woodyHTML 解析/提取器
woody 是一款基于 Java 的 HTML 解析/提取器,用法非常类似 webmagic,是对其抽取模块的完全重写。
功能:
- 多种结果数据类型(String, char, byte, short int, long, double, float, string[], Set, List,Data)
- 支持用户之定义脚本处理函数(目前支持 Javascript 函数配置处理)
- 支持 css、xpath 内核替换
- 支持 filter 功能
- 对 css、xpath 内核对象的缓存
一个完整的例子:
public class OsChinaBlog { public static void main(String[] args) throws Exception { Document doc = Jsoup.connect("http://www.oschina.net/news/43879/webmagic-0-3-0").timeout(60000) .userAgent("Mozilla/5.0 (Macintosh; Intel Mac OS X 10.8; rv:23.0) Gecko/20100101 Firefox/23.0").get(); String html = doc.html(); OsChinaBlogModel model = AnnotationExtractor.me().process(html, OsChinaBlogModel.class); System.out.println(model.toJson()); } public static class OsChinaBlogModel extends Model { public OsChinaBlogModel() { //use to reflect } @Inject @ComboExtract(value = { @ExtractBy(value = "h1.OSCTitle", type = ExprType.CSS), @ExtractBy(value = "//title/text()", type = ExprType.XPATH) }, op = OP.OR) public String title; @Inject @ExtractBy(value = "div.PubDate a[href~=http://my\\.oschina\\.net/]", type = ExprType.CSS) public String author; @Inject @ExtractBy(value = "发布于.\\s*(\\d+年\\d+月\\d+日)", type = ExprType.REGEX) public Date publishDate; @Inject @ComboExtract(value = { @ExtractBy(value = "div.PubDate", type = ExprType.CSS, setting = @Setting(outerHtml = true)), @ExtractBy(value = "(\\d+)评", type = ExprType.REGEX) }, op = OP.AND) public int commentNum; @Inject @ExtractBy(value = "span#p_favor_count", type = ExprType.CSS, setting = @Setting(function = @Function(value = "replace", args = { "+", "" }))) public int collectNum; @Inject @ComboExtract(value = { @ExtractBy(value = "div[id=userComments]", type = ExprType.CSS, setting = @Setting(outerHtml = true)), @ExtractBy(value = "div.TextContent", type = ExprType.CSS) }, op = OP.AND, multi = true) public List commentContents; @Inject @ExtractBy(value = "div[id=toolbar_wrapper]", setting = @Setting(fliters = { "b", "span" }), type = ExprType.CSS, impl = Document.class) public String weibo; } }
评论