Search for two sequential patterns with any number of characters in between using grep

Question

I want to look for lines with "word1 ... word2" where '...' could be any different characters. So far I have used two greps for the same like this:

grep "$word1" $filename | grep "$word2"

Is there a faster way to do this by suppose something like this:

grep "$word1*$word2" $filename

where maybe * could be some special character which can be any other character(s)?

pLumo · Accepted Answer · 2018-06-21T11:53:45.683

2

Use .*:

grep "${word1}.*${word2}" "$filename"

. matches any character
* matches any number of the preceding character

edited Jun 21 '18 at 11:53

answered Jun 21 '18 at 11:43

pLumo

22,565

2

More precisely, .* matches any chararacter repeated 0 or more times (IOW, it matches any string, including the empty string). – NickD Jun 21 '18 at 11:50
This is precisely what I was looking for – Pratik Mayekar Jun 21 '18 at 11:52

score 0 · Answer 2 · answered Jun 21 '18 at 12:03

If you need the two words to be delimited, i.e. if you do not want to match abba if one of the words is bb, then use

grep "\\<$word1\\>.*\\<$word2\\>" "$filename"

The pattern \< (here \\< to escape the first backslash from the shell) matches just before a word, and \> works similarly but just after a word.

There's alo \b that matches either before and after, and [[:<:]] and [[:>:]] that work just like \< and \>. Which ones that are implemented by grep varies.

Search for two sequential patterns with any number of characters in between using grep

2 Answers2