grep to sed, append after string match but instead on end of line

Question

I have the following text file with the following lines:

<test="123">
<test="456">
<test="789">

My aim is to have the above text file to be appended with a keyword "HELLO" after the above numbers, as following:

  <test="123.HELLO">
  <test="456.HELLO">
  <test="789.HELLO">

with the grep command and cut, I manage to get the value between the quotation mark.

grep -o "test=".* test.txt | cut -d \" -f2

I tried to use sed on top of it, with this line

grep -o "test=".* test.txt | cut -d \" -f2 | sed -i -- 's/$/.HELLO/' test.txt

however the closest I manage to get is instead a ".HELLO" which directly appended on the end of the line (and not after the numbers in between the quotes)

<test="123">.HELLO 
<test="456">.HELLO 
<test="789">.HELLO

How can I fix my sed statement to provide me with the requested line?

can you clarify: 1) spaces in front of expected output lines is needed or falsely shown here because of formatting 2) does your input file contain lines not containing test=? if so, should they be part of output as well — Sundeep
– Sundeep, Commented Oct 5, 2016 at 15:53

K Richardson · Accepted Answer · 2016-10-05 16:12:41Z

2

You can accomplish this with sed directly. Cut should not be necessary:

grep "test=" test.txt | sed 's/"\(.*\)"/"\1.HELLO"/'

edited Oct 5, 2016 at 16:12

answered Oct 5, 2016 at 13:48

K Richardson

1,7201 gold badge12 silver badges14 bronze badges

Sign up to request clarification or add additional context in comments.

3 Comments

Sundeep Over a year ago

grep is unnecessary too, as well as -o option should not be used in this case.. and you could add some sample filename to grep command...

K Richardson Over a year ago

Good call, edited my answer. I left the grep in, in case there are extraneous lines in the file that the OA wants left out. Of course it could be done with sed directly, but this seems more simplistic.

Sundeep Over a year ago

fine now :) sed -n '/test=/ s/"\(.*\)"/"\1.HELLO"/p' test.txt is simple too, but I do get your point :)

Tamas Rev · Accepted Answer · 2016-10-05 13:54:12Z

1

You can do it with groups in sed. To create new output, you can do this:

 sed 's/\(test="[^"]*\)"/\1.HELLO"/g' test.txt

To modify it in-place, you can use the -i switch:

 sed -i 's/\(test="[^"]*\)"/\1.HELLO"/g' test.txt

Explanation:

() is a group. You can refer to it with \1. In sed we have to escape the parentheses: \(\)
[^"]* matches everything that's not a quote. So the match will stop before the quote
In the replacement, you have to add the quote manually, since it's outside of the group. So you can put stuff before the quote.

answered Oct 5, 2016 at 13:54

Tamas Rev

7,1745 gold badges37 silver badges52 bronze badges

Comments

User9102d82 · Accepted Answer · 2016-10-05 13:53:49Z

1

Try this:

This is how your file looks like.

bash > cat a.txt
<test="123">
<test="456">
<test="789">

Your text piped to SED

bash > cat a.txt |sed 's/">/.HELLO">/g'
<test="123.HELLO">
<test="456.HELLO">
<test="789.HELLO">
bash >

Let me know if this worked out for you.

answered Oct 5, 2016 at 13:53

User9102d82

1,1901 gold badge10 silver badges23 bronze badges

Comments

zxy · Accepted Answer · 2016-10-05 14:41:37Z

1

awk 'sub("[0-9]+","&.HELLO")' file

answered Oct 5, 2016 at 14:41

zxy

1581 silver badge2 bronze badges

Collectives™ on Stack Overflow

grep to sed, append after string match but instead on end of line

4 Answers 4

3 Comments

Comments

Comments

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

4 Answers 4

3 Comments

Comments

Comments

Comments

Your Answer

Sign up or log in

Post as a guest

Related