CODEWITHSUNDEEP
bashawksed
achille
achille

Reputation: 315

fetch string between special 2 characters

input: 

random abcD .and$ 5487>1.2.3.4:random fghij & 9101112

output:

1.2.3.4

What I have done so far:

sed -e 's/'>'\(.*\)':'/\1/'

I have also tried: 

sed -e 's/>\(.*\):/\1/'

note: I can do it with awk: 

awk 'BEGIN { FS = ">" } ; { print $2 }' | awk 'BEGIN { FS = ":" } ; { print $1 }'

but it's a little bit heavy, isnt it?

Upvotes: 1

Views: 43

Answers (2)

Ali ISSA
Ali ISSA

Reputation: 408

if there are other chars ":" and ">" in the line, and you are sure that the substring to match contain only "." and digitis [ 0 to 9 ], it will be better to use :

 sed "s/.*>\([0-9|.][0-9|.]*\):.*/\1/g"

or better :

 sed "s/.*>\([0-9|.]\+\):.*/\1/g"

or ( to match the first one occurence ) :

 sed "s/.*>\([0-9|.]\+\):.*/\1/"

you can also use:

  grep -o ">[0-9|.]*:" |  grep -o "[0-9|.]*"

test :

 echo "random > abcD .and$ 5487>1.2.3.4:random : fghij >  & 9101112" |  sed "s/.*>\([0-9|.][0-9|.]*\):.*/\1/g"
 1.2.3.4

 echo "random > abcD .and$ 5487>1.2.3.4:random : fghij >  & 9101112" |  sed "s/.*>\([0-9|.]\+\):.*/\1/"
 1.2.3.4

Upvotes: 0

RavinderSingh13
RavinderSingh13

Reputation: 133680

Solution 1st: This simple sed may help you on same.

sed 's/.*>//;s/:.*//'  Input_file

Solution 2nd: Adding awk solution too now.

awk '{gsub(/.*>|:.*/,"")} 1'  Input_file

Solution 3rd: Using another awk with field separator logic.

awk -F'[>:]' '{print $2}'  Input_file

Solution 4th: Using another sed with back references logic.

sed 's/\(.*>\)\([^:]*\)\(.*\)/\2/'  Input_file

Upvotes: 2

Related Questions