for

find projects -name "*.*" | sort -n >filecheck_output

The script performs a cd earlier on, so the output file will obviously be created in that directory. If you want to avoid that, either don't cd, or do the cd in a subshell.

( cd somewhere; find projects ) | sort -n >filecheck_output

find somewhere/projects | sort -n >filecheck_output

In the latter case, the output from find will include somewhere in the path of every generated result. You can postprocess with sed to remove it like I show here, although that seems more brittle than the subshell solution.

find somewhere/projects | sed 's%^somewhere/%%' | sort -n

Not doing the redirection in the script itself seems like the best way to make it reusable; then you can choose a different output file each time you run it, if you like, so that makes it usable e.g. in a while loop over a set of directories, or whatever.

Collectives™ on Stack Overflow

Pipe results of bash script to a text file

2 Answers 2

Comments

Comments

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

Comments

Comments

Related