CODEWITHSUNDEEP
regex
coolent
coolent

Reputation: 21

Regex user_agent fields

I am trying to regex this single line user_agent field.

user_agent="Mozilla/5.0 (Macintosh; Intel Mac OS X 10_11_6) AppleWebKit/437.38 (KHTML, like Gecko) Chrome/49.0.3477.100 Safari/437.38"

cat myfile | grep -oP '(user_agent=[^ ]*)' | awk {'print $1'}

The command above returns

“user_agent="Mozilla/5.0 “

only. However I need a whole text

"Mozilla/5.0 (Macintosh; Intel Mac OS X 10_11_6) AppleWebKit/437.38 (KHTML, like Gecko) Chrome/49.0.3477.100 Safari/437.38"

to be matched.

Please help modify the regex pattern I used.

Upvotes: 0

Views: 126

Answers (1)

Amit Bhardwaj
Amit Bhardwaj

Reputation: 333

The problem you are facing is 2 fold.

  1. Your regex * is only fetching the first letter only since there is a space after that and that is not captured.
  2. In addition, even if you fix first part awk will again print only first content delimited by space.

So you need to let go of the awk print and use .* in place of * with a positive lookahead.

Check the following:

cat myfile | grep -oP '(user_agent=\".*\")(?=\saccept)'

Here, \".*\" is searching for everything that is within double quotes (?=\saccept)is a positive lookahead statement which stops your search as soon as it finds a space followed by accept.

Upvotes: 1

Related Questions