CODEWITHSUNDEEP
bashunixgrep
Fabian
Fabian

Reputation: 31

Can I grep for the number of same results of a grep result?

This is the grep command I have:

grep 'Query failed' file.txt | grep -o '\bT\w*'

Search for string 'Query failed' and then it gives me the words starting with T.

Result is for example:

Test1
Test1
Test2
Test2
Test2
Test3
Test4
Test4

Is it possible to get a result like:

Test1 = 2
Test2 = 3
Test3 = 1
Test4 = 2

Thanks a lot in advance

Upvotes: 3

Views: 61

Answers (2)

Tom Fenech
Tom Fenech

Reputation: 74665

One option, extending your current pipeline, would be to use uniq to count the occurrences and awk to reformat the line:

your_commands | uniq -c | awk '{ print $2 " = " $1 }'

Note that uniq expects to receive sorted input, so you may need to pipe to sort beforehand.

Alternatively, you could do the whole thing in awk:

awk '/Query failed/ { for (i = 1; i <= NF; ++i) if ($i ~ /^T/) ++seen[$i] } 
    END { for (i in seen) print i " = " seen[i] }' file

On lines matching the pattern, loop through all the words and keep a count of any starting with "T". Once the file has been processed, loop through the array and print the results.

Note that associative arrays are unsorted in awk, so the order of the output may vary.

Upvotes: 3

Barmar
Barmar

Reputation: 781370

Pipe it to sort and uniq -c to count all the repetitions.

grep 'Query failed' file.txt | grep -o '\bT\w*' | sort | uniq -c

Output will be:

   2 Test1
   3 Test2
   1 Test3
   2 Test4

From your example, it looks like the original file is already sorted. If so, you can omit the sort step.

Upvotes: 3

Related Questions