README.md

EBNF-inspired pattern matching written in golang

Following are some applications for pattern matching:

Wrap all strings for translate

input

function sayHello() {
    console.log("hello \"world\"");
    console.log('how \'are\' you "doing" today');
}

output

function sayHello() {
    console.log(_t("hello \"world\""));
    console.log(_t('how \'are\' you "doing" today'));
}

Full code

package main

import (
	. "github.com/phannam1412/go-pattern-matching/parser"
	"strings"
)

func main() {
	input := `
		function sayHello() {
			console.log("hello \"world\"");
			console.log('how \'are\' you "doing" today');
		}
	`
	doubleQuote := And(
		Text(`"`),
		Any(Or(Text(`\"`),NotToken(`"`))),
		Text(`"`),
	)
	singleQuote := And(
		Text(`'`),
		Any(Or(Text(`\'`),NotToken(`'`))),
		Text(`'`),
	)
	formula := FullSearch(Or(doubleQuote, singleQuote), -1)

	tokens := Tokenize(input)

	parsed := formula(tokens, 0)

	output := input
	for _, v := range parsed.Children {
		output = strings.ReplaceAll(output, v.Value, "_t(" + v.Value + ")")
	}

	println(output)

}

Extract url

input

extract some url from https://example.com 
or http://www.w3school.com
or with hyphen http://www.w3-school.com

output

https://example.com
http://www.w3school.com
http://www.w3-school.com

formula

domainName := And(
    Alphabet,
    Any(And(Hyphen, Alphabet)),
)
formula := FullSearch(And(
    Or(Text("http"),Text("https")),
    Text("://"),
    Any(Text("www.")),
    domainName,
    Text("."),
    Alphabet,
), -1)

Extract emails from a bulk of text

input

extract some emails like [email protected], [email protected], 
and exclude invalid emails like ^[email protected] 

output

[email protected]
[email protected]
[email protected]

formula

name := And(Alphabet, Any(And(Hyphen, Alphabet)))
formula := FullSearch(And(name,At,Alphabet,Dot,Alphabet,), -1)