Search Query Syntax Parser

A simple parser for advanced search query syntax.

It parses a string like this:

from:hi@retrace.io,foo@gmail.com to:me subject:vacations date:1/10/2013-15/04/2014 photos

And turns it into an object like this:

{
  from: ['hi@retrace.io', 'foo@gmail.com'],
  to: 'me',
  subject: 'vacations',
  date: {
    from: '1/10/2013',
    to: '15/04/2014'
    },
  text: 'photos',
  offsets: 
   [ { keyword: 'from', value: 'hi@retrace.io,foo@gmail.com', offsetStart: 0, offsetEnd: 32 },
     { keyword: 'to', value: 'me', offsetStart: 33, offsetEnd: 38 },
     { keyword: 'subject', value: 'vacations', offsetStart: 39, offsetEnd: 56 },
     { keyword: 'date', value: '1/10/2013-15/04/2014', offsetStart: 57, offsetEnd: 82 },
     { text: 'photos', offsetStart: 83, offsetEnd: 89 } ]
}

Installation

$ npm install search-query-parser

Usage

var searchQuery = require('search-query-parser');

var query = 'from:hi@retrace.io,foo@gmail.com to:me subject:vacations date:1/10/2013-15/04/2014 photos';
var options = {keywords: ['from', 'to', 'subject'], ranges: ['date']}

var searchQueryObj = searchQuery.parse(query, options);

// searchQueryObj.from is now ['hi@retrace.io', 'foo@gmail.com']
// searchQueryObj.to is now 'me'
// searchQueryObj.date is now {from: '1/10/2013', to: '15/04/2014'}
// searchQueryObj.text is now 'photos'

You can configure what keywords and ranges the parser should accept with the options argument. It accepts 5 values:

keywords, that can be separated by commas (,). Accepts an array of strings.
ranges, that can be separated by a hyphen (-). Accepts an array of strings.
tokenize, that controls the behaviour of text search terms. If set to true, non-keyword text terms are returned as an array of strings where each term in the array is a whitespace-separated word, or a multi-word term surrounded by single- or double-quotes.
alwaysArray, a boolean that controls the behaviour of the returned query. If set to true, all matched keywords would always be arrays instead of strings. If set to false they will be strings if matched a single value. Defaults to false.
offsets, a boolean that controls the behaviour of the returned query. If set to true, the query will contain the offsets object. If set to false, the query will not contain the offsets object. Defaults to true.

If no keywords or ranges are specified, or if none are present in the given search query, then searchQuery.parse will return a string if tokenize is false, or an array of strings under the key text if tokenize is true.

var searchQuery = require('search-query-parser');

var query = 'a query with "just text"';
var parsedQuery = searchQuery.parse(query);
// parsedQuery is now 'a query with "just text"'

var options = {keywords: ['unused']};
var parsedQueryWithOptions = searchQuery.parse(query, options);
// parsedQueryWithOptions is now 'a query with "just text"'

var options2 = {tokenize: true};
var parsedQueryWithTokens = searchQuery.parse(query, options2);
// parsedQueryWithTokens is now: ['a', 'query', 'with', 'just text']

You can also use exclusion syntax, like -from:sep@foobar.io name:hello,world. This also works with non-keyword text terms when tokenize is set to true.

{
  name: ['hello', 'world'],
  exclude: {
    from: ['sep@foobar.io']
  }
}

Sometimes checking against whether a keyword holds string or not can be excessive and prone to errors; it's often easier to simply expect everything is an array even if it means doing 1-iteration loops often.

var searchQuery = require('search-query-parser');

var query = 'test:helloworld fun:yay,happy';
var options = {keywords: ['test', 'fun']};
var parsedQueryWithOptions = searchQuery.parse(query, options);
// parsedQueryWithOptions is now:
// {
//   test: 'helloworld',
//   fun: ['yay', 'happy']
// }

var optionsAlwaysArray = {keywords: ['test', 'fun'], alwaysArray: true};
var parsedQueryWithOptions = searchQuery.parse(query, options);
// parsedQueryWithOptions is now:
// {
//   test: ['helloworld'], //No need to check whether test is a string or not!
//   fun: ['yay', 'happy']
// }

The offsets object could become pretty huge with long search queries which could be an unnecessary use of space if no functionality depends on it. It can simply be turned off using the option offsets: false.

Anytime, you can go back and stringify the parsed search query. This could be handy if you would like to manipulate the parsed search query object.

var searchQuery = require('search-query-parser');

var query = 'from:hi@retrace.io,foo@gmail.com to:me subject:vacations date:1/10/2013-15/04/2014 photos';
var options = {keywords: ['from', 'to', 'subject'], ranges: ['date']}

var searchQueryObj = searchQuery.parse(query, options);

searchQueryObj.to = 'you';
var newQuery = searchQuery.stringify(query, options);

// newQuery is now: photos from:hi@retrace.io,foo@gmail.com to:you subject:vacations date:1/10/2013-15/04/2014

Typescript

Typescript types are available for this library in the docs directory. Browse type documentation here.

Documentation is generated using node_modules/.bin/typedoc index.d.ts

Testing

The 29 tests are written using the BDD testing framework should.js, and run with mocha.

Run npm install should and npm install -g mocha to install them both.

Run tests with make test.

License

The MIT License (MIT)

Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:

The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software.