/// <summary> Finds a text node, however embedded it might be, and returns
/// it. The text node will retain links to its parents, so
/// further navigation is possible.
/// </summary>
/// <param name="searchText">The text to search for.
/// </param>
/// <returns> The list of text nodes (recursively) found.
/// </returns>
public virtual IText[] DigupStringNode(System.String searchText)
{
NodeList nodeList = SearchFor(searchText);
NodeList stringNodes = new NodeList();
for (int i = 0; i < nodeList.Size(); i++)
{
INode node = nodeList.ElementAt(i);
if (node is IText)
{
stringNodes.Add(node);
}
else
{
if (node is CompositeTag)
{
CompositeTag ctag = (CompositeTag)node;
IText[] nodes = ctag.DigupStringNode(searchText);
for (int j = 0; j < nodes.Length; j++)
{
stringNodes.Add(nodes[j]);
}
}
}
}
IText[] stringNode = new IText[stringNodes.Size()];
for (int i = 0; i < stringNode.Length; i++)
{
stringNode[i] = (IText)stringNodes.ElementAt(i);
}
return(stringNode);
}
public List <String> start(String htmlContent)
{
Parser parser = new Parser();
parser.InputHTML = htmlContent;
NodeList nodelist = parser.Parse(nodefilter);
int size = nodelist.Size();
List <String> results = new List <String>();
for (int i = 0; i < size; i++)
{
INode node = nodelist.ElementAt(i);
if (node is INode)
{
ITag tag = node as ITag;
if (needValue == "href")
{
results.Add(tag.GetAttribute(needValue));
}
else
{
results.Add(tag.FirstChild.GetText());
}
}
}
return(results);
}
public System.Collections.Specialized.StringCollection Extract()
{
String strPattern = "([A-Za-z0-9](([_\\.\\-]?[a-zA-Z0-9]+)*)@([A-Za-z0-9]+)(([\\.\\-]?[a-zA-Z0-9]+)*)\\.([A-Za-z]{2,}))";
System.Collections.Specialized.StringCollection collAddresses = new System.Collections.Specialized.StringCollection();
INodeFilter filter = new RegexFilter(strPattern, true);
NodeList nodes = m_obParser.Parse(filter);
if (null != nodes &&
0 != nodes.Size())
{
RegexOptions options = RegexOptions.IgnoreCase | RegexOptions.Multiline;
Regex obRegex = new Regex(strPattern, options);
for (Int32 i = 0; i < nodes.Size(); i++)
{
INode obNode = nodes.ElementAt(i);
String strText = obNode.GetText();
Match m = obRegex.Match(strText);
while (m.Success)
{
collAddresses.Add(m.Groups[0].Value);
// Advance to the next match.
m = m.NextMatch();
}
}
}
return(collAddresses);
}
private List <Centroid> initialCentroidRandom(int k, NodeList <BuildAction> observations)
{
List <Centroid> centroids = new List <Centroid>();
while (k > 0)
{
centroids.Add(new Centroid(observations.ElementAt(rand.Next(observations.Count - 1))));
k--;
}
return(centroids);
}
public static void RemoveMeaninglessNodes(this NodeList list)
{
for (int i = 0; i < list.Count; i++)
{
INode node = list.ElementAt(i);
if (node is TextNode && node.ToPlainTextStringEx().Trim().Equals(""))
{
list.Remove(node);
}
}
}
public RJOutline(String docs)
{
HTMLParser p = HTMLParser.GetByHTML(docs);
NodeList nodes = p.GetFirstNode("id", "work_outline").Children;
nodes.KeepAllNodesThatMatch(new TagNameFilter("tr"));
for (int i = 0; i < nodes.Count; i++)
{
INode node = nodes.ElementAt(i);
if (node != null)
{
node.Children.RemoveMeaninglessNodes();
this.data.Add(node.FirstChild.ToPlainTextStringEx().Trim(), node.LastChild.ToDividedTextString(" ").TrimAll());
}
}
}
private void DeleteExecute(object obj)
{
int index = NodeList.IndexOf(selectValue);
NodeList.Remove(selectValue);
if (NodeList.Count <= 0)
{
return;
}
if (index > 0)
{
SelectValue = NodeList.ElementAt(--index);
}
else
{
SelectValue = NodeList.ElementAt(0);
}
}
/// <summary>
/// 第二种页面：数据格式为文本
/// </summary>
/// <returns></returns>
public static string[] GetDataFromText(string link)
{
//定义一个数组存放需要的解析数据
string[] tempData = new string[6];
try
{
//过滤条件：添加第二次筛选条件
NodeFilter secondFilter = new AndFilter(new TagNameFilter("div"), new HasAttributeFilter("class", "nrl"));
NodeList secondNodeList = GetNodeList(link, secondFilter);
//获取目标div标签的子节点
secondNodeList = secondNodeList.ElementAt(0).Children;
//开始解析需要的数据项
TableTag targetTable = (TableTag)secondNodeList[0];
//获取表中的所有行
TableRow[] tr = targetTable.Rows;
//标题
tempData[0] = tr[0].ToPlainTextString().Trim();
//文章来源及更新时间
string temp = tr[1].ToPlainTextString().Trim();
int index1 = temp.IndexOf('：');
int index2 = temp.LastIndexOf('：');
tempData[1] = temp.Substring(index1 + 1, 3);
temp = temp.Substring(index2 + 1).Trim();
int year, month, date;
year = temp.LastIndexOf('年');
month = temp.LastIndexOf('月');
date = temp.LastIndexOf('日');
tempData[2] = temp.Substring(0, year) + "-" + temp.Substring(year + 1, month - year - 1)
+ temp.Substring(month + 1, date - month - 1);
//网址
tempData[3] = link;
//水位表
TableColumn[] tc = tr[3].Columns;
tempData[4] = tc[0].ChildAt(2).ToHtml();
}
catch (Exception ex)
{
Console.WriteLine("第二层解析出错啦" + ex.Message);
tempData[5] = "failed";
}
return(tempData);
}
/// <summary>
/// 从Html中获取所有超链接,必须以<html>包裹,自动过滤javascript:;等无效Url
/// </summary>
/// <param name="html">需要筛选的Html代码</param>
/// <param name="pre">链接前加</param>
/// <param name="end">链接后加</param>
/// <param name="charcontain">必须包含指定字符</param>
/// <returns></returns>
public string GetAllLink(string html, M_Collect_ListFilter model)
{
string list = "";
NodeList nodeList = GetTagList(html, "A");
for (int i = 0; i < nodeList.Size(); i++)
{
ATag link = (ATag)nodeList.ElementAt(i);
string href = link.GetAttribute("href");
if (string.IsNullOrEmpty(href) || href.ToLower().IndexOf("javascript") > -1)
{
continue;
}
//必须包含有指定字符串
if (!string.IsNullOrEmpty(model.CharContain) && !href.Contains(model.CharContain))
{
bool flag = false;
foreach (string start in model.CharContain.Split("|".ToCharArray(), StringSplitOptions.RemoveEmptyEntries))
{
if (href.Contains(start))
{
flag = true; break;
}
}
if (flag == false)
{
continue;
}
}
if (!string.IsNullOrEmpty(model.CharRegex) && !Regex.IsMatch(href, model.CharRegex))
{
continue;
}
list += (model.FillStart + link.GetAttribute("href") + model.FillEnd) + "\n";
}
return(list);
}
private List<Centroid> initialCentroidRandom(int k, NodeList<BuildAction> observations)
{
List<Centroid> centroids = new List<Centroid>();
while (k > 0)
{
centroids.Add(new Centroid(observations.ElementAt(rand.Next(observations.Count - 1))));
k--;
}
return centroids;
}
public INode GetFirstNode(String attribute, String regex)
{
NodeList list = GetNodes(attribute, regex);
return(list.ElementAt(0));
}
/// <summary> Collect the children.
/// <p>An initial test is performed for an empty XML tag, in which case
/// the start tag and end tag of the returned tag are the same and it has
/// no children.<p>
/// If it's not an empty XML tag, the lexer is repeatedly asked for
/// subsequent nodes until an end tag is found or a node is encountered
/// that matches the tag ender set or end tag ender set.
/// In the latter case, a virtual end tag is created.
/// Each node found that is not the end tag is added to
/// the list of children. The end tag is special and not a child.<p>
/// Nodes that also have a CompositeTagScanner as their scanner are
/// recursed into, which provides the nested structure of an HTML page.
/// This method operates in two possible modes, depending on a private boolean.
/// It can recurse on the JVM stack, which has caused some overflow problems
/// in the past, or it can use the supplied stack argument to nest scanning
/// of child tags within itself. The former is left as an option in the code,
/// mostly to help subsequent modifiers visualize what the internal nesting
/// is doing.
/// </summary>
/// <param name="tag">The tag this scanner is responsible for.
/// </param>
/// <param name="lexer">The source of subsequent nodes.
/// </param>
/// <param name="stack">The parse stack. May contain pending tags that enclose
/// this tag.
/// </param>
/// <returns> The resultant tag (may be unchanged).
/// </returns>
public override ITag Scan(ITag tag, Lexer lexer, NodeList stack)
{
INode node;
ITag next;
System.String name;
IScanner scanner;
ITag ret;
ret = tag;
if (ret.EmptyXmlTag)
{
ret.SetEndTag(ret);
}
else
{
do
{
node = lexer.NextNode(false);
if (null != node)
{
if (node is ITag)
{
next = (ITag)node;
name = next.TagName;
// check for normal end tag
if (next.IsEndTag() && name.Equals(ret.TagName))
{
ret.SetEndTag(next);
node = null;
}
else if (IsTagToBeEndedFor(ret, next))
// check DTD
{
// backup one node. insert a virtual end tag later
lexer.Position = next.StartPosition;
node = null;
}
else if (!next.IsEndTag())
{
// now recurse if there is a scanner for this type of tag
scanner = next.ThisScanner;
if (null != scanner)
{
if (mUseJVMStack)
{
// JVM stack recursion
node = scanner.Scan(next, lexer, stack);
AddChild(ret, node);
}
else
{
// fake recursion:
if (scanner == this)
{
if (next.EmptyXmlTag)
{
next.SetEndTag(next);
FinishTag(next, lexer);
AddChild(ret, next);
}
else
{
stack.Add(ret);
ret = next;
}
}
else
{
// normal recursion if switching scanners
node = scanner.Scan(next, lexer, stack);
AddChild(ret, node);
}
}
}
else
{
AddChild(ret, next);
}
}
else
{
if (!mUseJVMStack && !mLeaveEnds)
{
// Since all non-end tags are consumed by the
// previous clause, we're here because we have an
// end tag with no opening tag... this could be bad.
// There are two cases...
// 1) The tag hasn't been registered, in which case
// we just add it as a simple child, like it's
// opening tag
// 2) There may be an opening tag further up the
// parse stack that needs closing.
// So, we ask the factory for a node like this one
// (since end tags never have scanners) and see
// if it's scanner is a composite tag scanner.
// If it is we walk up the parse stack looking for
// something that needs this end tag to finish it.
// If there is something, we close off all the tags
// walked over and continue on as if nothing
// happened.
System.Collections.ArrayList attributes = System.Collections.ArrayList.Synchronized(new System.Collections.ArrayList(10));
attributes.Add(new TagAttribute(name, null));
ITag opener = lexer.NodeFactory.CreateTagNode(lexer.Page, next.StartPosition, next.EndPosition, attributes);
scanner = opener.ThisScanner;
if ((null != scanner) && (scanner == this))
{
// uh-oh
int index = -1;
for (int i = stack.Size() - 1; (-1 == index) && (i >= 0); i--)
{
// short circuit here... assume everything on the stack has this as it's scanner
// we'll need to stop if either of those conditions isn't met
ITag boffo = (ITag)stack.ElementAt(i);
if (name.Equals(boffo.TagName))
{
index = i;
}
else if (IsTagToBeEndedFor(boffo, next))
{
// check DTD
index = i;
}
}
if (-1 != index)
{
// finish off the current one first
FinishTag(ret, lexer);
AddChild((ITag)stack.ElementAt(stack.Size() - 1), ret);
for (int i = stack.Size() - 1; i > index; i--)
{
ITag fred = (ITag)stack.Remove(i);
FinishTag(fred, lexer);
AddChild((ITag)stack.ElementAt(i - 1), fred);
}
ret = (ITag)stack.Remove(index);
node = null;
}
else
{
AddChild(ret, next); // default behaviour
}
}
else
{
AddChild(ret, next); // default behaviour
}
}
else
{
AddChild(ret, next);
}
}
}
else
{
AddChild(ret, node);
node.DoSemanticAction();
}
}
if (!mUseJVMStack)
{
// handle coming out of fake recursion
if (null == node)
{
int depth = stack.Size();
if (0 != depth)
{
node = stack.ElementAt(depth - 1);
if (node is ITag)
{
ITag precursor = (ITag)node;
scanner = precursor.ThisScanner;
if (scanner == this)
{
stack.Remove(depth - 1);
FinishTag(ret, lexer);
AddChild(precursor, ret);
ret = precursor;
}
else
{
node = null; // normal recursion
}
}
else
{
node = null; // normal recursion
}
}
}
}
}while (null != node);
}
FinishTag(ret, lexer);
return(ret);
}
/// <summary>
/// 函数名称：ItemRetrival_2
/// 功能说明：用于提取帖子列表页面的url,帖子标题，帖子时间
/// 参数：string url表示帖子列表url
/// 参数 ref Encoding encode 用于获取网页字符集编码
/// 参数： ref List<string> listUrl,listTitle,listTime用于存放提取出的各项信息
///
/// </summary>
/// <param name="url"></param>
/// <param name="encode"></param>
/// <param name="listurl"></param>
/// <param name="listtitle"></param>
/// <param name="listtime"></param>
public static void ItemRetrival_2(string url, ref Encoding encode, ref List <string> listUrl, ref List <string> listTitle,
ref List <string> listTime)
{
//获取网页源码；
string rawtext = GetDataFromUrl(url);
string reg1 = @"<style[\s\S]+?/style>|<select[\s\S]+?/select>|<script[\s\S]+?/script>|<\!\-\-[\s\S]*?\-\->";
rawtext = new Regex(reg1, RegexOptions.Multiline | RegexOptions.IgnoreCase).Replace(rawtext, "");
//将无关的style，script等标签去掉；
//以下操作用于提取帖子页面的发帖时间、帖子URL，帖子标题等信息
//用htmlparser获取目标li元素
Lexer lexer = new Lexer(rawtext);
Parser parser = new Parser(lexer);
NodeFilter filter = new TagNameFilter("li");//解析出其中的li元素
NodeList htmlNodes = parser.Parse(filter);
//去掉其中不含有时间的条目
Regex f2 = new Regex(@"\d\d:\d\d");
for (int i = htmlNodes.Count - 1; i >= 0; i--)
{
if (!f2.IsMatch(htmlNodes[i].ToHtml()))
{
htmlNodes.Remove(i);
}
}
RegexFilter rf = new RegexFilter(@"\d\d:\d\d");
string final = htmlNodes.ToHtml();
for (int i = 0; i < htmlNodes.Count; i++)
{
Lexer lexerTmp = new Lexer(htmlNodes[i].ToHtml());
Parser parserTmp = new Parser(lexerTmp);
NodeList tmp = parserTmp.Parse(rf);
if (tmp.Count > 0)
{
for (int j = 0; j < tmp.Count; j++)
{
string temp = tmp[j].ToHtml();
ModifyRawText(ref temp);
listTime.Add(temp);
}
}
}
//提取帖子url和标题
string atagAssist = htmlNodes.ToHtml();
Lexer lex3 = new Lexer(atagAssist);
Parser par3 = new Parser(lex3);
NodeFilter filter3 = new TagNameFilter("a");
NodeList atagNodes = par3.Parse(filter3);
for (int i = 0; i < atagNodes.Count; i++)
{
string urlpart = new Regex(@"http://.*?(?=/)").Match(url).Value;
ATag link = (ATag)atagNodes.ElementAt(i);
string temp1 = link.GetAttribute("href");
string temp2 = link.StringText;
if (temp1 != null && !new Regex("http").IsMatch(temp1)) //如果提取出的url为相对url,则加上域名补全为绝对url
{
temp1 = urlpart + temp1; //将提取出的url构造完整，形成完整的url
}
ModifyRawText(ref temp2);
listUrl.Add(temp1);
listTitle.Add(temp2);
}
}
/// <summary>
/// 函数名称：ItemRetrival_1
/// 功能说明：用于提取帖子列表页面的url,帖子标题，帖子时间
/// 参数：string url表示帖子列表url
/// 参数 ref Encoding encode 用于获取网页字符集编码
/// 参数： ref List<string> listUrl,listTitle,listTime用于存放提取出的各项信息
/// </summary>
/// <param name="url"></param>
/// <param name="encode"></param>
/// <param name="listurl"></param>
/// <param name="listtitle"></param>
/// <param name="listtime"></param>
public static void ItemRetrival_1(string url, ref Encoding encode, ref List <string> listUrl, ref List <string> listTitle,
ref List <string> listTime)
{
//获取网页源码；
string rawtext = GetDataFromUrl(url);
//将无关的style，script等标签去掉；
string reg1 = @"<style[\s\S]+?/style>|<select[\s\S]+?/select>|<script[\s\S]+?/script>|<\!\-\-[\s\S]*?\-\->";
rawtext = new Regex(reg1, RegexOptions.Multiline | RegexOptions.IgnoreCase).Replace(rawtext, "");
//以下用htmlparser提取源码中的目标table;
Lexer lexer = new Lexer(rawtext);
//解析出其中的table元素
Parser parser = new Parser(lexer);
NodeFilter filter = new TagNameFilter("table");
NodeList htmlNodes = parser.Parse(filter);
//去除嵌套式table
Regex f1 = new Regex(@"<table.*?>");
for (int i = htmlNodes.Count - 1; i >= 0; i--)
{
MatchCollection myCollection = f1.Matches(htmlNodes[i].ToHtml());
if (myCollection.Count > 1)
{
htmlNodes.Remove(i);
}
}
//去除没有时间的table，认为这种table是无效table
Regex f2 = new Regex(@"\d\d:\d\d");
for (int i = htmlNodes.Count - 1; i >= 0; i--)
{
if (!f2.IsMatch(htmlNodes[i].ToHtml()))
{
htmlNodes.Remove(i);
}
}
//以下程序解析出以上三种目标信息
string final = htmlNodes.ToHtml();
Lexer lex2 = new Lexer(final);
Parser par2 = new Parser(lex2);
NodeFilter filter2 = new TagNameFilter("tr");
NodeList finalNodes = par2.Parse(filter2);
//提取发帖时间信息
RegexFilter rf = new RegexFilter(@"\d\d:\d\d");
for (int i = 0; i < finalNodes.Count; i++)
{
Lexer lexerTmp = new Lexer(finalNodes[i].ToHtml());
Parser parserTmp = new Parser(lexerTmp);
NodeList tmp = parserTmp.Parse(rf);
if (tmp.Count > 0)
{
for (int j = 0; j < tmp.Count; j++)
{
string temp = tmp[j].ToHtml();
ModifyRawText(ref temp);
listTime.Add(temp);
}
}
}
//提取帖子URL以及帖子标题
string atagAssist = finalNodes.ToHtml();
Lexer lex3 = new Lexer(atagAssist);
Parser par3 = new Parser(lex3);
NodeFilter filter3 = new TagNameFilter("a");
NodeList atagNodes = par3.Parse(filter3);
string urlpart = new Regex(@"http://.*?(?=/)").Match(url).Value;
for (int i = 0; i < atagNodes.Count; i++)
{
ATag link = (ATag)atagNodes.ElementAt(i);
string temp1 = link.GetAttribute("href");
string temp2 = link.StringText;
if (temp1 != null && !new Regex("http").IsMatch(temp1)) //如果提取出的url为相对url,则加上域名补全为绝对url
{
temp1 = urlpart + temp1; //将提取出的url构造完整，形成完整的url
}
ModifyRawText(ref temp2);
listUrl.Add(temp1);
listTitle.Add(temp2);
}
}
