Text extraction JavaScript library

• Text extraction

• Tag Recommended

Basically, moving it to read the following in this order:

Lib / lib.js

thing in common

lib/extract-content.js

本文抽出

The repository of route make package Then extract-content-all.js that is the concatenation of these are generated.

When you see you want to detail the actual use:

sketch/extract-content.test.js

本文抽出テスト

Lib / scoring-words.js

scoring tag (sample)

Use if you want to specify that you want / handler only text extraction.

var ex = new ExtractContentJS.LayeredExtractor(); //ex.addHandler( ex.factory.getHandler('Description') ); //ex.addHandler( ex.factory.getHandler('Scraper')); //ex.addHandler( ex.factory.getHandler('GoogleAdsence') ); ex.addHandler( ex.factory.getHandler('Heuristics') ); var res = ex.extract(document); if (res.isSuccess) { res.url; // URL string res.title; // title string res.engine; // handler itself used for extraction res.content; // content class of an instance (see below) }

Handler is far Heuristics only been implemented.

Returns an array of // body's determined to be leaves class instance that contains the leaf node (see below); content.asLeaves () content.asNode (); // return the things of the deepest of the common ancestor of all of the leaf nodes content.asTextFragment (); return a concatenation of the text of the node that is included in the // asLeaves () Return the textContent of // asNode (); content.toString ()

leaf.node; // leaf node leaf.depth; // depth from the body of the node

Ina Lintro

Copyright © 2009 INA Lintaro / Hatena. All rights reserved.

Copyright © 2007/2008 Nakatani Shuyo / Cybozu Labs Inc. All rights reserved.

• labs.cybozu.co.jp/blog/nakatani/2007/09/web_1.html

MIT License