One very common task when scripting with bash is to use a for loop to iterate over the contents of a directory or directory tree. There are two primary methods of accomplishing this task; using ls
and using find
. We’ll not consider the manual method as that would be completely unworthy of our attention.
I find it easy to start with ls
when I don’t need to recurse into a directory tree as that is a command that I use often. This often turns into a process such as this:
for dir in $(ls) do echo ${dir} done
Now the above method typically does not work for me. I have an alias setup to print out pretty colors when I issue the ls
command and that will cause each command which operates on the variable $dir
to fail with a “No such file or directory” error. I always have to remember this and re-write the command with the flag to disable color formatting:
for dir in $(ls --color=never) do echo ${dir} done
The above script will work every time.
The next option is using find
. find
is awesome and all powerful. Learn and use find
. The most common issue when using find is that you may have to filter out the current and/or parent directories when processing the results. Take this example:
for dir in $(find . -maxdepth 1 -type d) do echo ${dir} done
This loop will print out the current directory, as well as all other directories in the current working directory. If you are running some sort of processing within this loop, you may end up re-processing everything unless you discard the current working directory (noted by the dot).
This example will not process the current working directory:
for dir in $(find . -maxdepth 1 -type d) do if [ ${dir} == "." ] then continue fi echo ${dir} while pushd ${dir} do echo ${dir} done popd done
Bash for loops are incredibly useful and easy to work with. Use the above tips and make bash work for you.
Leave a Reply