kld-regex

A simple regular expression engine (and lexer) for ASCII text.

• Regular Expressions
  • Compile Regex
  • Get first match
  • Get all matches
  • Supported operators
• Lexers
  • Define a list of token types
  • Build Lexer
  • Match against source

Compile Regex

let Regex = require('kld-regex').Regex;
let runner = Regex.compile("(abc)+");

Get first match

var source = "abcdefabc";
var match = runner.next(source);

if (match !== null) {
    console.log(
        "source[%s:%s] = %s",
        match.startingOffset,
        match.endingOffset,
        match.text
    );
}

runner.next has optional 2nd and 3rd arguments. The 2nd argument is the starting offset where the Regex should begin its match in the specified string. The 3rd argument indicates the offset where the match should stop. Note that the ending offset is not included in the match range. This allows the user to say things like runner.next(source, currentOffset, source.length). If the starting offset is not specified, it defaults to 0. If the ending offset is not specified, it defaults to the specified string's length.

Get all matches

var source = "abcdefabc";
var matches = runner.all(source);

runner.all(source).forEach(match => {
    console.log(
        "source[%s:%s] = %s",
        match.startingOffset,
        match.endingOffset,
        match.text
    );
});

This will print:

Matches
=======
source = 'abcdefabc'
source[0:3] = abc
source[6:9] = abc

Supported operators

| Type | Example | |---------------------------------------|---------------- | | Sequence: | abc | | Alternation: | a|b|c | | Grouping: | (ab)|(cd) | | Repeat zero-or-more (Kleene closure): | \d*\.\d* | | Repeat one-or-more (Positive closure):| \d+\.\d+ | | Optional: | (abc)?def | | Character Classes: | [abc] | | Character Class Ranges: | [a-c] | | Character Class Negation: | [^abc] | | Empty Character Class: | [] | | Negated Empty Character Class: | [^] | | Match all except \r and \n: | . | | Digit Character Class: | \d | | Not-Digit Character Class: | \D | | Whitespace Character Class: | \s | | Not-Whitespace Character Class: | \S | | Word Character Class: | \w | | Not-Word Character Class: | \W | | Hex Character Values: | \x20 |

Define a list of token types

Each token type should have the following properties:

• type: a string
• pattern: a kld-regex string pattern (regex)
• caseSensitive: an optional boolean (defaults to true)

var tokens = [
    /* 0 */ { type: "whitespace",   pattern: "\\s+" },
    /* 1 */ { type: "comment",      pattern: "#[^\\r\\n]*" },
    /* 2 */ { type: "identifier",   pattern: "[a-z_][a-z0-9_]*", caseSensitive: false }
];

Build Lexer

let Lexer = require('kld-lexer').Lexer;
let lexer = new Lexer();
let runner = lexer.compile(tokens);

Match against source

let source = fs.readFileSync('sample.txt', { encoding: "utf8" });

runner.all(source).forEach(match => {
    // skip whitespace, but show comments
    if (match.type > 0) {
        console.log(
            "%s: [%s,%s]: %s",
            tokens[match.type].type,
            match.startingOffset,
            match.endingOffset,
            match.text
        );
    }
});