Lowden

now

用htmlparser分析并抽取正文

 


/**
* 收集HTML页面信息 调用抓取函数 按照自己的摸版 生成网页
* @param url
* @param urlEncode
*/
public void makeContext(ChannelLinkDO c) {
String metakeywords = "<META content={0} name=keywords>";
String metatitle = "<TITLE>{0}</TITLE>";
String metadesc = "<META content={0} name=description>";
String netshap = "<p> 正文快照: 时间{0}</p> ";

String tempLeate = "<LI class=active><A href=\"{0}\" target=_blank>{1}</A></LI>";
String crop = "<p><A href=\"{0}\" target=_blank>{1}</A></p> ";

try {
String siteUrl = getLinkUrl(c.getLink());
Parser parser = new Parser(c.getLink());
parser.setEncoding(c.getEncode());

for (NodeIterator e = parser.elements(); e.hasMoreNodes();) {
Node node = (Node) e.nextNode();

if (node instanceof Html) {
PageContext context = new PageContext();
context.setNumber(0);
context.setTextBuffer(new StringBuffer());
//抓取出内容
extractHtml(node, context, siteUrl);

StringBuffer testContext = context.getTextBuffer();
//.....生成网页
}
}
} catch (Exception e) {
System.out.println(e);
}
}

private String getLinkUrl(String link) {
String urlDomaiPattern = "(http://[^/]*?" + "/)(.*?)";

Pattern pattern = Pattern.compile(urlDomaiPattern,
Pattern.CASE_INSENSITIVE + Pattern.DOTALL);
Matcher matcher = pattern.matcher(link);
String url = "";

while (matcher.find()) {
int start = matcher.start(1);
int end = matcher.end(1);

url = link.substring(start, end - 1).trim();
}

return url;
}

/**
* 递归钻取正文信息
* @param nodeP
* @return
*/
protected List extractHtml(Node nodeP, PageContext context, String siteUrl)
throws Exception {
NodeList nodeList = nodeP.getChildren();
boolean bl = false;

if ((nodeList == null) || (nodeList.size() == 0)) {
if (nodeP instanceof ParagraphTag) {
ArrayList tableList = new ArrayList();
StringBuffer temp = new StringBuffer();
temp.append("<p style=\"TEXT-INDENT: 2em\">");
tableList.add(temp);
temp = new StringBuffer();
temp.append("</p>").append(lineSign);
tableList.add(temp);

return tableList;
}

return null;
}

if ((nodeP instanceof TableTag) || (nodeP instanceof Div)) {
bl = true;
}

if (nodeP instanceof ParagraphTag) {
ArrayList tableList = new ArrayList();
StringBuffer temp = new StringBuffer();
temp.append("<p style=\"TEXT-INDENT: 2em\">");
tableList.add(temp);
extractParagraph(nodeP, siteUrl, tableList);

temp = new StringBuffer();
temp.append("</p>").append(lineSign);

tableList.add(temp);

return tableList;
}

ArrayList tableList = new ArrayList();

try {
for (NodeIterator e = nodeList.elements(); e.hasMoreNodes();) {
Node node = (Node) e.nextNode();

if (node instanceof LinkTag) {
tableList.add(node);
setLinkImg(node, siteUrl);
} else if (node instanceof ImageTag) {
ImageTag img = (ImageTag) node;

if (img.getImageURL().toLowerCase().indexOf("http://") < 0) {
img.setImageURL(siteUrl + img.getImageURL());
} else {
img.setImageURL(img.getImageURL());
}

tableList.add(node);
} else if (node instanceof ScriptTag ||
node instanceof StyleTag || node instanceof SelectTag) {
} else if (node instanceof TextNode) {
if (node.getText().length() > 0) {
StringBuffer temp = new StringBuffer();
String text = collapse(node.getText()
.replaceAll("&nbsp;", "")
.replaceAll("　", ""));

temp.append(text.trim());

tableList.add(temp);
}
} else {
if (node instanceof TableTag || node instanceof Div) {
TableValid tableValid = new TableValid();
isValidTable(node, tableValid);

if (tableValid.getTrnum() > 2) {
tableList.add(node);

continue;
}
}

List tempList = extractHtml(node, context, siteUrl);

if ((tempList != null) && (tempList.size() > 0)) {
Iterator ti = tempList.iterator();

while (ti.hasNext()) {
tableList.add(ti.next());
}
}
}
}
} catch (Exception e) {
return null;
}

if ((tableList != null) && (tableList.size() > 0)) {
if (bl) {
StringBuffer temp = new StringBuffer();
Iterator ti = tableList.iterator();
int wordSize = 0;
StringBuffer node;
int status = 0;
StringBuffer lineStart = new StringBuffer(
"<p style=\"TEXT-INDENT: 2em\">");
StringBuffer lineEnd = new StringBuffer("</p>" + lineSign);

while (ti.hasNext()) {
Object k = ti.next();

if (k instanceof LinkTag) {
if (status == 0) {
temp.append(lineStart);
status = 1;
}

node = new StringBuffer(((LinkTag) k).toHtml());
temp.append(node);
} else if (k instanceof ImageTag) {
if (status == 0) {
temp.append(lineStart);
status = 1;
}

node = new StringBuffer(((ImageTag) k).toHtml());
temp.append(node);
} else if (k instanceof TableTag) {
if (status == 0) {
temp.append(lineStart);
status = 1;
}

node = new StringBuffer(((TableTag) k).toHtml());
temp.append(node);
} else if (k instanceof Div) {
if (status == 0) {
temp.append(lineStart);
status = 1;
}

node = new StringBuffer(((Div) k).toHtml());
temp.append(node);
} else {
node = (StringBuffer) k;

if (status == 0) {
if (node.indexOf("<p") < 0) {
temp.append(lineStart);
temp.append(node);
wordSize = wordSize + node.length();
status = 1;
} else {
temp.append(node);
status = 1;
}
} else if (status == 1) {
if (node.indexOf("</p") < 0) {
if (node.indexOf("<p") < 0) {
temp.append(node);
wordSize = wordSize + node.length();
} else {
temp.append(lineEnd);
temp.append(node);
status = 1;
}
} else {
temp.append(node);
status = 0;
}
}
}
}

if (status == 1) {
temp.append(lineEnd);
}

if (wordSize > context.getNumber()) {
context.setNumber(wordSize);
context.setTextBuffer(temp);
}

return null;
} else {
return tableList;
}
}

return null;
}

/**
* 设置图象连接
* @param nodeP
* @param siteUrl
*/
private void setLinkImg(Node nodeP, String siteUrl) {
NodeList nodeList = nodeP.getChildren();

try {
for (NodeIterator e = nodeList.elements(); e.hasMoreNodes();) {
Node node = (Node) e.nextNode();

if (node instanceof ImageTag) {
ImageTag img = (ImageTag) node;

if (img.getImageURL().toLowerCase().indexOf("http://") < 0) {
img.setImageURL(siteUrl + img.getImageURL());
} else {
img.setImageURL(img.getImageURL());
}
}
}
} catch (Exception e) {
return;
}

return;
}

/**
* 钻取段落中的内容
* @param nodeP
* @param siteUrl
* @param tableList
* @return
*/
private List extractParagraph(Node nodeP, String siteUrl, List tableList) {
NodeList nodeList = nodeP.getChildren();

if ((nodeList == null) || (nodeList.size() == 0)) {
if (nodeP instanceof ParagraphTag) {
StringBuffer temp = new StringBuffer();
temp.append("<p style=\"TEXT-INDENT: 2em\">");
tableList.add(temp);
temp = new StringBuffer();
temp.append("</p>").append(lineSign);
tableList.add(temp);

return tableList;
}

return null;
}

try {
for (NodeIterator e = nodeList.elements(); e.hasMoreNodes();) {
Node node = (Node) e.nextNode();

if (node instanceof ScriptTag || node instanceof StyleTag ||
node instanceof SelectTag) {
} else if (node instanceof LinkTag) {
tableList.add(node);
setLinkImg(node, siteUrl);
} else if (node instanceof ImageTag) {
ImageTag img = (ImageTag) node;

if (img.getImageURL().toLowerCase().indexOf("http://") < 0) {
img.setImageURL(siteUrl + img.getImageURL());
} else {
img.setImageURL(img.getImageURL());
}

tableList.add(node);
} else if (node instanceof TextNode) {
if (node.getText().trim().length() > 0) {
String text = collapse(node.getText()
.replaceAll("&nbsp;", "")
.replaceAll("　", ""));
StringBuffer temp = new StringBuffer();
temp.append(text);
tableList.add(temp);
}
} else if (node instanceof Span) {
StringBuffer spanWord = new StringBuffer();
getSpanWord(node, spanWord);

if ((spanWord != null) && (spanWord.length() > 0)) {
String text = collapse(spanWord.toString()
.replaceAll("&nbsp;", "")
.replaceAll("　", ""));

StringBuffer temp = new StringBuffer();
temp.append(text);
tableList.add(temp);
}
} else if (node instanceof TagNode) {
String tag = node.toHtml();

if (tag.length() <= 10) {
tag = tag.toLowerCase();

if ((tag.indexOf("strong") >= 0) ||
(tag.indexOf("b") >= 0)) {
StringBuffer temp = new StringBuffer();
temp.append(tag);
tableList.add(temp);
}
} else {
if (node instanceof TableTag || node instanceof Div) {
TableValid tableValid = new TableValid();
isValidTable(node, tableValid);

if (tableValid.getTrnum() > 2) {
tableList.add(node);

continue;
}
}

extractParagraph(node, siteUrl, tableList);
}
}
}
} catch (Exception e) {
return null;
}

return tableList;
}

protected void getSpanWord(Node nodeP, StringBuffer spanWord) {
NodeList nodeList = nodeP.getChildren();

try {
for (NodeIterator e = nodeList.elements(); e.hasMoreNodes();) {
Node node = (Node) e.nextNode();

if (node instanceof ScriptTag || node instanceof StyleTag ||
node instanceof SelectTag) {
} else if (node instanceof TextNode) {
spanWord.append(node.getText());
} else if (node instanceof Span) {
getSpanWord(node, spanWord);
} else if (node instanceof ParagraphTag) {
getSpanWord(node, spanWord);
} else if (node instanceof TagNode) {
String tag = node.toHtml().toLowerCase();

if (tag.length() <= 10) {
if ((tag.indexOf("strong") >= 0) ||
(tag.indexOf("b") >= 0)) {
spanWord.append(tag);
}
}
}
}
} catch (Exception e) {
}

return;
}

/**
* 判断TABLE是否是表单
* @param nodeP
* @return
*/
private void isValidTable(Node nodeP, TableValid tableValid) {
NodeList nodeList = nodeP.getChildren();

/**如果该表单没有子节点则返回**/
if ((nodeList == null) || (nodeList.size() == 0)) {
return;
}

try {
for (NodeIterator e = nodeList.elements(); e.hasMoreNodes();) {
Node node = (Node) e.nextNode();

/**如果子节点本身也是表单则返回**/
if (node instanceof TableTag || node instanceof Div) {
return;
} else if (node instanceof ScriptTag ||
node instanceof StyleTag || node instanceof SelectTag) {
return;
} else if (node instanceof TableColumn) {
return;
} else if (node instanceof TableRow) {
TableColumnValid tcValid = new TableColumnValid();
tcValid.setValid(true);
findTD(node, tcValid);

if (tcValid.isValid()) {
if (tcValid.getTdNum() < 2) {
if (tableValid.getTdnum() > 0) {
return;
} else {
continue;
}
} else {
if (tableValid.getTdnum() == 0) {
tableValid.setTdnum(tcValid.getTdNum());
tableValid.setTrnum(tableValid.getTrnum() + 1);
} else {
if (tableValid.getTdnum() == tcValid.getTdNum()) {
tableValid.setTrnum(tableValid.getTrnum() +
1);
} else {
return;
}
}
}
}
} else {
isValidTable(node, tableValid);
}
}
} catch (Exception e) {
return;
}

return;
}

/**
* 判断是否有效TR
* @param nodeP
* @param TcValid
* @return
*/
private void findTD(Node nodeP, TableColumnValid tcValid) {
NodeList nodeList = nodeP.getChildren();

/**如果该表单没有子节点则返回**/
if ((nodeList == null) || (nodeList.size() == 0)) {
return;
}

try {
for (NodeIterator e = nodeList.elements(); e.hasMoreNodes();) {
Node node = (Node) e.nextNode();

/**如果有嵌套表单**/
if (node instanceof TableTag || node instanceof Div ||
node instanceof TableRow ||
node instanceof TableHeader) {
tcValid.setValid(false);

return;
} else if (node instanceof ScriptTag ||
node instanceof StyleTag || node instanceof SelectTag) {
tcValid.setValid(false);

return;
} else if (node instanceof TableColumn) {
tcValid.setTdNum(tcValid.getTdNum() + 1);
} else {
findTD(node, tcValid);
}
}
} catch (Exception e) {
tcValid.setValid(false);

return;
}

return;
}

protected String collapse(String string) {
int chars;
int length;
int state;
char character;
StringBuffer buffer = new StringBuffer();
chars = string.length();

if (0 != chars) {
length = buffer.length();
state = ((0 == length) || (buffer.charAt(length - 1) == ' ') ||
((lineSign_size <= length) &&
buffer.substring(length - lineSign_size, length).equals(lineSign)))
? 0 : 1;

for (int i = 0; i < chars; i++) {
character = string.charAt(i);

switch (character) {
case '\u0020':
case '\u0009':
case '\u000C':
case '\u200B':
case '\u00a0':
case '\r':
case '\n':

if (0 != state) {
state = 1;
}

break;

default:

if (1 == state) {
buffer.append(' ');
}

state = 2;
buffer.append(character);
}
}
}

return buffer.toString();
}

posted on 2009-07-17 17:16 Lowden 阅读(396) 评论(0)  编辑  收藏 所属分类: Java - HtmlParser 、转载经典


只有注册用户登录后才能发表评论。
网站导航:

My Links

Blog Stats

常用链接

留言簿

随笔分类

随笔档案

文章分类

文章档案

好友链接

搜索

最新评论

阅读排行榜

评论排行榜