[SOLVED] Using grep for excluding words

Issue

For example my text is like this below.

mykey
nice comment
...
...
mykey
not bad comment
...
...
mykey
nice comment
...
...
mykey
excellent comment

I am using grep such as

$ grep 'mykey' * -1 | wc -l
$ 4

$ grep 'mykey' * -1 | grep 'nice comment' | wc -l
$ 2

This can check the number of mykey appears and nice comment is succeeded.

Then I want to see where the nice comment is not appeared after mykey.

At first I try this ,but it was wrong.

grep 'mykey' * -1 | grep -v 'nice comment'

I want to show such as,

mykey
not bad comment

mykey
excellent comment

Is it possible?

Solution

I don't think it is possible with grep, but you can use sed:

sed -n '/mykey/{N;/nice comment/!p;}' master.txt

Output:

mykey
not bad comment
mykey
excellent comment

If you only want to count number of "not nice" comments, you could still use two greps:

$ grep -A1 --group-separator= 'mykey' master.txt | grep -v 'nice comment\|mykey\|^$' | wc -l
2

Answered By - knittl

Answer Checked By - David Goodson (WPSolving Volunteer)