C＃需要将word,Excel等office文件及pdf文件以何种格式存入oracle才可以用于全文检索？

Question

string connectString = "Data Source=ydd;User ID=ydd;password=aaa";
        /// <summary>
        /// ear_id表的主键，ear：blob/clob型字段，用于存放电子档案
        /// </summary>
        string sqlAr_e = "Select ear_id, seq_dir, ear_name,"
                                    +" ear_tp, ear_size, ear_date, ear_creator,"
                                    +" ear_desc, dply_no,ear from ar_e";/// <summary>
        /// 将本地文件存入DataSet提交到服务器
        /// </summary>
        /// <param name="sender"></param>
        /// <param name="e"></param>
        private void btnUp_Click(object sender, EventArgs e)
        {
            byte[] bFile = GetStreamBytes(txtPath.Text);//将电子档案转换成byte[]类型；            if (bFile == null)
            {
                return;
            }            DataRow drAr_E = dtAr_E.NewRow();
            drAr_E["EAR_ID"] = Guid.NewGuid().ToString();
            drAr_E["SEQ_DIR"] = DBNull.Value;
            drAr_E["EAR_NAME"] = txtE_Name.Text;//电子档案名称
            drAr_E["EAR_DESC"] = txtDesc.Text;//文件备注
            drAr_E["EAR_CREATOR"] = txtAuthor.Text;//创建人
            drAr_E["EAR_TP"] = txtFileType.Text;//文件类型,ru：txt,doc,xls
            drAr_E["EAR_SIZE"] = txtSize.Text;//文件大小
            drAr_E["EAR_DATE"] = ModifyDate.Value;//最近修改时间
            drAr_E["DPLY_NO"] = DBNull.Value;//序号
            drAr_E["IS_DELETED"] = DBNull.Value;//删除标志
            drAr_E["TS"] = DateTime.Now;//时间戳            drAr_E["EAR"] = bFile;//电子档案
            dtAr_E.Rows.Add(drAr_E);            SubmitData();
        }        /// <summary>
        /// 提交本地数据
        /// </summary>
        /// <returns></returns>
        private bool SubmitData()
        {
            try
            {                string strTableName = dsData.Tables[0].TableName.ToString();
                OracleConnection conn = new OracleConnection(connectString);
                OracleDataAdapter oda = new OracleDataAdapter(sqlAr_e, conn);                OracleCommandBuilder cmdOracle = new OracleCommandBuilder(oda);                conn.Open();                oda.InsertCommand = cmdOracle.GetInsertCommand();
                oda.UpdateCommand = cmdOracle.GetUpdateCommand();
                int iSucceedRowCounts = oda.Update(dsData, strTableName);
                conn.Close();                dsData.AcceptChanges();                MessageBox.Show("上传成功。", "成功提示", MessageBoxButtons.OK, MessageBoxIcon.Information);
                return true;
            }
            catch (Exception ex)
            {
                if (DialogResult.Retry == MessageBox.Show("提交数据出错。原因如下：\n\n" + ex.Message, "出错提示", MessageBoxButtons.RetryCancel, MessageBoxIcon.Error))
                {
                   return SubmitData();
                }
                else
                {
                    dsData.RejectChanges();
                    return true;
                }
            }
            
        }        /// <summary>
        /// 将指定文件转换成byte[]，主要用于读写blob数据
        /// </summary>
        /// <param name="filePath">如果指定文件不存在，返回null</param>
        /// <returns></returns>
        public  byte[] GetStreamBytes(string filePath)
        {
            if (!File.Exists(filePath))
            {
                MessageBox.Show("文件不存在", "提示", MessageBoxButtons.OK, MessageBoxIcon.Information);
                return null;
            }
            try
            {
                FileStream fileStream = new FileStream(filePath, FileMode.Open);
                byte[] streamBytes = new byte[fileStream.Length];
                fileStream.Read(streamBytes, 0, streamBytes.Length);
                fileStream.Close();                return streamBytes;
            }
            catch (Exception ex)
            {                if (DialogResult.Retry == MessageBox.Show("转换指定文件出错。原因如下：\n\n" + ex.Message+"\n\n是否重试？", "出错提示", MessageBoxButtons.RetryCancel, MessageBoxIcon.Error))
                {
                    return GetStreamBytes(filePath);
                }
                else
                {
                    return null;
                }
            }
        }

Answer 1

/// <summary>
        /// 全文检索
        /// </summary>
        /// <param name="sender"></param>
        /// <param name="e"></param>
        private void btnSearch_Click(object sender, EventArgs e)
        {
            string strFilter = GetFilter();            GetData(strFilter);
        }        /// <summary>
        /// 获取查询用的where子串
        /// </summary>
        /// <returns></returns>
        private string GetFilter()
        {
            ArrayList alFilters = SplitText(txtContent.Text,rbSpecial.Checked);
             int iCount=alFilters.Count;             if (iCount == 0)
             {
                 return null;
             }             string strSubFilter = "";
             for (int i = 0; i < iCount; i++)
             {
                 strSubFilter += alFilters[i].ToString();                 if (i != iCount - 1)
                 {
                     strSubFilter += " or ";
                 }
             }             string[] strFields = new string[] { "ear_name", "ear" };             string strFilter = "";
            int iLength=strFields.Length;
            for(int i=0;i<iLength;i++)
            {
                strFilter += "contains(" + strFields[i] + ",'" + strSubFilter + "')>0";                if (i != iLength - 1)
                {
                    strFilter += " or ";
                }
            }
            return strFilter;
        }        /// <summary>
        /// 将输入的字符串去掉其中的特殊字符后拆分成n个子串，生成ArrayList返回
        /// </summary>
        /// <param name="strText"></param>
        /// <returns></returns>
        private ArrayList SplitText(string strText,bool bAll)
        {
            /*
            [^x] 匹配除了x以外的任意字符
            [^aeiou] 匹配除了aeiou这几个字母以外的任意字符
            */            string strReg = bAll ? " " : "[^A-Za-z0-9\u4e00-\u9fa5]";
            Regex r = new Regex(strReg); // Split on hyphens.
            string[] strFilters = r.Split(strText);            ArrayList alFilters = new ArrayList();
            
            foreach(string strFilter in strFilters)
            {
                if (strFilter != "" && !alFilters.Contains( strFilter))
                {
                    alFilters.Add(strFilter);
                }
            }
            return alFilters;
        }        /// <summary>
        /// 获取数据
        /// </summary>
        /// <param name="strFilter"></param>
        /// <returns></returns>
        private bool GetData(string strFilter)
        {
            try
            {
                strFilter = strFilter.Trim();                if (strFilter == "" || strFilter == null)
                {
                    MessageBox.Show("请设置查询条件后再点击查询按钮！", "出错提示...", MessageBoxButtons.OK, MessageBoxIcon.Error);                    return false;
                }                string sqlCmdText = sqlAr_e+" where "+strFilter;                OracleConnection conn = new OracleConnection(connectString);
                OracleCommand cmd = new OracleCommand(sqlCmdText, conn);
                OracleDataAdapter oda = new OracleDataAdapter(cmd);
                conn.Open();                oda.Fill(dsData, "ar_e");                conn.Close();                return true;
            }
            catch (Exception ex)
            {
                if (DialogResult.Retry == MessageBox.Show("获取数据出错。原因如下：\n\n" + ex.Message + "\n\n是否重试？", "出错提示", MessageBoxButtons.RetryCancel, MessageBoxIcon.Error))
                {
                    return GetData(strFilter);
                }
                else
                {
                    if (dsData != null)
                    {
                        dsData.RejectChanges();
                    }
                    return false;
                }
            }
        }

Answer 2

楼主,我现在也遇到同样的问题了。不过目前的最新问题是.文件DOC格式的存不到CLOB中间去.
提示为:对象必须实现 IConvertible
我是直接循环插入数据的所以不能提交到DATASET里.楼主如果有类似经验的话.
请帮忙,我代码如下: insertsql="insert into CORATTACH_INDEX(ATTACHID,TABLEID,RECORDID,TRUENAME,SEARCHTYPE,ATT_CONTENT)values('"+ATTACHID+"','"+TABLEID+"','"+RECORDID+"','"+TRUENAME+"',1,EMPTY_CLOB())";           
try
{
objoa.RunSqlString(insertsql);
}
catch(Exception error)
{
this.Response.Write(error.Message+"11111111111111111");
} FileStream   fsInfo = new  FileStream( Server.MapPath(ATTACHURL).ToString(), FileMode.Open, FileAccess.Read );
StreamReader  srInfo  =  new StreamReader( fsInfo, System.Text.Encoding.GetEncoding( "GB2312" ) );
srInfo.BaseStream.Seek( 0, SeekOrigin.Begin );
byte[] streambytes=new byte[fsInfo.Length];
fsInfo.Read(streambytes,0,Convert.ToInt32(fsInfo.Length));

try
{
string updatesql="select ATT_CONTENT into myclob from corattach_index where attachid='"+ATTACHID+"' FOR UPDATE";
//string updatesql="update corattach_index set ATT_CONTENT=@txtcontent where attachid='"+ATTACHID+"'";
Connection.Open();
OracleCommand cmd=new OracleCommand(updatesql,Connection);
OracleParameter op=new OracleParameter("@txtcontent",OracleType.Clob);
op.Value=streambytes;
cmd.Parameters.Add(op);
cmd.ExecuteNonQuery();
this.Response.Write("插入TXT文档成功");
Connection.Close();

}
catch(Exception error)
{

this.Response.Write(error.Message+"txt插入出错了 ");
}