CODEWITHSUNDEEP
linuxshellcommand-linescripting
Mammadj
Mammadj

Reputation: 153

How to replace Pipe with a new line in Linux?

Please, accept my apologies, if this question was asked before. I am new and do not know how to do it. I have a file containing the data like this:

name=1|surname=2|phone=3|email=4
phone=5|surname=6|name=7|email=8
surname=9|phone=10|email=11|name=12
phone=13|email=14|name=15|surname=6

I would like to have a file like this:

name=1
name=7
name=12
name=15

Thanks in advance!

Upvotes: 1

Views: 5819

Answers (2)

Walter A
Walter A

Reputation: 19982

I would go for the solution of @Lars, but I wanted to test this with "lookbehind". With grep you can get the matches only with grep -o, but the following line will also find surname:

grep -o "name=[0-9]*" names.txt

You can fix this a little by looking for the character before name (start of line with ^ or |).

grep -o "(^|\|)name=[0-9]*" names.txt

What a fix! Now you get the right names, but sometimes with an extra |.
With \K (and grep option -P) you can tell grep to use something for the matching but skip it during output.

grep -oP "(^|\|)\Kname=[0-9]*" names.txt

Upvotes: 0

Lars Fischer
Lars Fischer

Reputation: 10139

Say names.txt is your file, then use something like :

cat names.txt  | tr "|" "\n" | grep "^name="
  • tr transforms | to newlines
  • grep filters for the lines with name

And here is a one command solution with GNU awk:

 awk -v RS="[|\n]" '/^name=/' names.txt
  • the -v RS="[|\n]' set the record separatro to|` or newline
  • the /^name=/ filters for records starting with name= (and implicitly prints them)

Upvotes: 2

Related Questions