Notes about Bash.
Run command on each line in a file
Run the [command] on each line in a [file]:
while read l; do [command] $l & done < [file]
This will run [command] [line n] & for every line in [file], where [line n] is the next line of the file on each call.
E.g…
hello.txt:
hello
world
me@mine ~ $ while read l; do echo $l & done < hello.txt;
[1] 5048
hello
[2] 5049
world
Reference: Looping through the content of a file in Bash?
Show details of directory
Normally ls -l a_directory will show a details list of the contents of a_directory. To show the details of the directory itself:
ls -ld a_directory
Run script with a variable taken from lines in a file
#!/bin/sh
while read line
do
ls -l $line
done < "filenames.txt"
List only hidden directories
ls -bd .*/
List only hidden directories excluding ./ and ../
ls -bd .*/ | egrep -v '^.?./'
egrep '^.?./' matches ./ and ../ while the -v option inverts the match.
Grep contents of files in hidden directories
for i in $(ls -bd .*/ | egrep -v '^.?./'); do grep -r "the string" $i; done
Counting files/directories
Note: These examples don’t include hidden files/directories.
Files and directories: ls -l | sed 1d | wc -l
Files only: ls -l | sed 1d | grep -v ^d | wc -l
Directories only: ls -ld */ | wc -l
The grep -v ^d does an inverted match on directories (i.e. keeps lines that don’t begin with a d).
The sed 1d removes the first line - the total - from the top of the list.
Order directories by disk usage
See Order directories by disk usage in my du notes.
Loops
(This example updates and initialises all submodules in a git repository with git 1.5)
FILES=`find . -type f -name ".gitmodules"`
for f in $FILES
do
pushd `dirname $f`
git submodule update --init
popd
done
