1
Named capturing groups in go
2
The simplest way to gain access to any subgroup match in a regexp is by using indexes. This is still my preferred method
3
of choice for one-off tasks, and I use it daily to perform substitutions on my neovim buffers.
4
5
However, when introducing a regular expression into a codebase, I try to use named capture groups instead. They allow
6
you to access the match by a name, and I think they become especially useful for longer and more complex patterns where
7
they can help to serve as documentation.
8
9
Here is one way of using them in Go:
10
11
var re = regexp.MustCompile("This file is (?P<size>[0-9]+)(?P<unit>kb|mb|gb)")12
13
func main() {14
str := "This file is 100kb"15
match := re.FindSubmatch([]byte(str))16
fmt.Println(string(match[re.SubexpIndex("size")])) // 10017
fmt.Println(string(match[re.SubexpIndex("unit")])) // kb18
}
12
13
This is a fairly simple pattern, but I still think it is easier to read compared to its unnamed counter part:
14
15
var re = regexp.MustCompile("This file is ([0-9]+)(kb|mb|gb)")16
17
func main() {18
str := "This file is 100kb"19
match := re.FindSubmatch([]byte(str))20
fmt.Println(string(match[1])) // 10021
fmt.Println(string(match[2])) // kb22
}
15
16
The end
17
I usually tweet something when I've finished writing a new post. You can find me on Twitter
18
by clickingÂ