Racci

A lightweight search engine for Node.js, Chinese supported.

Feature

• Support two search mode: full-text search and field search

• Support Chinese query

• Customize field weight in full-text search

Usage

It is extremly simple.

Step 1: require racci

npm i racci --save

var racci = require('racci')

Step 2: import corpus

var docs = {
"1":{"singer":"Jay Chou",
		"composer":"Jay Chou"},
"2":{"singer":"Jay Chou",
		"composer":"Jay Chou"}}

// format: {identifier:[Object]}

racci.Parser.import(docs)

Step 3: Build models once and for all

// full-text search model
racci.Parser.init("corpus", ["lyrics", "singer", "composer", 'songwritter', 'album'], [1, 20, 3, 2, 1])

// name, search from these fields, field weight for ranking.

// field search model
racci.Parser.init("singer", ["singer","composer"], "commit_count")

// name, search from these fields, sort by this field.

Step 4: Search!

racci.Search.search("corpus", "流星雨",0,0)

// [ { doc: '1',},
//  { doc: '2'} ]

// first flag: show doc details or not
// second flag: show score or not

racci.Search.search("singer", "周杰伦",0,1)

// [ { doc: '1', score: '57' },
//   { doc: '2', score: '22' }]

racci.Search.search("corpus", "流星雨",1,0)

// [ { doc: [Object] },
//  { doc: [Object] } ]

racci.Search.search("singer", "周杰伦",1,1)

// [ { doc: [Object], score: '57' },
//   { doc: [Object], score: '22' }]

Examples

Implementation

A music search engine based on Express

HERE