Removing duplicates in grep output

Question

I have a case where i got a results file with the following pattern:

path:pattern found

for example

./user/home/file1:this is a game

in other words when i searched for some string i got the file and the line it found it.

Problem is sometimes i have multiple cases in the same file so i would like to remove the duplicates files (the cases would be different so it's not possible).

Any help or ideas are appreciated :)

End results is to turn this:

/user/home/desktop/file1:this is a game
/user/home/desktop/file1:what kind of game
/user/home/desktop/file1:fast action game

into just the first results found without losing all the rest of the data in the file.

Update1:

So the actual file looks like this:

/user/home/desktop/file1:this is a game
/user/home/desktop/file1:what kind of game
/user/home/desktop/file1:fast action game
/user/home/desktop/file2:a game
/user/home/desktop/file3:of game
/user/home/desktop/file4:fast game

i'm looking to get rid of the multiple occurrences in the same file so it should look like this:

/user/home/desktop/file1:this is a game
/user/home/desktop/file2:a game
/user/home/desktop/file3:of game
/user/home/desktop/file4:fast game

Answer 1

You could use sort -u:

grep pattern files | sort -t: -u -k1,1

• -t: - use : as the delimiter
• -k1,1 - sort based on the first field only
• -u - removed duplicates (based on the first field)

This will retain just one occurrence of files, removing any duplicates.

For your example, this is the output you get:

/user/home/desktop/file1:this is a game

In case you are looking for multiple distinct matches with a file, then:

grep pattern files | sort -u