1: /**
2: *
3: */
4: package com.taobao.cd.http.util;
5:
6: import java.io.IOException;
7: import java.io.InputStream;
8: import java.net.URL;
9: import java.net.URLConnection;
10: import java.util.ArrayList;
11: import java.util.HashMap;
12: import java.util.List;
13: import java.util.concurrent.Callable;
14: import java.util.concurrent.ExecutionException;
15: import java.util.concurrent.ExecutorService;
16: import java.util.concurrent.Executors;
17: import java.util.concurrent.FutureTask;
18:
19: import javax.imageio.ImageIO;
20: import javax.imageio.ImageReader;
21: import javax.imageio.stream.ImageInputStream;
22:
23: import org.htmlparser.Node;
24: import org.htmlparser.NodeFilter;
25: import org.htmlparser.Parser;
26: import org.htmlparser.Tag;
27: import org.htmlparser.Text;
28: import org.htmlparser.filters.TagNameFilter;
29: import org.htmlparser.tags.ImageTag;
30: import org.htmlparser.tags.LinkTag;
31: import org.htmlparser.util.NodeList;
32: import org.htmlparser.util.ParserException;
33:
34: /**
35: * @author zunyuan.jy
36: *
37: * @date 2011-10-25
38: */
39: public class HtmlStat { 40:
41: private HashMap<Class<? extends Node>, Integer> statMap; // 存放所有標簽的計數器
42:
43: private ExecutorService exec;
44:
45: private List<String> textList; // 存放文本數據的隊列
46:
47: private List<LinkObject> linkList; // 存放超鏈接對應的文本
48:
49: private List<ImageObject> imageObjectList; // 存放處理完后的圖像信息
50:
51: private Parser parser;
52:
53: private ParserPool parserPool;
54:
55: private String host;
56:
57: /**
58: * 先序遍歷dom樹 統計全文的字數 統計全文的標簽總數 統計各類標簽數，統計資源放入一個map
59: * 當發現有圖像標記時，可以將其放入一個隊列，然后開線程去預渲染圖片（只讀圖片頭） 圖片渲染完畢后，列出圖片的信息 統計全文的標簽文字數，非空
60: *
61: * @throws IOException
62: * @throws ParserException
63: *
64: */
65:
66: public HtmlStat() { 67: parserPool = ParserPool.getInstance();
68: textList = new ArrayList<String>(20);
69: linkList = new ArrayList<LinkObject>(20);
70: imageObjectList = new ArrayList<ImageObject>(20);
71: statMap = new HashMap<Class<? extends Node>, Integer>(100);
72: }
73:
74: private void createParser(String url) throws Exception { 75: URL u = new URL(url);
76: host = u.getProtocol() + "://" + u.getHost();
77: URLConnection con = u.openConnection();
78: con.setRequestProperty("User-Agent", HttpUtil.UA); 79: org.htmlparser.scanners.ScriptScanner.STRICT = false;
80: org.htmlparser.lexer.Lexer.STRICT_REMARKS = false;
81: parser = parserPool.borrowOne();
82: parser.setConnection(con);
83: }
84:
85: public void analyse(String url) throws Exception { 86: createParser(url);
87: exec = Executors.newCachedThreadPool();
88:
89: parse();
90:
91: exec.shutdown();
92: returnParser();
93: }
94:
95: private void parse() throws ParserException { 96: NodeFilter filter = new TagNameFilter("body"); 97: NodeList nodes = parser.extractAllNodesThatMatch(filter);
98: Node node = null;
99: if (nodes != null) { 100: for (int i = 0; i < nodes.size(); i++) { 101: node = nodes.elementAt(i);
102: recursiveParse(node);
103: }
104: }
105: }
106:
107: private void recursiveParse(Node node) { 108: if (node != null) { 109: NodeList nodes = node.getChildren();
110: Node tNode = null;
111: if (nodes != null) { 112: for (int i = 0; i < nodes.size(); i++) { 113: tNode = nodes.elementAt(i);
114: if (tNode instanceof Text) { 115: if (tNode.toPlainTextString().trim().equals("")) 116: continue;
117: if (node instanceof LinkTag) { 118: LinkTag aTag = (LinkTag) node;
119: if (aTag.isHTTPLink()
120: && aTag.getAttribute("href") != null) { 121: String link = aTag.getAttribute("href"); 122: LinkObject lo = new LinkObject();
123: lo.setLink(link);
124: lo.setText(tNode.toPlainTextString().trim());
125: linkList.add(lo);
126: }
127: } else { 128: textList.add(tNode.toPlainTextString().trim());
129: }
130: /*
131: * if (node instanceof LinkTag) { LinkTag aTag = 132: * (LinkTag) node; if (aTag.isHTTPLink() &&
133: * aTag.getAttribute("href") != null){ String link = 134: * aTag.getAttribute("href"); } } else if (node 135: * instanceof Div) { 136: *
137: * } else if (node instanceof TableTag || node
138: * instanceof TableHeader) { 139: *
140: * } else if (node instanceof TableRow || node
141: * instanceof TableColumn) { 142: *
143: * } else if (node instanceof ParagraphTag) { 144: *
145: * } else if (node instanceof DefinitionListBullet) { 146: *
147: * } else if (node instanceof BulletList) { 148: *
149: * } else if (node instanceof Span) { 150: *
151: * } else if (node instanceof HeadingTag) { 152: *
153: * }
154: */
155: } else if (tNode instanceof ImageTag) { 156: ImageTag img = (ImageTag) tNode;
157: String url = img.getImageURL();
158: if (!url.startsWith("http://") 159: && !url.startsWith("https://")) { 160: url = host + url;
161: }
162: final String imgSrc = url;
163: final String alt = img.getAttribute("alt"); 164: FutureTask<ImageObject> task = new FutureTask<ImageObject>(
165: new Callable<ImageObject>() { 166:
167: public ImageObject call() throws Exception { 168: // TODO Auto-generated method stub
169: URL u = new URL(imgSrc);
170: String suffix = imgSrc.substring(imgSrc
171: .lastIndexOf(".") + 1); 172: InputStream is = u.openStream();
173: ImageInputStream stream = ImageIO
174: .createImageInputStream(is);
175: ImageReader ir = ImageReaderFactory
176: .getInstance()
177: .createImageReader(suffix);
178: ImageObject io = new ImageObject();
179: if (ir != null) { 180: ir.setInput(stream, true, false);
181: int w = ir.getWidth(0);
182: int h = ir.getHeight(0);
183:
184: io.setUrl(imgSrc);
185: io.setWidth(w);
186: io.setHeight(h);
187: io.setFormat(suffix);
188: io.setAlt(alt);
189:
190: }
191: return io;
192: }
193:
194: });
195: try { 196: exec.execute(task);
197: imageObjectList.add(task.get());
198: } catch (InterruptedException e) { 199: e.printStackTrace();
200: } catch (ExecutionException e) { 201: e.printStackTrace();
202: }
203:
204: } else if (tNode instanceof Tag) { 205: int x = 1;
206: if (statMap.containsKey(tNode.getClass())) { 207: x = statMap.get(tNode.getClass());
208: x++;
209: }
210: statMap.put(tNode.getClass(), x);
211: recursiveParse(tNode);
212: }
213: }
214: }
215: }
216: }
217:
218: private void returnParser() throws Exception { 219: parserPool.returnOne(parser);
220: }
221:
222: public HashMap<Class<? extends Node>, Integer> getStatMap() { 223: return statMap;
224: }
225:
226: public List<String> getTextList() { 227: return textList;
228: }
229:
230: public List<LinkObject> getLinkList() { 231: return linkList;
232: }
233:
234: public List<ImageObject> getImageObjectList() { 235: return imageObjectList;
236: }
237:
238: public static void main(String[] args) { 239: HtmlStat hs = new HtmlStat();
240: try { 241: hs.analyse("http://163.com"); 242: System.out.println("該頁有標簽種類:"); 243: System.out.println(hs.getStatMap().size());
244: System.out.println("有圖像標簽:"); 245: System.out.println(hs.getImageObjectList().size());
246: System.out.println("有文本標簽:"); 247: System.out.println(hs.getTextList().size());
248: System.out.println("有超鏈接標簽"); 249: System.out.println(hs.getLinkList().size());
250: } catch (Exception e) { 251: e.printStackTrace();
252: }
253: }
254: }