CODEWITHSUNDEEP
linuxbashshellawk
Fattaneh Talebi
Fattaneh Talebi

Reputation: 767

How to retrieve a string which is located after a pattern in bash

I have a large file. I want to retrieve the word that is exactly after this string: "PatterStr()."

Two sample lines:

PatterStr().123232424 hhhhh 9999.  test, test32312
66666666698977.         PatterStr().8888

The output should be:

123232424
8888

when I use grep the whole line will be printed And when two patterns are find in a line, both should be printed, for instance:

PatterStr().123232424 hhhhh 9999.  test, test32312.  PatterStr().11111111
66666666698977.         PatterStr().8888

the correct result:

123232424
11111111
8888

Upvotes: 0

Views: 482

Answers (5)

Claes Wikner
Claes Wikner

Reputation: 1517

With the help of ORS we get a "\n" after each statement on each line.

awk -F'[. ]' 'NR == 1{print $2 ORS $NF}NR == 2{print $NF}' file

123232424
11111111
8888

Upvotes: 0

chden
chden

Reputation: 31

You can reduce the output of grep with the option -o or --only-matching. This will print only the matched parts of a matching line. To suppress the output of PatterStr() you can use a LookBehind as described here.

cat bigfile | grep -Po '(?<=PatterStr\(\)\.)[\w]+'

Upvotes: 3

RavinderSingh13
RavinderSingh13

Reputation: 133428

Could you please try following.

awk '
{
  while(match($0,/PatterStr\(\)\.[0-9]+/)){
    value=substr($0,RSTART,RLENGTH)
    sub(/.*\./,"",value)
    print value
    $0=substr($0,RSTART+RLENGTH)
    value=""
  }
}'  Input_file

Output will be as follows.

123232424
11111111
8888

Explanation of above code: Adding detailed level of explanation for above code.

awk '                                              ##Starting awk program from here.
{
  while(match($0,/PatterStr\(\)\.[0-9]+/)){        ##Starting while loop which has match function to match regex of PatterStr(). till all digits here.
    value=substr($0,RSTART,RLENGTH)                ##Creating variable value which has sub-string value of current line, startin point RSTART tioll RLENGTH.
    sub(/.*\./,"",value)                           ##Substituting everything till DOT with NULL in variable value here.
    print value                                    ##Printing variable value here.
    $0=substr($0,RSTART+RLENGTH)                   ##Setting rest of sub-string value starting from RSTART+RLENGTH to last of line of current line here.
    value=""                                       ##Nullify variable value here.
  }
}'  Input_file                                     ##Mentioning Input_file name here.

Upvotes: 4

V.V.
V.V.

Reputation: 44

There are many ways how you can achieve this, you can for example do it with sed:

sed 's/ /\n/g' text-file.txt | sed -n 's/^PatterStr()\.\(.*\)/\1/p'

The first sed will split the content to separate lines by replacing space by new line, the second will match the lines with PatterStr(). and print what comes directly after it.

Upvotes: 0

Francesco Gasparetto
Francesco Gasparetto

Reputation: 1963

This line does what you need

grep 'PatterStr()' large-file | sed "s/ /\n/g" | grep 'PatterStr()' | cut -f2 -d\.

Output:

123232424
11111111
8888

Upvotes: 0

Related Questions