A small CLI program to read from the standard input and output consequent unique lines like uniq, but without needing to sort them and with ability to specify a field (or a range of fields) to determine uniqueness.
Either compile it using Nim compiler, with
> nim c -d:release unimq.nim
…or grab the binary from the releases page.
-d --delimiter="your char/string here"
Specify the delimiter to use. Note that if this option is used, --fields must be used as well.
-f --fields="0" ="3" ="0..2" ="1..4"
Specify the fields to use. Note that if this option is used, --delimiter must be used as well.
-h --help help Print out available options and some usage examples.
Or use it without arguments to uniq by whole lines.
# Just get unique consecutive lines like you would with `uniq`
> echo "1
2
1
3
3
1" | unimq
1
2
1
3
1
# Specify a field
> echo "file1:10:0
file2:10:2
file1:5:3
file3:10:22
file4:10:11
file3:15:5" | unimq --delimiter=":" --fields="1"
file1:10:0
file1:5:3
file3:10:22
file3:15:5
# Specify a range of fields
> echo "file1:10:0
file1:10:20
file1:13:1" | unimq --delimiter=":" --fields="0..1"
file1:10:0
file1:13:1