UNIX for Dummies Questions & Answers

In a file a pattern is occured many times randomly. Even it may appear more then once in the same line too. How i can get the number of times that pattern appeared in the file? let the file name is abc.txt and the pattern is "xyz".

I used the following code:
grep -ic "xyz" abc.txt

but it is not perfectly showing the result. if the pattern is repeated in a line two or more times it consider only once. so this code actually shows number of lines where pattern is found but not showing how many times in the entire file the pattern appeared? Shall i need a script for it? how can i overcome this situation?

Can anybody help me please?

You can try something like this:

Code:

awk -v var="xyz" '{for(i=1;i<=NF;i++){if($i==var){n++}}}END{print n}' file

Regards

One cheat method:

Code:

sed 's/xyz/xyz\n/g' abc.txt | grep -c xyz

If the pattern always consists of alphabetics, try something like

Code:

tr -c A-Za-z '
' <abc.txt | grep -c xyz

This replaces any non-alphabetics with newlines, so every word is on a separate line, and then, of course, it's okay that grep -c counts lines.

(Yes, the second non-option argument to tr is a literal line break between single quotes. If your tr understands some more readable notation, like maybe '\n' or '\012', then by all means use that instead.)