【小記】：大膽嘗試才能突破，某個中醫藥大學有一批圖片需要處理（ORC），然後進行資料探勘。之前沒有接觸過ORC這個東西，但是還是應允了。在網上搜索一番，關於中文圖片識別，最終敲定為基於微軟的OneNote，其識別率相對較高。網上這個技術點的資料真心不多，後來於部落格園找到一篇博文，但是那個程式還是bug百出，而且只是單處理。後來經過一番摸索逐個突破，批處理完成。然後進行介面設計，這些零碎工作完成後，便是入庫處理。由於OneNote生成的xml檔案封裝好的，即不可視的。便將其程式碼處理生成txt檔案，再進行Oracle入庫處理。入庫前需要檔案內容稽核，並且在WPF開發中資料繫結和分頁中做了獨特處理。現在經過半個月的工作，本專案做個階段總結。一則知識總結便於二次開發，儘量保持程式流暢性，核心知識做以梳理；另外，相關

篇三：批量處理後的txt檔案入庫處理

【開篇概述】：本文繼上述文章再做深度剖析，本篇主要介紹txt入庫（oracle）和wpf分頁，在此篇涉及的頁面設計前篇已經細述不再概說。裡面涉及到oracle 64位系統安裝處理問題，以前oracle的表，序列，觸發器，自增種子等基本操作。外加檔案操作和利用使用者控制元件分頁等技術。大致效果圖如下，其次貼上完整程式碼，其後就一一概說本文核心要點和難點突破，最後梳理技術知識點，使其一則保持完整性，二則梳理整理知識點便於隨時取用。（本篇暫做效果圖如下，至於最終效果圖和稽核介面，待終篇。）

效果圖：

完整程式碼：

namespace OnenoteOCRDemo
{
    /// <summary>
    /// TextData.xaml 的互動邏輯
    /// </summary>
    public partial class TextData : Window
    {
        public TextData()
        {
            InitializeComponent();
            this.Loaded += new RoutedEventHandler(MainWindow_Loaded);
        }
        void MainWindow_Loaded(object sender, RoutedEventArgs e)
        {
            this.Loaded += delegate
            {
                InitData();
            };
            dataGrid1.LoadingRow += new EventHandler<DataGridRowEventArgs>(dataGrid1_LoadingRow);
        }
        void dataGrid1_LoadingRow(object sender, DataGridRowEventArgs e)
        {
            e.Row.Header = e.Row.GetIndex() + 1;    //設定行表頭的內容值   
        }
        //查詢
        private void InitData()
        {
            string sql = "select * from T_TCM_THESIS  order by ID desc";
            DataSet ds = MYHelper.SQLHelper.GetOrlData(sql);
            if (ds.Tables[0].Rows.Count > 0)
            {
                page1.ShowPages(this.dataGrid1, ds, 10);
            }
            else
            {
                System.Windows.Forms.MessageBox.Show("Erroy");
            }
        }
      
        private void Window_Loaded(object sender, RoutedEventArgs e)
        {
            InitData();
        }
       //批量刪除
        List<int> selectFID = new List<int>();  

        private void CheckBox_Click(object sender, RoutedEventArgs e)
        {
            System.Windows.Controls.CheckBox dg = sender as System.Windows.Controls.CheckBox;
            int FID = int.Parse(dg.Tag.ToString());   //獲取該行的FID   
            var bl = dg.IsChecked;
            if (bl == true)
            {
                selectFID.Add(FID);         //如果選中就儲存FID   
            }
            else
            {
                selectFID.Remove(FID);  //如果選中取消就刪除裡面的FID   
            }  
           
        }

        private void btnDelete_Click(object sender, RoutedEventArgs e)
        {
            foreach (int FID in selectFID)
            {
                string sql = "delete  T_TCM_THESIS where id=" + FID;
                int i = SQLHelper.OrlExecuteQuery(sql);
                if (i > 0)
                {
                    System.Windows.Forms.MessageBox.Show("success!!!");
                }
                else
                {
                    System.Windows.Forms.MessageBox.Show("Erroy!!!");
                }
                InitData();
            }  
        }
        public void TextIsNull()
        {
            txtdocname.Text = "";
            txtyears.Text = "";
            txtdoctype.Text = "";
            txtsummary.Text = "";
            txtauthor.Text = "";
            txtsource.Text = "";
            txtsourcetype.Text = "";
            txtmemo.Text = "";
            txttitle.Text = "";
            txtenglishtitle.Text = "";
            txtcompany.Text = "";
            txtkeyword.Text = "";
            txtcontent.Text = "";
            txtdigest.Text = "";
            txtpath.Text = "";
            txtdisease.Text = "";
        }
        private void btnAdd_Click(object sender, RoutedEventArgs e)
        {
            string docname = txtdocname.Text.ToString();
            string years = txtyears.Text.ToString();
            string doctype = txtdoctype.Text.ToString();
            string summary = txtsummary.Text.ToString();
            string author = txtauthor.Text.ToString();
            string source = txtsource.Text.ToString();
            string sourcetype = txtsourcetype.Text.ToString();
            string memo = txtmemo.Text.ToString();
            string title = txttitle.Text.ToString();
            string englishtitle = txtenglishtitle.Text.ToString();
            string authorcompany = txtcompany.Text.ToString();
            string keyword = txtkeyword.Text.ToString();
            string content = txtcontent.Text.ToString().Trim();
            string digest = txtdigest.Text.ToString();
            string path = txtpath.Text.ToString();
            string disease = txtdisease.Text.ToString();
            int auditflag = 1;
            try
            {
                string sql = "INSERT INTO T_TCM_THESIS(docname,years,doctype,summary,author,source,sourcetype,memo,title,englishtitle,";
                sql += "authorcompany,keyword,content,digest,path,disease,auditflag) Values('" + docname + "','" + years + "','" + doctype + "','" + summary + "',";
                sql += "'" + author + "','" + source + "','" + sourcetype + "','" + memo + "','" + title + "','" + englishtitle + "','" + authorcompany + "','" + keyword + "',";
                sql += "