List of usage examples for org.jsoup.nodes Document select
public Elements select(String cssQuery)
From source file:Main.java
public static void main(String[] args) throws Exception { Document doc = Jsoup.parse(new File("t2.html"), "UTF-8"); doc.select("area#area1").after("<area id=\"newArea\" />"); System.out.println(doc.html()); }
From source file:Main.java
public static void main(String[] args) throws Exception { String URL = "http://www.your server.com"; Document doc = Jsoup.connect(URL).get(); String cheapest = doc.select("b.item-price").first().text(); System.out.println(cheapest); }
From source file:Main.java
public static void main(String[] args) throws Exception { String URL = "http://www.your server.com"; Document doc = Jsoup.connect(URL).get(); String cheapest = doc.select("input[name=p-min]").first().attr("value"); System.out.println(cheapest); }
From source file:Main.java
public static void main(String[] args) throws Exception { String html = "<description>" + "..." + "</description>"; Document doc = Jsoup.parse(html); for (Element desc : doc.select("description")) { String unescapedHtml = desc.text(); String src = Jsoup.parse(unescapedHtml).select("img").first().attr("src"); System.out.println(src);/*from w w w .j a va 2 s .c o m*/ } System.out.println("Done"); }
From source file:Main.java
public static void main(String[] args) throws Exception { String xmlStr = "<style>" + // "v\\:* {behavior:url(#default#VML);}" + // "o\\:* {behavior:url(#default#VML);}" + // "w\\:* {behavior:url(#default#VML);}" + // ".shape {behavior:url(#default#VML);}" + // "</style>" + // "<xml>" + // "<w:WordDocument>" + // "<w:View>Normal</w:View>" + // "<w:Zoom>0</w:Zoom>" + // "<w:TrackMoves>false</w:TrackMoves>" + // "</xml>"; Document doc = Jsoup.parse(xmlStr, "", Parser.xmlParser()); doc.select("style").remove(); System.out.println(doc);//from www . j a va 2 s . c o m }
From source file:Main.java
public static void main(String[] args) { String html = "<html> " + " <body> " + " <table id='myTable'> " + " <tbody> " + " <tr> " + " <th>header</th> " + " <td>1</td> " + " <table> " + " <tbody> " + " <tr> " + " <td>a1</td> " + " <td>a2</td> " + " <td>a3</td> " + " </tr> " + " </tbody> " + " </table> " + " </td> " + " <td>high level2</td>" + " <td>high level3</td>" + " </tr> " + " </tbody> " + " </table> " + " </body> " + "</html> "; Document doc = Jsoup.parse(html); Elements highLevelTDs = doc.select("#myTable > tbody > tr > td"); System.out.println(highLevelTDs.size()); for (Element td : highLevelTDs) { System.out.println(td);//from ww w. ja v a 2s . c o m } }
From source file:Main.java
public static void main(String[] args) throws IOException { String url = "data.html"; Document doc = Jsoup.parse(new File(url), "UTF-8"); Elements rows = doc.select("tr"); for (Element row : rows) { print("---------"); Elements data = row.getElementsByTag("td"); print("First Name:%s", data.get(0).text()); print("Last Name:%s", data.get(1).text()); print("Date:%s", data.get(2).text()); print("City:%s", data.get(3).text()); }//ww w . j a v a 2 s .c o m }
From source file:Main.java
public static void main(String[] args) throws IOException { Document doc = Jsoup.connect("http://www.your server.com/").get(); Elements tableRows = doc.select("tr"); for (Element row : tableRows) { Elements cls1 = row.getElementsByClass("cls1"); Elements cls2 = row.getElementsByClass("cls2"); Elements cls3 = row.getElementsByClass("cls3"); if (!cls1.isEmpty() && !cls2.isEmpty() && !cls3.isEmpty()) { System.out.println(cls1.get(0).text()); System.out.println(cls2.get(0).text()); System.out.println(cls3.get(0).text()); }/*w w w . jav a 2 s . c o m*/ } }
From source file:Main.java
public static void main(String[] args) { Document doc = Jsoup.parse("html with frame", "UTF-8"); Document noFramesDoc = Jsoup.parseBodyFragment(doc.select("noframes").text()); System.out.println("Age = " + noFramesDoc.select("input[id=age]").attr("value")); System.out.println("Class = " + noFramesDoc.select("input[id=class]").attr("value")); }
From source file:Main.java
public static void main(String[] args) throws IOException { String docId = "1"; String url = "http://www.your server.com/" + docId; Document doc = Jsoup.connect(url).timeout(60000).userAgent("Mozilla/25.0").get(); Elements authors = doc.select("div"); System.out.println("authors=" + authors); System.out.println("authors.length=" + authors.size()); for (Element a : authors) { System.out.println(" author: " + a); }/* w ww. ja va2 s.c om*/ }