/// <summary>
/// 用于推送的查询
/// </summary>
/// <param name="query">检索语句</param>
/// <param name="filter">等级不同与否</param>
/// <param name="addtime">添加时间</param>
/// <returns>检索结果</returns>
public string SearchForPush(string query, int filter, string addtime)
{
string str = "";
Xapian.MSet xm = searchforpush(query, filter, addtime);
if (!Directory.Exists(localdb))
{
Directory.CreateDirectory(localdb);
}
if (xm != null)
{
if (xm.Size() == 0)
{
return(null);
}
else
{
str = "订阅词" + query + "有" + xm.Size().ToString() + "条更新！\n";
int i = 0;
for (Xapian.MSetIterator iter = xm.Begin(); iter != xm.End(); ++iter)
{
//构造检索结果
Xapian.Document iterdoc = iter.GetDocument();
str = str + (++i).ToString() + " 、标题：" + iterdoc.GetValue(3) + "\n 链接：" + iterdoc.GetValue(1) + "\n";
}
}
}
else
{
return(null);
}
return(str);
}
/// <summary>
/// 在数据库中删除一篇文档
/// </summary>
/// <param name="dbname">数据库名</param>
/// <param name="hashcodelist">文章路径hashcode列表</param>
/// <returns>是否成功，成功为1，失败为0</returns>
public int delDocument(string dbname, List <string> hashcodelist)
{
string DBName = dbname;
try
{
Xapian.WritableDatabase database;
database = new Xapian.WritableDatabase(DBName, Xapian.Xapian.DB_CREATE_OR_OPEN);
foreach (var item in hashcodelist)
{
Xapian.Enquire enquire = new Xapian.Enquire(database);
//设置检索的前缀
Xapian.QueryParser qp = new Xapian.QueryParser();
qp.SetDatabase(database);
qp.SetDefaultOp(Xapian.Query.op.OP_ELITE_SET);
qp.SetStemmingStrategy(Xapian.QueryParser.stem_strategy.STEM_NONE);
//检索hash值
string querystr = item;
qp.AddPrefix("", "Q"); //hash前缀为Q
Xapian.Query query = qp.ParseQuery(querystr);
Console.WriteLine("query is" + query.GetDescription() + "\n");
//开始检索
enquire.SetQuery(query);
//返回结果
Xapian.MSet XapAns = enquire.GetMSet(0, int.MaxValue);
var a = XapAns.Size();
for (Xapian.MSetIterator iter = XapAns.Begin(); iter != XapAns.End(); ++iter)
{
Xapian.Document iterdoc = iter.GetDocument();
if (iterdoc.GetValue(VALUE_HASHCODE) != item) //防止hash检查出错
{
continue;
}
else
{
uint docid = iter.GetDocId(); //获取唯一id
database.DeleteDocument(docid); //删除文档
}
}
}
database.Commit(); //提交数据库
database.Close(); //关闭数据库
}
catch (Exception e)
{
log.Error(e.Message);
Console.Error.WriteLine("Exception: " + e.ToString());
return(0);
}
return(1);
}
/// <summary>
/// 更新一篇文章列表
/// </summary>
/// <param name="dbname">数据库路径</param>
/// <param name="list">文章列表</param>
/// <returns>是否成功，成功为1，失败为0</returns>
public int updateDocument(string dbname, List <xapIndex> list)
{
ChineseSeg cs = new ChineseSeg();
string DBName = dbname;
try
{
Xapian.WritableDatabase database;
database = new Xapian.WritableDatabase(DBName, Xapian.Xapian.DB_CREATE_OR_OPEN);
foreach (var item in list)
{
Xapian.Enquire enquire = new Xapian.Enquire(database);
//设置检索的前缀
Xapian.QueryParser qp = new Xapian.QueryParser();
qp.SetDatabase(database);
qp.SetDefaultOp(Xapian.Query.op.OP_ELITE_SET);
qp.SetStemmingStrategy(Xapian.QueryParser.stem_strategy.STEM_NONE);
//通过hash查找文章
string querystr = item.hashcode;
qp.AddPrefix("", "Q"); //hash前缀为Q
Xapian.Query query = qp.ParseQuery(querystr);
Console.WriteLine("query is" + query.GetDescription() + "\n");
//开始检索
enquire.SetQuery(query);
//返回结果
Xapian.MSet XapAns = enquire.GetMSet(0, int.MaxValue);
for (Xapian.MSetIterator iter = XapAns.Begin(); iter != XapAns.End(); ++iter)
{
Xapian.Document iterdoc = iter.GetDocument();
if (iterdoc.GetValue(VALUE_HASHCODE) != item.hashcode) //以防出现hash筛选错误
{
continue;
}
else
{
uint docid = iter.GetDocId(); //获取唯一id
Xapian.Document doc = new Xapian.Document();
Xapian.TermGenerator indexer = new Xapian.TermGenerator();
doc.SetData(HttpUtility.HtmlEncode(item.content)); //设置负载域
DateTime DateTimestart = DateTime.Now;
doc.AddValue(VALUE_TIME, DateTimestart.ToString("yyyy/MM/dd")); //插入时间
doc.AddValue(VALUE_AHREF, item.ahref); //原文链接
doc.AddValue(VALUE_LOCALINK, item.link); //本地链接
doc.AddValue(VALUE_TITLE, item.title); //文章标题
doc.AddValue(VALUE_SOURCE, item.source.ToString()); //来源类型
doc.AddValue(VALUE_SECLEVEL, item.seclevel.ToString()); //等级
doc.AddValue(VALUE_EXTENSION, item.extension.ToString()); //扩展名
doc.AddValue(VALUE_HASHCODE, item.hashcode); //hash
indexer.SetDocument(doc);
indexer.SetStemmingStrategy(Xapian.TermGenerator.stem_strategy.STEM_NONE); //设置不解析策略
string strcut = cs.JiebaSeg(item.content);
string titlecut = cs.JiebaSeg(item.title);
indexer.IndexText(strcut, 1, "C"); //设置内容前缀
indexer.IndexText(titlecut, 1, "T"); //设置标题前缀
indexer.IndexText(item.hashcode, 1, "Q"); //设置文档名hash
indexer.IndexText(item.ahref, 1, "A"); //设置链接前缀（用于推送文件夹订阅）
database.ReplaceDocument(docid, doc); //替换文档
}
}
}
database.Commit(); //提交数据库
database.Close(); //关闭数据库
}
catch (Exception e)
{
log.Error(e.Message);
Console.Error.WriteLine("Exception: " + e.ToString());
return(0);
}
return(1);
}
public static void Main()
{
try {
// Test the version number reporting functions give plausible
// results.
string v = "";
v += Xapian.Version.Major();
v += ".";
v += Xapian.Version.Minor();
v += ".";
v += Xapian.Version.Revision();
string v2 = Xapian.Version.String();
if (v != v2)
{
System.Console.WriteLine("Unexpected version output (" + v + " != " + v2 + ")");
System.Environment.Exit(1);
}
Xapian.Stem stem = new Xapian.Stem("english");
Xapian.Document doc = new Xapian.Document();
// Currently SWIG doesn't generate zero-byte clean code for
// transferring strings between C# and C++.
/*
* doc.SetData("a\0b");
* if (doc.GetData() == "a") {
* System.Console.WriteLine("GetData+SetData truncates at a zero byte");
* System.Environment.Exit(1);
* }
* if (doc.GetData() != "a\0b") {
* System.Console.WriteLine("GetData+SetData doesn't transparently handle a zero byte");
* System.Environment.Exit(1);
* }
*/
doc.SetData("is there anybody out there?");
doc.AddTerm("XYzzy");
doc.AddPosting(stem.Apply("is"), 1);
doc.AddPosting(stem.Apply("there"), 2);
doc.AddPosting(stem.Apply("anybody"), 3);
doc.AddPosting(stem.Apply("out"), 4);
doc.AddPosting(stem.Apply("there"), 5);
Xapian.WritableDatabase db = new Xapian.WritableDatabase("", Xapian.Xapian.DB_BACKEND_INMEMORY);
db.AddDocument(doc);
if (db.GetDocCount() != 1)
{
System.Environment.Exit(1);
}
if (doc.TermListCount() != 5)
{
System.Environment.Exit(1);
}
int count = 0;
Xapian.TermIterator i = doc.TermListBegin();
while (i != doc.TermListEnd())
{
++count;
++i;
}
if (count != 5)
{
System.Environment.Exit(1);
}
// Check exception handling for Xapian::DocNotFoundError.
try {
Xapian.Document doc2 = db.GetDocument(2);
System.Console.WriteLine("Retrieved non-existent document: " + doc2.ToString());
System.Environment.Exit(1);
} catch (System.Exception e) {
// We expect DocNotFoundError
if (e.Message.Substring(0, 16) != "DocNotFoundError")
{
System.Console.WriteLine("Unexpected exception from accessing non-existent document: " + e.Message);
System.Environment.Exit(1);
}
}
Xapian.QueryParser qp = new Xapian.QueryParser();
// Check QueryParser parsing error.
try {
qp.ParseQuery("test AND");
System.Console.WriteLine("Successfully parsed bad query");
System.Environment.Exit(1);
} catch (System.Exception e) {
if (e.Message != "QueryParserError: Syntax: <expression> AND <expression>")
{
System.Console.WriteLine("Exception string not as expected, got: '" + e.Message + "'");
System.Environment.Exit(1);
}
}
// FIXME: It would be better if the (uint) cast wasn't required here.
qp.ParseQuery("hello world", (uint)Xapian.QueryParser.feature_flag.FLAG_BOOLEAN);
// Test wrapping of null-able grouping parameter.
qp.AddBooleanPrefix("colour", "XC");
qp.AddBooleanPrefix("color", "XC");
qp.AddBooleanPrefix("foo", "XFOO", null);
qp.AddBooleanPrefix("bar", "XBAR", "XBA*");
qp.AddBooleanPrefix("baa", "XBAA", "XBA*");
// FIXME: It would be better if the (uint) cast wasn't required here.
Xapian.DateRangeProcessor rpdate = new Xapian.DateRangeProcessor(1, (uint)Xapian.Xapian.RP_DATE_PREFER_MDY, 1960);
qp.AddRangeprocessor(rpdate);
qp.AddRangeprocessor(rpdate, null);
qp.AddRangeprocessor(rpdate, "foo");
if (Xapian.Query.MatchAll.GetDescription() != "Query(<alldocuments>)")
{
System.Console.WriteLine("Unexpected Query.MatchAll.toString()");
System.Environment.Exit(1);
}
if (Xapian.Query.MatchNothing.GetDescription() != "Query()")
{
System.Console.WriteLine("Unexpected Query.MatchNothing.toString()");
System.Environment.Exit(1);
}
// Check that OP_ELITE_SET works (in 0.9.6 and earlier it had the
// wrong value in C#).
try {
Xapian.Query foo = new Xapian.Query(Xapian.Query.op.OP_OR, "hello", "world");
Xapian.Query foo2 = new Xapian.Query(Xapian.Query.op.OP_ELITE_SET, foo, foo);
foo = foo2; // Avoid "unused variable" warning.
} catch (System.Exception e) {
System.Console.WriteLine("Using OP_ELITE_SET cause exception '" + e.Message + "'");
System.Environment.Exit(1);
}
// Feature test for MatchDecider.
doc = new Xapian.Document();
doc.SetData("Two");
doc.AddPosting(stem.Apply("out"), 1);
doc.AddPosting(stem.Apply("source"), 2);
doc.AddValue(0, "yes");
db.AddDocument(doc);
Xapian.Query query = new Xapian.Query(stem.Apply("out"));
Xapian.Enquire enquire = new Xapian.Enquire(db);
// Check Xapian::BAD_VALUENO is wrapped suitably.
enquire.SetCollapseKey(Xapian.Xapian.BAD_VALUENO);
enquire.SetQuery(query);
Xapian.MSet mset = enquire.GetMSet(0, 10, null, new TestMatchDecider());
if (mset.Size() != 1)
{
System.Console.WriteLine("MatchDecider found " + mset.Size().ToString() + " documents, expected 1");
System.Environment.Exit(1);
}
if (mset.GetDocId(0) != 2)
{
System.Console.WriteLine("MatchDecider mset has wrong docid in");
System.Environment.Exit(1);
}
mset = enquire.GetMSet(0, 10);
for (Xapian.MSetIterator m = mset.Begin(); m != mset.End(); m++)
{
// In Xapian 1.2.6 and earlier, the iterator would become
// eligible for garbage collection after being advanced.
// It didn't actually get garbage collected often, but when
// it did, it caused a crash. Here we force a GC run to make
// this issue manifest if it is present.
System.GC.Collect();
System.GC.WaitForPendingFinalizers();
}
// Test setting and getting metadata
if (db.GetMetadata("Foo") != "")
{
System.Console.WriteLine("db.GetMetadata(\"Foo\") returned wrong value \"" + db.GetMetadata("Foo") + "\" - expected \"\"");
System.Environment.Exit(1);
}
db.SetMetadata("Foo", "Foo");
if (db.GetMetadata("Foo") != "Foo")
{
System.Console.WriteLine("db.GetMetadata(\"Foo\") returned wrong value \"" + db.GetMetadata("Foo") + "\" - expected \"Foo\"");
System.Environment.Exit(1);
}
// Test OP_SCALE_WEIGHT and corresponding constructor
Xapian.Query query4 = new Xapian.Query(Xapian.Query.op.OP_SCALE_WEIGHT, new Xapian.Query("foo"), 5.0);
if (query4.GetDescription() != "Query(5 * foo)")
{
System.Console.WriteLine("Unexpected query4.GetDescription()");
System.Environment.Exit(1);
}
} catch (System.Exception e) {
System.Console.WriteLine("Exception: " + e.ToString());
System.Environment.Exit(1);
}
}
/// <summary>
/// 检索结果
/// </summary>
/// <param name="query">检索词</param>
/// <param name="page">分页号</param>
/// <param name="filter">来源/等级不同过滤</param>
/// <param name="filetype">类型过滤</param>
/// <param name="num">返回结果数目</param>
/// <param name="XapResList">返回检索结果</param>
/// <param name="ts">返回检索时间</param>
public void SearchReturn(string query, int page, int filter, string filetype, out uint num, out List <SearchResult> XapResList, out TimeSpan ts)
{
DateTime DateTimestart = DateTime.Now;
DateTime DateTimeend;
Xapian.MSet xm;
XapResList = new List <SearchResult>();
{
query = query.Replace("\\", "");
query = query.Replace("/", "");
}
string querystr = _cs.JiebaSegnotSearch(query); //分词
if (!Directory.Exists(localdb))
{
Directory.CreateDirectory(localdb);
}
if (filetype == "1980/01/01")
{
xm = searchforpush(query, filter, filetype); //检索
}
else
{
xm = search(querystr, filter, filetype); //检索
}
//若返回不为空
if (xm != null)
{
num = xm.Size(); //结果数目
int pagecount = 0;
for (Xapian.MSetIterator iter = xm.Begin(); iter != xm.End(); ++iter)
{
SearchResult sr = new SearchResult();
++pagecount;
if (pagecount <= ((page - 1) * 10)) //获得分页
{
continue;
}
else
{
if (XapResList.Count >= 10) //每页10个结果
{
break;
}
Xapian.Document iterdoc = iter.GetDocument();
bool ftpflag = false; //ftp标记，转码用
bool emflag = false;
string strcontent = iterdoc.GetData(); //取出正文
string strtitle = iterdoc.GetValue(3); //取出标题 ValueTitle
string strahref = iterdoc.GetValue(1); //取出链接
string source = iterdoc.GetValue(0);
string strcut = "";
int contentlen = strcontent.Length; //判断正文长度，为下面筛选含有关键词片段做准备
uint docid = iter.GetDocId();
if (source == "4")
{
sr.allcontent = strcontent;
}
if (source == "2")
{
ftpflag = true;
strahref = UrlEncode(strahref); //若为ftp链接，需要转码
}
string[] strquerycut = querystr.Split(' ');
string emlink = "";
List <string> tmp = new List <string>();
foreach (var item in strquerycut)
{
if (item == "e" || item == "E" || item == "m" || item == "M" ||
item == "em" || item == "Em" || item == "Em" || item == "EM" ||
item == "<" || item == ">")
{
emflag = true;
if (emlink != "")
{
emlink = emlink + "|" + item;
}
else
{
emlink = item;
}
}
else
{
tmp.Add(item);
}
}
HashSet <string> hs = new HashSet <string>(tmp); //此时已经去掉重复的数据保存在hashset中
String[] strunique = new String[hs.Count];
hs.CopyTo(strunique);
int cutlen = strunique.Length;
int count = 0;
if (emlink != "" && cutlen == 0)
{
foreach (var item in strquerycut)
{
//消掉*问号空格
if (item == " " || item == "")
{
continue;
}
CompareInfo Compare = CultureInfo.InvariantCulture.CompareInfo;
int conpos = Compare.IndexOf(strcontent, item, CompareOptions.IgnoreCase); //根据位置标红
//int conpos = strcontent.IndexOf(item); //根据位置标红
if (conpos != -1)
{
if (contentlen - conpos > 150 && conpos > 50)
{
//截取150字作为cache
strcut = strcontent.Substring(conpos - 50, 200);
if (emflag)
{
strtitle = ReplaceCntent(emlink, strtitle);
strcut = ReplaceCntent(emlink, strcut);
}
strcut = "..." + strcut + "...";
goto Finally;
}
else if (conpos > 50)
{
////截取150字作为cache
strcut = strcontent.Substring(conpos - 50, contentlen - conpos + 50);
if (emflag)
{
strtitle = ReplaceCntent(emlink, strtitle);
strcut = ReplaceCntent(emlink, strcut);
}
strcut = "..." + strcut + "...";
goto Finally;
}
else if (contentlen - conpos > 150)
{
//截取150字作为cache
strcut = strcontent.Substring(0, conpos + 150);
if (emflag)
{
strtitle = ReplaceCntent(emlink, strtitle);
strcut = ReplaceCntent(emlink, strcut);
}
strcut = "..." + strcut + "...";
goto Finally;
}
else
{
//strcut = HttpUtility.HtmlEncode(strcut);
//不够150的全拿出
strcut = strcontent;
if (emflag)
{
strtitle = ReplaceCntent(emlink, strtitle);
strcut = ReplaceCntent(emlink, strcut);
}
strcut = "..." + strcut + "...";
goto Finally;
}
}
else
{
CompareInfo Comparetitle = CultureInfo.InvariantCulture.CompareInfo;
int conpostitle = Comparetitle.IndexOf(strtitle, item, CompareOptions.IgnoreCase); //根据位置标红
if (conpostitle != -1)
{
if (contentlen > 200)
{
strcut = strcontent.Substring(0, 200);
if (emflag)
{
strtitle = ReplaceCntent(emlink, strtitle);
strcut = ReplaceCntent(emlink, strcut);
}
strcut = "..." + strcut + "...";
goto Finally;
}
else
{
strcut = strcontent;
if (emflag)
{
strtitle = ReplaceCntent(emlink, strtitle);
strcut = ReplaceCntent(emlink, strcut);
}
strcut = "..." + strcut + "...";
goto Finally;
}
}
else
{
++count;
}
}
}
}
else
{
//每一个词都查一遍
foreach (var item in strunique)
{
//消掉*问号空格
if (item == " " || item == "")
{
continue;
}
CompareInfo Compare = CultureInfo.InvariantCulture.CompareInfo;
int conpos = Compare.IndexOf(strcontent, item, CompareOptions.IgnoreCase); //根据位置标红
//int conpos = strcontent.IndexOf(item); //根据位置标红
if (conpos != -1)
{
if (contentlen - conpos > 150 && conpos > 50)
{
//截取150字作为cache
strcut = strcontent.Substring(conpos - 50, 200);
if (emflag)
{
strtitle = ReplaceCntent(emlink, strtitle);
strcut = ReplaceCntent(emlink, strcut);
}
//strcut = HttpUtility.HtmlEncode(strcut);
for (; count < cutlen; count++)
{
if (strunique[count] == " " || strunique[count] == "")
{
continue;
}
//标红，大小写不敏感，regex替换法，replace大小写敏感
strtitle = Regex.Replace(strtitle, Regex.Escape(strunique[count]), "<em>" + strunique[count] + "</em>", RegexOptions.IgnoreCase);
//strtitle = strtitle.Replace(strquerycut[count], "<font color = red>" + strquerycut[count] + "</font>");
//strcut = strcut.Replace(strquerycut[count], "<font color = red>" + strquerycut[count] + "</font>");
strcut = Regex.Replace(strcut, Regex.Escape(strunique[count]), "<em>" + strunique[count] + "</em>", RegexOptions.IgnoreCase);
}
strcut = "..." + strcut + "...";
goto Finally;
}
else if (conpos > 50)
{
////截取150字作为cache
strcut = strcontent.Substring(conpos - 50, contentlen - conpos + 50);
if (emflag)
{
strtitle = ReplaceCntent(emlink, strtitle);
strcut = ReplaceCntent(emlink, strcut);
}
//strcut = HttpUtility.HtmlEncode(strcut);
for (; count < cutlen; count++)
{
if (strunique[count] == " " || strunique[count] == "")
{
continue;
}
//标红
strtitle = Regex.Replace(strtitle, Regex.Escape(strunique[count]), "<em>" + strunique[count] + "</em>", RegexOptions.IgnoreCase);
//strtitle = strtitle.Replace(strquerycut[count], "<font color = red>" + strquerycut[count] + "</font>");
//strcut = strcut.Replace(strquerycut[count], "<font color = red>" + strquerycut[count] + "</font>");
strcut = Regex.Replace(strcut, Regex.Escape(strunique[count]), "<em>" + strunique[count] + "</em>", RegexOptions.IgnoreCase);
}
strcut = "..." + strcut + "...";
goto Finally;
}
else if (contentlen - conpos > 150)
{
//截取150字作为cache
strcut = strcontent.Substring(0, conpos + 150);
if (emflag)
{
strtitle = ReplaceCntent(emlink, strtitle);
strcut = ReplaceCntent(emlink, strcut);
}
//strcut = HttpUtility.HtmlEncode(strcut);
for (; count < cutlen; count++)
{
if (strunique[count] == " " || strunique[count] == "")
{
continue;
}
//标红
strtitle = Regex.Replace(strtitle, Regex.Escape(strunique[count]), "<em>" + strunique[count] + "</em>", RegexOptions.IgnoreCase);
//strtitle = strtitle.Replace(strquerycut[count], "<font color = red>" + strquerycut[count] + "</font>");
//strcut = strcut.Replace(strquerycut[count], "<font color = red>" + strquerycut[count] + "</font>");
strcut = Regex.Replace(strcut, Regex.Escape(strunique[count]), "<em>" + strunique[count] + "</em>", RegexOptions.IgnoreCase);
}
strcut = "..." + strcut + "...";
goto Finally;
}
else
{
//strcut = HttpUtility.HtmlEncode(strcut);
//不够150的全拿出
strcut = strcontent;
if (emflag)
{
strtitle = ReplaceCntent(emlink, strtitle);
strcut = ReplaceCntent(emlink, strcut);
}
for (; count < cutlen; count++)
{
if (strunique[count] == " " || strunique[count] == "")
{
continue;
}
//标红
strtitle = Regex.Replace(strtitle, Regex.Escape(strunique[count]), "<em>" + strunique[count] + "</em>", RegexOptions.IgnoreCase);
//strtitle = strtitle.Replace(strquerycut[count], "<font color = red>" + strquerycut[count] + "</font>");
//strcut = strcut.Replace(strquerycut[count], "<font color = red>" + strquerycut[count] + "</font>");
strcut = Regex.Replace(strcut, Regex.Escape(strunique[count]), "<em>" + strunique[count] + "</em>", RegexOptions.IgnoreCase);
}
strcut = "..." + strcut + "...";
goto Finally;
}
}
else
{
CompareInfo Comparetitle = CultureInfo.InvariantCulture.CompareInfo;
int conpostitle = Comparetitle.IndexOf(strtitle, item, CompareOptions.IgnoreCase); //根据位置标红
if (conpostitle != -1)
{
if (contentlen > 200)
{
strcut = strcontent.Substring(0, 200);
if (emflag)
{
strtitle = ReplaceCntent(emlink, strtitle);
strcut = ReplaceCntent(emlink, strcut);
}
//strcut = HttpUtility.HtmlEncode(strcut);
for (; count < cutlen; count++)
{
if (strunique[count] == " " || strunique[count] == "")
{
continue;
}
//标红
strtitle = Regex.Replace(strtitle, Regex.Escape(strunique[count]), "<em>" + strunique[count] + "</em>", RegexOptions.IgnoreCase);
//strtitle = strtitle.Replace(strquerycut[count], "<font color = red>" + strquerycut[count] + "</font>");
//strcut = strcut.Replace(strquerycut[count], "<font color = red>" + strquerycut[count] + "</font>");
strcut = Regex.Replace(strcut, Regex.Escape(strunique[count]), "<em>" + strunique[count] + "</em>", RegexOptions.IgnoreCase);
}
strcut = "..." + strcut + "...";
goto Finally;
}
else
{
strcut = strcontent;
if (emflag)
{
strtitle = ReplaceCntent(emlink, strtitle);
strcut = ReplaceCntent(emlink, strcut);
}
//strcut = HttpUtility.HtmlEncode(strcut);
for (; count < cutlen; count++)
{
if (strunique[count] == " " || strunique[count] == "")
{
continue;
}
//标红
strtitle = Regex.Replace(strtitle, Regex.Escape(strunique[count]), "<em>" + strunique[count] + "</em>", RegexOptions.IgnoreCase);
//strtitle = strtitle.Replace(strquerycut[count], "<font color = red>" + strquerycut[count] + "</font>");
//strcut = strcut.Replace(strquerycut[count], "<font color = red>" + strquerycut[count] + "</font>");
strcut = Regex.Replace(strcut, Regex.Escape(strunique[count]), "<em>" + strunique[count] + "</em>", RegexOptions.IgnoreCase);
}
strcut = "..." + strcut + "...";
goto Finally;
}
}
else
{
++count;
}
}
}
}
//找到合适的内容之后返回结果
Finally:
sr.ahref = iterdoc.GetValue(1);
if (ftpflag) //判断是否需要转码链接
{
sr.ahrefencode = strahref; //ftp则使用转码链接
}
else
{
sr.ahrefencode = sr.ahref;
}
sr.link = iterdoc.GetValue(2);
sr.title = strtitle;
sr.snippet = strcut;
XapResList.Add(sr);
}
}
}
else
{
num = 0;
}
DateTimeend = DateTime.Now;
ts = DateTimeend - DateTimestart;
ts.TotalMilliseconds.ToString(); //查询时间返回
}
private void SearchReturn(Xapian.MSet xm, SearchInfo Searchwords, out AnsInfo ai)
{
ai = new AnsInfo();
List <SearchResult> XapResList = new List <SearchResult>();
string query = Searchwords.SearchString;
query = query.Replace("\\", "");
query = query.Replace("/", "");
int page = Searchwords.page;
var segmenter = new JiebaSegmenter();
var segments = segmenter.Cut(query);
string querystr = string.Join(" ", segments); //分词
//若返回不为空
if (xm != null)
{
ai.totalnum = xm.Size(); //结果数目
int pagecount = 0;
for (Xapian.MSetIterator iter = xm.Begin(); iter != xm.End(); ++iter)
{
SearchResult sr = new SearchResult();
++pagecount;
if (pagecount <= ((page - 1) * 10)) //获得分页
{
continue;
}
else
{
if (XapResList.Count >= 10) //每页10个结果
{
break;
}
Xapian.Document iterdoc = iter.GetDocument();
bool ftpflag = false; //ftp标记，转码用
bool emflag = false;
string strcontent = iterdoc.GetData(); //取出正文
string strtitle = iterdoc.GetValue(3); //取出标题 ValueTitle
string strahref = iterdoc.GetValue(1); //取出链接
string source = iterdoc.GetValue(0);
string strcut = "";
int contentlen = strcontent.Length; //判断正文长度，为下面筛选含有关键词片段做准备
uint docid = iter.GetDocId();
if (source == "4")
{
sr.allcontent = strcontent;
}
if (source == "2")
{
ftpflag = true;
strahref = UrlEncode(strahref); //若为ftp链接，需要转码
}
string[] strquerycut = querystr.Split(' ');
string emlink = "";
List <string> tmp = new List <string>();
foreach (var item in strquerycut)
{
if (item == "e" || item == "E" || item == "m" || item == "M" ||
item == "em" || item == "Em" || item == "Em" || item == "EM" ||
item == "<" || item == ">")
{
emflag = true;
if (emlink != "")
{
emlink = emlink + "|" + item;
}
else
{
emlink = item;
}
}
else
{
tmp.Add(item);
}
}
HashSet <string> hs = new HashSet <string>(tmp); //此时已经去掉重复的数据保存在hashset中
String[] strunique = new String[hs.Count];
hs.CopyTo(strunique);
int cutlen = strunique.Length;
int count = 0;
if (emlink != "" && cutlen == 0)
{
foreach (var item in strquerycut)
{
//消掉*问号空格
if (item == " " || item == "")
{
continue;
}
CompareInfo Compare = CultureInfo.InvariantCulture.CompareInfo;
int conpos = Compare.IndexOf(strcontent, item, CompareOptions.IgnoreCase); //根据位置标红
//int conpos = strcontent.IndexOf(item); //根据位置标红
if (conpos != -1)
{
if (contentlen - conpos > 150 && conpos > 50)
{
//截取150字作为cache
strcut = strcontent.Substring(conpos - 50, 200);
if (emflag)
{
strtitle = ReplaceCntent(emlink, strtitle);
strcut = ReplaceCntent(emlink, strcut);
}
strcut = "..." + strcut + "...";
goto Finally;
}
else if (conpos > 50)
{
////截取150字作为cache
strcut = strcontent.Substring(conpos - 50, contentlen - conpos + 50);
if (emflag)
{
strtitle = ReplaceCntent(emlink, strtitle);
strcut = ReplaceCntent(emlink, strcut);
}
strcut = "..." + strcut + "...";
goto Finally;
}
else if (contentlen - conpos > 150)
{
//截取150字作为cache
strcut = strcontent.Substring(0, conpos + 150);
if (emflag)
{
strtitle = ReplaceCntent(emlink, strtitle);
strcut = ReplaceCntent(emlink, strcut);
}
strcut = "..." + strcut + "...";
goto Finally;
}
else
{
//strcut = HttpUtility.HtmlEncode(strcut);
//不够150的全拿出
strcut = strcontent;
if (emflag)
{
strtitle = ReplaceCntent(emlink, strtitle);
strcut = ReplaceCntent(emlink, strcut);
}
strcut = "..." + strcut + "...";
goto Finally;
}
}
else
{
CompareInfo Comparetitle = CultureInfo.InvariantCulture.CompareInfo;
int conpostitle = Comparetitle.IndexOf(strtitle, item, CompareOptions.IgnoreCase); //根据位置标红
if (conpostitle != -1)
{
if (contentlen > 200)
{
strcut = strcontent.Substring(0, 200);
if (emflag)
{
strtitle = ReplaceCntent(emlink, strtitle);
strcut = ReplaceCntent(emlink, strcut);
}
strcut = "..." + strcut + "...";
goto Finally;
}
else
{
strcut = strcontent;
if (emflag)
{
strtitle = ReplaceCntent(emlink, strtitle);
strcut = ReplaceCntent(emlink, strcut);
}
strcut = "..." + strcut + "...";
goto Finally;
}
}
else
{
++count;
}
}
}
}
else
{
//每一个词都查一遍
foreach (var item in strunique)
{
//消掉*问号空格
if (item == " " || item == "")
{
continue;
}
CompareInfo Compare = CultureInfo.InvariantCulture.CompareInfo;
int conpos = Compare.IndexOf(strcontent, item, CompareOptions.IgnoreCase); //根据位置标红
//int conpos = strcontent.IndexOf(item); //根据位置标红
if (conpos != -1)
{
if (contentlen - conpos > 150 && conpos > 50)
{
//截取150字作为cache
strcut = strcontent.Substring(conpos - 50, 200);
if (emflag)
{
strtitle = ReplaceCntent(emlink, strtitle);
strcut = ReplaceCntent(emlink, strcut);
}
//strcut = HttpUtility.HtmlEncode(strcut);
for (; count < cutlen; count++)
{
if (strunique[count] == " " || strunique[count] == "")
{
continue;
}
//标红，大小写不敏感，regex替换法，replace大小写敏感
strtitle = Regex.Replace(strtitle, Regex.Escape(strunique[count]), "<em>" + strunique[count] + "</em>", RegexOptions.IgnoreCase);
//strtitle = strtitle.Replace(strquerycut[count], "<font color = red>" + strquerycut[count] + "</font>");
//strcut = strcut.Replace(strquerycut[count], "<font color = red>" + strquerycut[count] + "</font>");
strcut = Regex.Replace(strcut, Regex.Escape(strunique[count]), "<em>" + strunique[count] + "</em>", RegexOptions.IgnoreCase);
}
strcut = "..." + strcut + "...";
goto Finally;
}
else if (conpos > 50)
{
////截取150字作为cache
strcut = strcontent.Substring(conpos - 50, contentlen - conpos + 50);
if (emflag)
{
strtitle = ReplaceCntent(emlink, strtitle);
strcut = ReplaceCntent(emlink, strcut);
}
//strcut = HttpUtility.HtmlEncode(strcut);
for (; count < cutlen; count++)
{
if (strunique[count] == " " || strunique[count] == "")
{
continue;
}
//标红
strtitle = Regex.Replace(strtitle, Regex.Escape(strunique[count]), "<em>" + strunique[count] + "</em>", RegexOptions.IgnoreCase);
//strtitle = strtitle.Replace(strquerycut[count], "<font color = red>" + strquerycut[count] + "</font>");
//strcut = strcut.Replace(strquerycut[count], "<font color = red>" + strquerycut[count] + "</font>");
strcut = Regex.Replace(strcut, Regex.Escape(strunique[count]), "<em>" + strunique[count] + "</em>", RegexOptions.IgnoreCase);
}
strcut = "..." + strcut + "...";
goto Finally;
}
else if (contentlen - conpos > 150)
{
//截取150字作为cache
strcut = strcontent.Substring(0, conpos + 150);
if (emflag)
{
strtitle = ReplaceCntent(emlink, strtitle);
strcut = ReplaceCntent(emlink, strcut);
}
//strcut = HttpUtility.HtmlEncode(strcut);
for (; count < cutlen; count++)
{
if (strunique[count] == " " || strunique[count] == "")
{
continue;
}
//标红
strtitle = Regex.Replace(strtitle, Regex.Escape(strunique[count]), "<em>" + strunique[count] + "</em>", RegexOptions.IgnoreCase);
//strtitle = strtitle.Replace(strquerycut[count], "<font color = red>" + strquerycut[count] + "</font>");
//strcut = strcut.Replace(strquerycut[count], "<font color = red>" + strquerycut[count] + "</font>");
strcut = Regex.Replace(strcut, Regex.Escape(strunique[count]), "<em>" + strunique[count] + "</em>", RegexOptions.IgnoreCase);
}
strcut = "..." + strcut + "...";
goto Finally;
}
else
{
//strcut = HttpUtility.HtmlEncode(strcut);
//不够150的全拿出
strcut = strcontent;
if (emflag)
{
strtitle = ReplaceCntent(emlink, strtitle);
strcut = ReplaceCntent(emlink, strcut);
}
for (; count < cutlen; count++)
{
if (strunique[count] == " " || strunique[count] == "")
{
continue;
}
//标红
strtitle = Regex.Replace(strtitle, Regex.Escape(strunique[count]), "<em>" + strunique[count] + "</em>", RegexOptions.IgnoreCase);
//strtitle = strtitle.Replace(strquerycut[count], "<font color = red>" + strquerycut[count] + "</font>");
//strcut = strcut.Replace(strquerycut[count], "<font color = red>" + strquerycut[count] + "</font>");
strcut = Regex.Replace(strcut, Regex.Escape(strunique[count]), "<em>" + strunique[count] + "</em>", RegexOptions.IgnoreCase);
}
strcut = "..." + strcut + "...";
goto Finally;
}
}
else
{
CompareInfo Comparetitle = CultureInfo.InvariantCulture.CompareInfo;
int conpostitle = Comparetitle.IndexOf(strtitle, item, CompareOptions.IgnoreCase); //根据位置标红
if (conpostitle != -1)
{
if (contentlen > 200)
{
strcut = strcontent.Substring(0, 200);
if (emflag)
{
strtitle = ReplaceCntent(emlink, strtitle);
strcut = ReplaceCntent(emlink, strcut);
}
//strcut = HttpUtility.HtmlEncode(strcut);
for (; count < cutlen; count++)
{
if (strunique[count] == " " || strunique[count] == "")
{
continue;
}
//标红
strtitle = Regex.Replace(strtitle, Regex.Escape(strunique[count]), "<em>" + strunique[count] + "</em>", RegexOptions.IgnoreCase);
//strtitle = strtitle.Replace(strquerycut[count], "<font color = red>" + strquerycut[count] + "</font>");
//strcut = strcut.Replace(strquerycut[count], "<font color = red>" + strquerycut[count] + "</font>");
strcut = Regex.Replace(strcut, Regex.Escape(strunique[count]), "<em>" + strunique[count] + "</em>", RegexOptions.IgnoreCase);
}
strcut = "..." + strcut + "...";
goto Finally;
}
else
{
strcut = strcontent;
if (emflag)
{
strtitle = ReplaceCntent(emlink, strtitle);
strcut = ReplaceCntent(emlink, strcut);
}
//strcut = HttpUtility.HtmlEncode(strcut);
for (; count < cutlen; count++)
{
if (strunique[count] == " " || strunique[count] == "")
{
continue;
}
//标红
strtitle = Regex.Replace(strtitle, Regex.Escape(strunique[count]), "<em>" + strunique[count] + "</em>", RegexOptions.IgnoreCase);
//strtitle = strtitle.Replace(strquerycut[count], "<font color = red>" + strquerycut[count] + "</font>");
//strcut = strcut.Replace(strquerycut[count], "<font color = red>" + strquerycut[count] + "</font>");
strcut = Regex.Replace(strcut, Regex.Escape(strunique[count]), "<em>" + strunique[count] + "</em>", RegexOptions.IgnoreCase);
}
strcut = "..." + strcut + "...";
goto Finally;
}
}
else
{
++count;
}
}
}
}
//找到合适的内容之后返回结果
Finally:
sr.ahref = iterdoc.GetValue(1);
if (ftpflag) //判断是否需要转码链接
{
sr.ahrefencode = strahref; //ftp则使用转码链接
}
else
{
sr.ahrefencode = sr.ahref;
}
sr.link = iterdoc.GetValue(2);
sr.title = strtitle;
sr.snippet = strcut;
XapResList.Add(sr);
}
}
ai.retinfo = XapResList;
}
else
{
ai.totalnum = 0;
ai.retinfo = null;
}
}
