void ReportAbandonURI(UriResource u, string reason)
{
try
{
AbandonLinks++;
lbAbandonLinks.Text = strAbandonLinks + AbandonLinks;
AbandonUri a = new AbandonUri();
a.Info = u;
a.Reason = reason;
AbandonLinkList.Add(a);
if (AbandonLinkList.Count > 100)
{
AbandonLinkList.RemoveAt(0);
}
try
{
a.Keyword = hunter.projectInfo.strategy.GetKeyword(u.Keyword);
}
catch
{
a.Keyword = "[获取出错]";
}
WriteAbandonUri(a);
}
catch (Exception ex) { WriteException(ex); }
}
public void ReportAbandonURI(UriResource u, string reason)
{
if (onReportAbandonURI != null)
{
onReportAbandonURI(u, reason);
}
}
/// <summary>
/// 下载指定uri中的所有文件。如果为null表示跳过下载，为Empty表示下载没有问题，为路径表示下载不成功。
/// </summary>
/// <param name="uri">文件资源定位</param>
/// <returns>下载信息</returns>
public DownloadInfo DownloadFile(UriResource uriRes, Hunter h, HunterDownloadThread thisThread)
{
Database db = null;
if (h.projectInfo.DatabaseHelper != null)
{
db = h.projectInfo.DatabaseHelper.GetDatabaseInstance();
}
flowCalculator.Interval = 1000;
flowCalculator.Elapsed += new ElapsedEventHandler(flowCalculator_Elapsed);
try
{
HunterWebClient wc = new HunterWebClient();
#region 判断文件是否重复
bool isExist = false;
//检查文件在本地是否重复
isExist = database.LinkExists(uriRes.Url);
//如果文件不重复，而又为网络模式，则要检查数据库内的内容
if (!isExist && (h.projectInfo.CurrentMode == ProjectInfo.HunterMode.network)) //网络模式需要对比数据库和HunterXML
{
if (db == null)
{
hunterConsole.WriteException(new Exception("数据库连接失败，使用本地模式判重：" + uriRes.Url));
}
else
{
bool OpenFailed = false;
try
{
db.DbOpen();
}
catch (Exception ex)
{
hunterConsole.WriteException(ex);
}
bool KRESULT = db.IsRecordExists("tb_file_infos", out isExist, new FieldValue("file_link", uriRes.Url.Replace("'", "\\'").Replace("\"", "\\\"")));
if (!KRESULT || OpenFailed)
{
//hunterConsole.WriteException(new Exceptions("数据库连接失败，使用本地模式判重：" + uriRes.Url));
isExist = false;
}
try
{
db.DbClose();
}
catch (Exception ex)
{
hunterConsole.WriteException(ex);
}
}
}
#endregion
if (isExist)
#region 链接重复对应的措施
{
hunterConsole.ReportAbandonURI(uriRes, "链接重复");
return(null);
}
#endregion
else
{
#region 载部分
try
{
hunterConsole.WriteDownload("正在下载文件：" + uriRes.Url);
hunterConsole.WriteDownload(
"线程ID：" + Thread.CurrentThread.ManagedThreadId + Environment.NewLine +
"正在下载的文件：" + Environment.NewLine +
"下载地址：" + uriRes.Url + Environment.NewLine +
"下载的关键字：" + strategy.GetKeyword(uriRes.Keyword) + Environment.NewLine +
"下载的页面页码：" + uriRes.index);
}
catch (Exception ex)
{
hunterConsole.WriteException(ex);
}
wc.DownloadFileCompleted += new System.ComponentModel.AsyncCompletedEventHandler(wc_DownloadFileCompleted); //绑定文件下载事件
wc.DownloadProgressChanged += new DownloadProgressChangedEventHandler(wc_DownloadProgressChanged); //绑定下载进度改变事件
//临时文件命名
string filepath = HunterUtilities.GetFilenameFromUrl(pInfo, strategy, uriRes); //获取合适的文件名
int timeout = 0;
flowCalculator.Start(); //开始计算流量
receive1 = 0; //最开始第一次获得的数据量为0
wc.DownloadKeyword = strategy.GetKeyword(uriRes.Keyword);
wc.DownloadSource = uriRes.Url;
wc.DownloadDestination = filepath;
if (!Directory.Exists(Path.GetDirectoryName(filepath)))
{
Directory.CreateDirectory(Path.GetDirectoryName(filepath));
}
wc.DownloadFileAsync(new Uri(uriRes.Url), filepath); //开始下载
bool downloadProblem = false;
while (wc.IsBusy)
{
Thread.Sleep(1000); //使用进程休眠
timeout++;
if (timeout >= pInfo.timeout || h.DownloadCancelled) //如果超时或下载取消
{
wc.CancelAsync();
wc.Dispose();
if (!downloadProblem)
{
hunterConsole.WriteDownload(
"线程ID：" + Thread.CurrentThread.ManagedThreadId + Environment.NewLine +
"下载超时，取消下载。");
try
{
hunterConsole.ReportAbandonDownloadInfo(new DownloadInfo("", uriRes.Keyword, uriRes.index, uriRes.Url, "", true, "超时", hunterProject.projectInfo.strategy.GetKeyword(uriRes.Keyword))
, "超时");
}
catch { }
}
downloadProblem = true;
}
}
flowCalculator.Stop();
receive1 = 0;
receive2 = 0; //清空流量计
hunterConsole.outputSpeedInfo(DateTime.Now, 0); //流量计清零
wc.Dispose();
#endregion
if (!downloadProblem)
{ //如果没有下载问题
string MD5 = string.Empty;
try
{
MD5 = HunterUtilities.GetMD5Hash(filepath);
}
catch (Exception e)
{
hunterConsole.WriteException(new Exception("无法获取MD5。"));
hunterConsole.WriteException(e);
}
#region 文件MD5是否重复
bool isDuplicate; //记录文件是否重复
//判断是否与本地XML重复
isDuplicate = database.isDuplicate(uriRes.Url, MD5);
if (!isDuplicate && (h.projectInfo.CurrentMode == ProjectInfo.HunterMode.network)) //网络模式需要对比数据库和HunterXML
{
if (db == null)
{
hunterConsole.WriteException(new Exception("数据库连接失败，使用本地模式判重：" + MD5));
}
else
{
bool OpenFailed = false;
try
{
db.DbOpen();
}
catch (Exception ex)
{
hunterConsole.WriteException(ex);
}
bool KRESULT = db.IsFileExists("tb_file_infos", out isDuplicate, MD5);
if (!KRESULT || OpenFailed)
{
hunterConsole.WriteException(new Exception("数据库连接失败，使用本地模式判重：" + MD5));
isDuplicate = false;
}
try
{
db.DbClose();
}
catch (Exception ex)
{
hunterConsole.WriteException(ex);
}
}
}
#endregion
#region 文件重复、不重复对应的动作
if (!isDuplicate)
{ //检测是否重复。如果不重复则入库
wc.XMLFile = Path.Combine(hunterProject.projectInfo.filefolder, "$__" + Path.GetFileName(wc.DownloadDestination)) + ".xml";
try
{
HunterUtilities.WriteDownloadFileXML(wc.DownloadSource, wc.DownloadKeyword, Path.GetFileName(wc.DownloadDestination), (hunterProject.projectInfo.search_language == ProjectInfo.Language.none ? null : hunterProject.projectInfo.search_language.ToString()),
wc.XMLFile);
}
catch (Exception ex)
{
hunterConsole.WriteException(ex);
}
database.addNewRecord(uriRes.Url, wc.DownloadKeyword, filepath, MD5);
DownloadInfo d = new DownloadInfo(filepath, uriRes.Keyword, uriRes.index, uriRes.Url, MD5, false, "已下载", hunterProject.projectInfo.strategy.GetKeyword(uriRes.Keyword));
//Network模式：自动上传样张
if (h.projectInfo.CurrentMode == ProjectInfo.HunterMode.network)
#region 网络模式上传样张
{
String t_md5 = HunterUtilities.GetMD5Hash(filepath); //获得文件MD5码
String filename = t_md5 + "_" + Path.GetFileName(filepath);
String combinedPath = Path.Combine(pInfo.share_remote_path, "cache", ProjectInfo.IP_ADDRESS + " (" + pInfo.name + ")");
String combinedFullPath = Path.Combine(combinedPath, filename);
if (!Directory.Exists(combinedPath))
{
Directory.CreateDirectory(combinedPath);
}
bool fileMoveSuccess = false;
const int maxMoveCount = 5;
int moveCount = 0;
try
{
if (File.Exists(wc.XMLFile))
{
File.Delete(Path.Combine(combinedPath, Path.GetFileName(wc.XMLFile)));
File.Move(wc.XMLFile, Path.Combine(combinedPath, Path.GetFileName(wc.XMLFile)));
}
}
catch (Exception e)
{
hunterConsole.WriteException(e);
}
while (!fileMoveSuccess)
{
try
{
if (moveCount > maxMoveCount)
{
break;
}
File.Move(filepath, combinedFullPath);
fileMoveSuccess = true;
}
catch (Exception e)
{
moveCount++;
hunterConsole.WriteException(e);
}
}
}
#endregion
count++;
hunterConsole.outputDownloadedFileNum(DateTime.Now, count);
hunterConsole.ReportDownloadInfo(d);
return(d);
}
else
{
DownloadInfo d = new DownloadInfo(filepath, uriRes.Keyword,
uriRes.index, uriRes.Url, MD5, true, "MD5重复", hunterProject.projectInfo.strategy.GetKeyword(uriRes.Keyword));
hunterConsole.ReportAbandonDownloadInfo(d, "MD5重复");
return(d); //删除文件
}
#endregion
}
return(new DownloadInfo(filepath, uriRes.Keyword, uriRes.index,
uriRes.Url, null, true, "重复", hunterProject.projectInfo.strategy.GetKeyword(uriRes.Keyword))); //删除文件
}
}
catch (Exception e)
{
//*此处预留错误处理
hunterConsole.WriteException(e);
return(null);
}
}
/// <summary>
/// 捕获带有filetype结尾的链接。proxy为代理，为null表示不使用代理。
/// </summary>
/// <returns>返回一个本次页面中捕获的链接序列</returns>
public List <string> HuntUris(HunterProxy proxy, HunterForm main)
{
Regex linkReg = null;
String htmlCode = null;
List <string> thisURL = new List <string>(); //记录本次匹配的所有URL项
try
{
linkReg = new Regex(strategy.StrategyData.configuration.Regex); //超链接+超链接文本
WebProxy webproxy;
if (proxy != null)
{
webproxy = new WebProxy(proxy.IPAndPort);
}
else
{
webproxy = null;
}
if (proxy != null)
{
mHunterConsole.WriteDetails("正在使用代理：" + proxy.IPAndPort + "(" + proxy.Description + ")");
}
mHunterConsole.WriteDetails("准备分析页面：" + urlAddress);
htmlCode = GetPageHtml(webproxy, main);
mHunterConsole.WriteHTML(htmlCode);
}
catch (WebException ex) //如果是返回超时，返回一个Count>0的随机结果
{
thisURL.Add("{/WebException/}" + new Random().Next().ToString());
mHunterConsole.WriteDetails("页面" + urlAddress + "请求失败。原因：" + ex.Message);
mHunterConsole.ReportAbandonURI(new UriResource(urlAddress, strategy.CurrentKeywordProgress,
strategy.CurrentSearchProgress, null), ex.Message);
return(thisURL);
}
catch (Exception ex)
{
mHunterConsole.WriteException(ex);
}
try
{
Match m = linkReg.Match(htmlCode);
while (m.Success)
{
allCount++;
mHunterConsole.outputAnalysedUris(DateTime.Now, allCount);
//得到一个网址后，保存起来
string linkText = m.Result("${text}");
thisURL.Add(linkText); //记录本次获取到的linkText
if (strategy.HasForbiddenWord(linkText))
{
m = m.NextMatch();
continue; //如果含有违禁词语 则放弃下载 继续下一个
}
//对linkText中的内容进行处理，去掉里面的尖括号
Regex r = new Regex("<(.*?)>");
linkText = r.Replace(linkText, "");
string uri = null;
try
{
uri = (strategy.StrategyData.configuration.Redirect.ToLower() == "true") ? GetTheRedirectUrl(m.Result("${url}")) : (m.Result("${url}"));
}
catch (WebException)
{
mHunterConsole.WriteDetails("链接" + (m.Result("${url}") + "重定向超时。"));
mHunterConsole.ReportAbandonURI(new UriResource((m.Result("${url}")), strategy.CurrentKeywordProgress,
strategy.CurrentSearchProgress, null), "重定向超时");
m = m.NextMatch();
continue;
}
if (uri.EndsWith("." + strategy.Filetype))
{
availableCount++;
mHunterConsole.outputAvailableUris(DateTime.Now, availableCount);
UriResource u = new UriResource(uri, strategy.CurrentKeywordProgress
, strategy.CurrentSearchProgress, linkText); //封装成一个Uri资源
if (!uriQueue.Contains(u)) //考虑在多线程中，可能会出现重复项目
{
uriQueue.Enqueue(u); //将一个资源放入队列
}
mHunterConsole.outputDownloadingUriInfo(DateTime.Now,
"找到的资源的URL:" + u.Url + Environment.NewLine +
"标题：" + u.Text + Environment.NewLine +
"关键字：" + strategy.GetKeyword(u.Keyword) + Environment.NewLine +
"搜索页码：" + u.index + Environment.NewLine +
"已列入下载队列。");
mHunterConsole.WriteDetails("正在获得有效URI：" + uri);
}
m = m.NextMatch();
}
}
catch (Exception ex)
{
mHunterConsole.WriteException(ex);
}
return(thisURL);
}
/// <summary>
/// 从UriResource获得合适的保存文件名
/// </summary>
/// <param name="pInfo"></param>
/// <param name="strategy"></param>
/// <param name="uriRes"></param>
/// <returns></returns>
public static string GetFilenameFromUrl(ProjectInfo pInfo, Strategy strategy, UriResource uriRes)
{
/* 得到文件名的3套策略：
* 1、获得超链接的标题
* 2、如果超链接的标题有乱码，则获得网络路径
* 3、如果不能正常匹配到网络路径，则以时间命名（一般不会出现这种情况）
*/
//删除某些标记，并Html解码，然后合法化用户名
string filepath, basicName = HunterUtilities.LegalizeFile(
WebUtility.HtmlDecode(uriRes.Text)
);
if (strategy.HasConfusionString(basicName)) //如果含有违禁词（如乱码）
{
Regex regFilename = new Regex("^http://(.*)/(?<filename>.*)\\." + strategy.Filetype);
Match mFilename = regFilename.Match(uriRes.Url); //获得网络路径
if (mFilename.Success)
{
basicName = mFilename.Result("${filename}");
}
else
{
basicName = DateTime.Now.ToFileTime().ToString();
}
}
//文件判定重复措施
filepath = Path.Combine(pInfo.filefolder, basicName + "." + strategy.Filetype);
int i = 0;
bool keepOriginal = true; string tempName = null;
while (File.Exists(filepath))
{
i++;
keepOriginal = false;
tempName = basicName + "_" + i.ToString();
filepath = Path.Combine(pInfo.filefolder, tempName + "." + strategy.Filetype);
}
;
if (!keepOriginal)
{
basicName = tempName; //出现重名，则在后面添加编号
}
filepath = Path.Combine(pInfo.filefolder, basicName + "." + strategy.Filetype);
return(filepath);
}
