public static void msdn_channel9process(string path, string link)
{
Denoising de = new Denoising();
de.Path = path;
de.Work(); // work后就获得了原始数据，去噪数据等
MainWindow._rawdata = de.Rawdata; // 获得原始数据
MainWindow._denoisingdata = de.Denoisingdata; // 获得去噪数据
if (MainWindow._denoisingdata == null) // 如果得到的去噪数据是null，说明可能存在编码或其他一些问题
{
MainWindow._denoisingdata = "";
}
// 初始化一个用来分词的类，对去噪数据进行分词处理
WordSegment ws = new WordSegment();
//bool isCHineseWebPage = WebTesting(MainWindow._denoisingdata);
MainWindow._wordSegmentResult = ws.cutwords(MainWindow._denoisingdata, WordSegmentData.preWordSegmentor);
foreach (string word in MainWindow._wordSegmentResult)
{
if (MainWindow.wordFreq.ContainsKey(word))
{
int freq = (int)MainWindow.wordFreq[word];
MainWindow.wordFreq[word] = freq + 1;
}
else
{
MainWindow.wordFreq.Add(word, 1);
}
}
MainWindow._tags = new List <string>();
MainWindow.tagPreview = new Hashtable();
// 初始化一个数据挖掘类的实例
DataMining dm = new DataMining();
// 获得原始数据和去噪数据
dm.Rawdata = MainWindow._rawdata;
dm.Denoisingdata = MainWindow._denoisingdata;
// 开始工作
dm.Work();
// 获得标题，作者，发布日期，其实很不准确
MainWindow._title = dm.Title;
MainWindow._author = dm.Author;
MainWindow._postdate = dm.Postdate;
Page pg = new Page();
PageItem pgitem = new PageItem();
MainWindow.curwid = MainWindow.getWebpageNo() + 1;
pgitem.setwid(MainWindow.curwid);
//item1.settitle(MainWindow._title);
//pgitem.settitle(translate(MainWindow._title));
pgitem.settitle(MainWindow._title);
pgitem.setauthor(MainWindow._author);
pgitem.setpostdate(MainWindow._postdate);
pgitem.setlink(MainWindow.curUrl);
pgitem.setreferred("");
pg.savePage(pgitem);
HtmlDocument doc = new HtmlDocument();
try
{
doc.Load(path, Encoding.UTF8);
}
catch (Exception e)
{
return;
}
HtmlNode rootNode = doc.DocumentNode;
string CategoryListXPath = "//title|//img";
HtmlNodeCollection categoryNodeList = rootNode.SelectNodes(CategoryListXPath);
foreach (HtmlNode child in categoryNodeList)
{
HtmlNode hn = HtmlNode.CreateNode(child.OuterHtml);
if (hn.SelectSingleNode("//title") != null)
{
MainWindow._title = hn.SelectSingleNode("//title").InnerText;
}
else if (hn.SelectSingleNode("//*[@class=\"thumb\"]") != null)
{
string str = hn.SelectSingleNode("//*[@class=\"thumb\"]").OuterHtml;
string reg = "(?<=alt=\")[^\"]*\"";
string key_words = Regex.Match(str, reg).Value;
MainWindow._tags.Add(key_words);
}
}
//save to table TAGS
foreach (string tag in MainWindow._tags)
{
if (!MainWindow._occurredTags.Contains(tag))
{
Tag thistag = new Tag();
TagItem tgitem = new TagItem();
tgitem.settid(MainWindow.getTagNo() + 1);
//MainWindow._occurredTags.Add(translate(tag));
MainWindow._occurredTags.Add(tag);
//tgitem.setname(translate(tag));
tgitem.setname(tag);
thistag.saveTag(tgitem);
}
}
foreach (string tag in MainWindow._tags)
{
Preview b = new Preview();
PreviewItem item3 = new PreviewItem();
//MainWindow.curtid = MainWindow.getTagNo(translate(tag));
MainWindow.curtid = MainWindow.getTagNo(tag);
item3.settid(MainWindow.curtid);
item3.setwid(MainWindow.curwid);
//item3.setpreview(translate((string)tagPreview[tag]));
item3.setpreview((string)MainWindow.tagPreview[tag]);
//item3.setpreview((string)tagPreview[tag]);
b.savePreview(item3);
}
}
public static void Process(string path, bool needWriteToDB) // @path:文件路径 @needWriteToDB:表示是否需要写入数据库
{
// 初始化一个用来去噪的类
Denoising de = new Denoising();
de.Path = path;
de.Work(); // work后就获得了原始数据，去噪数据等
MainWindow._rawdata = de.Rawdata; // 获得原始数据
MainWindow._denoisingdata = de.Denoisingdata; // 获得去噪数据
if (MainWindow._denoisingdata == null) // 如果得到的去噪数据是null，说明可能存在编码或其他一些问题
{
MainWindow._denoisingdata = "";
}
// 初始化一个用来分词的类，对去噪数据进行分词处理
WordSegment ws = new WordSegment();
//bool isCHineseWebPage = WebTesting(MainWindow._denoisingdata);
MainWindow._wordSegmentResult = ws.cutwords(MainWindow._denoisingdata, WordSegmentData.preWordSegmentor);
foreach (string word in MainWindow._wordSegmentResult)
{
if (MainWindow.wordFreq.ContainsKey(word))
{
int freq = (int)MainWindow.wordFreq[word];
MainWindow.wordFreq[word] = freq + 1;
}
else
{
MainWindow.wordFreq.Add(word, 1);
}
}
/*if (isCHineseWebPage) // 只对中文页面采用分词算法
* {
* MainWindow._wordSegmentResult = ws.DoWordSegment(MainWindow._denoisingdata); // 使用分词算法进行分词
*
* // 对分词结果中每个词的词频进行统计
* string word;
* Regex r = new Regex(@"\{<([^>]*)>\}", RegexOptions.Multiline);
* MatchCollection matches = r.Matches(MainWindow._wordSegmentResult);
* foreach (Match match in matches)
* {
* if (match.Success)
* {
* word = match.Groups[1].Value;
* if (MainWindow.wordFreq.ContainsKey(word))
* {
* int freq = (int)MainWindow.wordFreq[word];
* MainWindow.wordFreq[word] = freq + 1;
* }
* else
* {
* MainWindow.wordFreq.Add(word, 1);
* }
* }
* }
* }
* else // 对英文页面，直接采用分隔符分离的方法
* {
* MainWindow._wordSegmentResult = MainWindow._denoisingdata;
* string[] words = MainWindow._wordSegmentResult.Split(new char[] {' ', '`', '~', '!', '@', '$', '%',
* '^', '&', '*','(', ')', '_', '+', '=', '{', '[', '}', ']', ':', ';', '\"', '\'', ',',
* '<', '.', '>', '?', '/', '\\', '\r', '\n', '0', '1', '2', '3', '4', '5', '6', '7', '8', '9'});
* foreach (string word in words)
* {
* if (word.Length > 1)
* {
* if (MainWindow.wordFreq.ContainsKey(word.ToLower()))
* {
* int freq = (int)MainWindow.wordFreq[word.ToLower()];
* MainWindow.wordFreq[word.ToLower()] = freq + 1;
* }
* else
* {
* MainWindow.wordFreq.Add(word.ToLower(), 1);
* }
* }
* }
* }*/
// 数据挖掘
// 先清空两个容器
MainWindow._tags = new List <string>();
MainWindow.tagPreview = new Hashtable();
// 初始化一个数据挖掘类的实例
DataMining dm = new DataMining();
// 获得原始数据和去噪数据
dm.Rawdata = MainWindow._rawdata;
dm.Denoisingdata = MainWindow._denoisingdata;
// 开始工作
dm.Work();
// 获得标题，作者，发布日期，其实很不准确
MainWindow._title = dm.Title;
MainWindow._author = dm.Author;
MainWindow._postdate = dm.Postdate;
// 已经得到了想要的信息，开始
//DB OPERATION
//信息已经提取完成，存入数据库，包括tag webpage tag_webpage 3张表
//请把新的tag存入tag表中，注意编号
if (needWriteToDB)
{
/*suiyuhao*/
//save to table PAGE
Page pg = new Page();
PageItem pgitem = new PageItem();
MainWindow.curwid = MainWindow.getWebpageNo() + 1;
pgitem.setwid(MainWindow.curwid);
//item1.settitle(MainWindow._title);
//pgitem.settitle(translate(MainWindow._title));
pgitem.settitle(MainWindow._title);
pgitem.setauthor(MainWindow._author);
pgitem.setpostdate(MainWindow._postdate);
pgitem.setlink(MainWindow.curUrl);
pgitem.setreferred("");
pg.savePage(pgitem);
string body = dealhtml(MainWindow._rawdata);
string keywords = "";
//save to table TAGS
foreach (string tag in MainWindow._tags)
{
if (!MainWindow._occurredTags.Contains(tag))
{
Tag thistag = new Tag();
TagItem tgitem = new TagItem();
tgitem.settid(MainWindow.getTagNo() + 1);
//MainWindow._occurredTags.Add(translate(tag));
MainWindow._occurredTags.Add(tag);
//tgitem.setname(translate(tag));
tgitem.setname(tag);
thistag.saveTag(tgitem);
}
keywords += tag;
}
MainWindow.insertintosolr(MainWindow._title, "", MainWindow.curUrl, MainWindow._postdate, keywords, 0, body);
foreach (string tag in MainWindow._tags)
{
Preview b = new Preview();
PreviewItem item3 = new PreviewItem();
//MainWindow.curtid = MainWindow.getTagNo(translate(tag));
MainWindow.curtid = MainWindow.getTagNo(tag);
item3.settid(MainWindow.curtid);
item3.setwid(MainWindow.curwid);
//item3.setpreview(translate((string)tagPreview[tag]));
item3.setpreview((string)MainWindow.tagPreview[tag]);
//item3.setpreview((string)tagPreview[tag]);
b.savePreview(item3);
}
/*suiyuhao*/
}
}