How to search for a string in linux - linux

Question

How to extract only /app/xxxx field in the below lines and add df -P -T command to the extracted strings and save the output in the below format
edlp_nps_app:x:23449:5000:EDLP_NPS_APP (HP):/app/edlp_nps_app:/bin/bash
genxp_app:x:23414:15887:GENXP_APP (HP):/app/genxp_app:/bin/bash
icegnapp:x:21697:15954:ICEGNAPP (HP):/app/icegnapp:/bin/bash
icegnftp:x:21554:15416:ICEGNFTP
(HP):/app/icegnftp:/usr/libexec/openssh/sftp-server
df -P -T /app/XXXXX
df -P -T /app/edlp_nps_app
O/P:
Filesystem Type 1024-blocks Used Available Capacity
Mounted on
/dev/mapper/rootvg-rootvol ext4 144365708 27057836 110769428 20% /
Output i require is FS, type, Mounted on, appname
/dev/mapper/rootvg-rootvol ext4 / /app/edlp_nps_app
I tried it awk command but it didnt work

Answer 1

Using awk test if 6th field starts with /app, if true print 6th field
awk -F':' '$6~/^\/app/{print "df -P -T "$6}' infile | bash
To save in file
awk -F':' '$6~/^\/app/{print "df -P -T "$6}' infile | bash > outfile
To avoid printing of header multiple times you may pipe another awk like below
awk -F':' '$6~/^\/app/{print "df -P -T "$6}' infile | bash | awk 'FNR==1{print;next}/Filesystem/{next}1' >outfile
Test Results :
input :
$ cat file
edlp_nps_app:x:23449:5000:EDLP_NPS_APP (HP):/app/edlp_nps_app:/bin/bash
genxp_app:x:23414:15887:GENXP_APP (HP):/app/genxp_app:/bin/bash
icegnapp:x:21697:15954:ICEGNAPP (HP):/app/icegnapp:/bin/bash
icegnftp:x:21554:15416:ICEGNFTP
(HP):/app/icegnftp:/usr/libexec/openssh/sftp-server
Output:
$ awk -F: '$6~/^\/app/{print "df -P -T "$6}' file
df -P -T /app/edlp_nps_app
df -P -T /app/genxp_app
df -P -T /app/icegnapp

Answer 2

extract /app/xxx
first we drop the lines that do not contain /app
sed '/\/app/!d'
then we extract the /app/...
's|.*\(/app/[^:]*\).*|\1|'
combined:
sed '/\/app/!d;s|.*\(/app/[^:]*\).*|\1|'
Output i require is FS, type, Mounted on, appname
/dev/mapper/rootvg-rootvol ext4 / /app/edlp_nps_app
Let's tweak the df's output a little bit by having it supply only the relevant fields
df /app/foo --output=source,fstype,target
I do not believe we can get df to drop the header, so we strip it
df /app/foo --output=source,fstype,target | sed '1d'
I will append the app name manually and put it all together in a script named "magic.sh":
#!/bin/sh
if [ ! -f "$1" ]
then
echo "Input file does not exist!"
exit 1
fi
if [ -f "$2" ]
then
echo "Output file already exists!"
exit 2
fi
sed '/\/app/!d;s|.*\(/app/[^:]*\).*|\1|' $1 | while read -r app
do
echo "$(df $app --output=source,fstype,target | sed '1d') $app" >> $2
done
Assuming the specified input is in a file named "input" and you want the result in a file named "output", this is how to call it:
sh magic.sh input output