tokenspliter v0.0.4

tokenspliter

Token spliter used to split tokens from text.

exmaple

let tokenspliter = Tokenspliter([{
    type: 'identify',
    regular: /[a-zA-Z][a-zA-Z0-9]*/
}, {
    type: 'number',
    regular: /[0-9]+/
}, {
    type: 'whitespace',
    regular: /\s+/
}]);

let tokenStream = tokenspliter('a09 bcd 12');
let t1 = tokenStream.next(); // {type: 'identity', 'lexion': 'a09'}
let t2 = tokenStream.next(); // {type: 'whitespace', 'lexion': ' '}

install

npm i tokenspliter --save

usage

initial tokenspliter

let tokenspliter = Tokenspliter([{
    type: 'identify',
    regular: /[a-zA-Z][a-zA-Z0-9]*/
}, {
    type: 'number',
    regular: /[0-9]+/
}, {
    type: 'whitespace',
    regular: /\s+/
}]);

Initial a token spliter, by pass an array which item contains a type and a regular expression.

accept text

let tokenStream = tokenspliter('a09 bcd 12');

operate token stream

• next

Move to next, and return next token.

let t1 = tokenStream.next(); // {type: 'identity', 'lexion': 'a09'}

• lookAhead

Look ahead one step, and return next token, but token stream will hold.

let t1 = tokenStream.lookAhead(); // {type: 'identity', 'lexion': 'a09'}

• lookAheads

Look ahead more steps.

First parameter is steps, second paramter is filter (optional). If has filter, only when filter returns true, it will count one step ahead.

let t1 = tokenStream.lookAheads(3, (token) => token.type !== 'whitespace');

• isEmpty

If token stream comes to end, will return true, otherwise will return false;

tokenStream.isEmpty()

error token

• default handling

By default, throw an exception.

let tokenspliter = Tokenspliter([{
    type: 'identify',
    regular: /[a-zA-Z][a-zA-Z0-9]*/
}, {
    type: 'number',
    regular: /[0-9]+/
}, {
    type: 'whitespace',
    regular: /\s+/
}]);
let tokenStream = tokenspliter('a09 & bcd 12');
tokenStream.next();
tokenStream.next();
tokenStream.next(); // throw an error

• ignore handling

Set tokespliter's second parameter as null, will ignore exception, and when meet unrecoginized token, will return it as error token. The data structure of error token is:

{
    type: 'error_type',
    lexcion: 'some characters'
}

let tokenspliter = Tokenspliter([{
    type: 'identify',
    regular: /[a-zA-Z][a-zA-Z0-9]*/
}, {
    type: 'number',
    regular: /[0-9]+/
}, {
    type: 'whitespace',
    regular: /\s+/
}]);
let tokenStream = tokenspliter('a09 & bcd 12', null); // null
tokenStream.next();
tokenStream.next();
tokenStream.next(); // {type: 'error_type', lexion: '&'}