sql server 導出的datetime結果 CAST(0x00009E0E0095524F AS DateTime) 如何向mysql,oracle等數據庫進行轉換

時間 2020-06-22

標籤 sql server 導出 datetime 結果 cast 0x00009e0e0095524f 如何 mysql oracle 數據庫進行轉換欄目 SQL 简体版

原文原文鏈接

1. 處理 sql server 導出的 datetime 類型的字段java

在進行sql server向mysql等其餘數據進行遷移數據時，會發現使用sql server導出的datetime類型的結果是16進製表示的二進制的結果，相似於：CAST(0x00009E0E0095524F AS DateTime)，這樣形式的datetime是沒法向其餘數據庫插入的，因此須要將這種表現形式進行轉換。搜索了好久，纔在在stackoverflow上找到正確的轉換方法。在網上看到不少人都這個問題都不知道解決辦法，本文采用Java語言根據stackoverflow介紹的原理，進行編碼實現轉換。mysql

注意，由於datetime分爲了4字節形式(SmallDateTime)和8字節形式(DateTime)，全部對應的又兩種轉換方法。另外本轉義方法，只能精確到秒級別，毫秒級別是不精確的。若是須要徹底精確的轉換，請採用其餘工具進行轉換，好比sqlyog, mysql workbench等工具。正則表達式

下面是具體的代碼：sql

import java.io.BufferedReader;
import java.io.BufferedWriter;
import java.nio.charset.StandardCharsets;
import java.nio.file.Files;
import java.nio.file.Paths;
import java.text.SimpleDateFormat;
import java.util.Calendar;
import java.util.regex.Matcher;
import java.util.regex.Pattern;

/**
 * sql server數據庫在向mysql等數據庫遷移數據時，會遇到一個麻煩，sql server導出的datetime的結果是16進制形式的二進制結果，
 * 相似於 CAST(0x00009E0E0095524F AS DateTime)，這樣的導出結果是沒法直接向mysql數據庫中導入的，因此須要對sql server
 * 導出的腳本中的全部的 datetime 字段類型進行轉換，轉換成mysql等數據庫承認的結果：2010-10-13 09:03:39
 * 才能正確的完成sql server數據向mysql等數據庫的遷移。
 * 注意本方法只能精確到秒級別，毫秒級別是不精確的。
 * 具體轉換原理，參見：http://stackoverflow.com/questions/12033598/cast-hex-as-datatime-how-to-get-date
 * @author digdeep@126.com
 */
public class SqlServerDateTimeUtils {
    private static final Calendar cal = Calendar.getInstance();
    
    static{
        cal.set(Calendar.YEAR, 1900);
        cal.set(Calendar.MONTH, Calendar.JANUARY);
        cal.set(Calendar.DATE, 1);
        cal.set(Calendar.HOUR_OF_DAY, 0);
        cal.set(Calendar.MINUTE, 0);
        cal.set(Calendar.SECOND, 0);    // 1900-01-01 00:00:00
    }

    /**
     * 將sql server導出的16進制datetime字段： CAST(0x00009E0E0095524F AS DateTime) 形式轉換成字符串形式：
     * 2010-10-13 09:03:39 ，以利於 sql server 向 mysql, oracle 遷移數據 
     * @param rawDateTime 
     * @return 
     * reference: http://stackoverflow.com/questions/12033598/cast-hex-as-datatime-how-to-get-date
     */
    public static String convertRawDateTimeToStr(String rawDateTime){
        return convertRawToStr(rawDateTime, false);
    }
    

    /**
     * 將sql server導出的datetime字段結果 CAST(0x00009E0E0095524F AS DateTime)，轉換成
     * 2010-10-13 09:03:39 或者 2010-10-13 09:03:39.394 注意毫秒部分不精確
     * @param rawDateTime
     * @param millisecondFlag 結果是否須要帶毫秒部分 
     * @return
     * reference: http://stackoverflow.com/questions/12033598/cast-hex-as-datatime-how-to-get-date
     */
    public static String convertRawToStr(String rawDateTime, boolean millisecondFlag) {
        if (rawDateTime == null || rawDateTime.trim().equals(""))
            return null;

        String rawData = rawDateTime.substring("CAST(".length(), rawDateTime.toUpperCase().indexOf(" AS DATETIME"));
        if (rawData == null || rawData.trim().equals(""))    
            return null;
        
        rawData = rawData.trim();
        String result = null;
        
        if(rawData.length() <= 10){        // rowData = "0x993a02CE"
            result = getDateTimeStr4Bytes(rawData);
        }
        if(rawData.length() > 10 && rawData.length() <= 18){    // rowData = "0x00009E0E0095524F"
            result = getDateTimeStr8Bytes(rawData, millisecondFlag);
        }
        return result;
    }
    
    /**
     * sql server 利用 SmallDateTime 4個字節  
     * select CAST(0x993902CE as  SmallDateTime); 
     * 2007-05-25 11:58:00 (只精確到分鐘？？？)
     *
     * mysql:
     * SELECT "0x993902CE" INTO @raw_data; 
     * SELECT conv(substr(@raw_data, 3, 4), 16, 10) INTO @days; 
     * SELECT conv(substr(@raw_data, 7, 4), 16, 10) INTO @minutes;
     * SELECT "1900-01-01 00:00:00" INTO @start_date; 
     * SELECT date_add(@start_date, interval @days DAY) INTO @date_plus_years; 
     * SELECT date_add(@date_plus_years, interval @minutes MINUTE) INTO @final_date;
     * select @final_date;
      * 2007-05-25 11:58:00
     * @param rawData
     * @return
     */
    private static String getDateTimeStr4Bytes(String rawData){
        String day = rawData.substring(2, 2 + 4);    // rowData = "0x993a02CE"
        String minutes = rawData.substring(6, 6 + 4);

        Calendar calendar =  Calendar.getInstance();
        calendar.setTimeInMillis(cal.getTimeInMillis());    // 1900-01-01 00:00:00

        calendar.add(Calendar.DATE, Integer.parseInt(day, 16));
        calendar.add(Calendar.MINUTE, Integer.parseInt(minutes, 16));
        
        SimpleDateFormat sdf = new SimpleDateFormat("yyyy-MM-dd HH:mm:ss");
        
        return sdf.format(calendar.getTime());
    }
    
    /**
     * sql server DateTime 利用 8 字節表示， 該轉換精確到秒，無毫秒部分
     * @param rawData
     * @param millisecondFlag 結果是否須要帶毫秒部分 
     * @return 
     * 
     */
    private static String getDateTimeStr8Bytes(String rawData, boolean millisecondFlag){
        String day = rawData.substring(2, 2 + 8);            // 4字節表示距離1900年的天數
        String fraction = rawData.substring(10, 10 + 8);    // 4字節表示剩餘的部分 faction

        int millis =  (int)(Integer.parseInt(fraction, 16) * 3.33333);    // faction*3.3333 以後表示的是毫秒數
        int seconds= millis / 1000;    // 獲得秒數
        
        Calendar calendar =  Calendar.getInstance();
        calendar.setTimeInMillis(cal.getTimeInMillis());    // 1900-01-01 00:00:00

        calendar.add(Calendar.DATE, Integer.parseInt(day, 16));    // 加上天數
        calendar.add(Calendar.SECOND,  seconds);    // 加上秒數
//      calendar.add(Calendar.MILLISECOND,  millis); // 採用 Calendar.MILLISECOND 計算會致使在秒級別的出現偏差
        
        SimpleDateFormat sdf = null;
        if(millisecondFlag)
            sdf = new SimpleDateFormat("yyyy-MM-dd HH:mm:ss.SSS");    // 毫秒部分是不精確的，毫秒部分每次運行的結果不相同！    
        else
            sdf = new SimpleDateFormat("yyyy-MM-dd HH:mm:ss");    // 不帶毫秒部分，秒級別是精確的，每次運行的結果相同    
        
        return sdf.format(calendar.getTime());
    }
    
    public static void main(String[] args) {        
        //sql server: select CAST(0x00009E0E0095524F AS DateTime) == 2010-10-13 09:03:39.783  正確的日期值
        String str = "CAST(0x00009E0E0095524F AS DateTime)";    
        System.out.println(convertRawToStr(str, true));        // 2010-10-13 09:03:39.374  毫秒部分不精確
        System.out.println(convertRawDateTimeToStr(str));    // 2010-10-13 09:03:39  秒級別是精確的
        
        try {
            // 這裏的字符集通常是 StandardCharsets.UTF_16 或者 StandardCharsets.UTF_8，具體看你導出時採用的是那種字符集
            BufferedReader reader = Files.newBufferedReader(Paths.get("F:\\Members.sql"), StandardCharsets.UTF_16);
            BufferedWriter writer = Files.newBufferedWriter(Paths.get("F:\\Members_mysql.sql"), StandardCharsets.UTF_8);
            
            String line = null; 
            Matcher matcher = null; 
            String matcherStr = null;
            String reg = ".*((?i)CAST\\(0x[0-9-a-f-A-F]+ AS DateTime\\)).*"; // ( 爲特殊字符，表示普通的( 須要用 \\( 來轉義表示
            Pattern pattern = Pattern.compile(reg);

            while ((line = reader.readLine()) != null) {
                while (line.matches(reg)) {
                    matcher = pattern.matcher(line);
                    matcherStr = null;
                    if (matcher.find()) {
                        matcherStr = matcher.group(1);    // matcherStr = "CAST(0x00009E0E0095524F AS DateTime)"
                        if (matcherStr != null) {
                            String mysqlStr = convertRawDateTimeToStr(matcherStr);    
                            if(mysqlStr != null)    
                                line = line.replace(matcherStr, " '"+mysqlStr+"' "); // mysqlStr = '2010-10-13 09:03:39'
                        }
                    } else {
                        break;    // break inner while loop
                    }
                }                
                writer.write(line);
                writer.newLine();
            }            
        } catch (Exception e) {
            e.printStackTrace();
        }        
        System.out.println("done.");
    }    
}

正則表達式 String reg = ".*((?i)CAST\\(0x[0-9-a-f-A-F]+ AS DateTime\\)).*"; 其中的 (?i) 表示不區分字符的大小寫，相似於js中的 /xx/i數據庫

在 F:\\Members.sql 中保存的是sql server導出的insert語句形式的數據：oracle

INSERT [Members] ([g_MemberID], [FK_GroupId], [FK_UserId], [FK_C_UserId], [g_Member_State], [g_Member_Jobtitle], [g_Member_Open], [g_Member_Nickname], [g_Member_FirstnamePY], [g_Member_Firstname], [g_Member_LastnamePY], [g_Member_Lastname], [g_Member_sex], [g_Member_Birthday], [g_Member_MaritalStatus], [FK_City], [g_Member_FacePath], [g_Member_Address], [g_Member_Mobile], [g_Member_hometel], [g_Member_Email], [g_Member_QQ], [g_Member_Wangwang], [g_Member_Msn], [g_Member_Note], [g_Member_JoinTime], [FK_User_AttnUserID], [g_push], [g_Member_Order], [gmt_last_modified]) VALUES (1, 1, 4, 4, 40, NULL, 30, N'大手機', N'', N'', N'總', N'總', N'M', N'2013-11-31', NULL, 11, N'20131231/2013123115MF0DD3.jpg', N'', N'13900000000', N'', N'', N'', N'', N'', N'', NULL, NULL, NULL, NULL, CAST(0x0000A2A500FC2E4F AS DateTime))
INSERT [Members] ([g_MemberID], [FK_GroupId], [FK_UserId], [FK_C_UserId], [g_Member_State], [g_Member_Jobtitle], [g_Member_Open], [g_Member_Nickname], [g_Member_FirstnamePY], [g_Member_Firstname], [g_Member_LastnamePY], [g_Member_Lastname], [g_Member_sex], [g_Member_Birthday], [g_Member_MaritalStatus], [FK_City], [g_Member_FacePath], [g_Member_Address], [g_Member_Mobile], [g_Member_hometel], [g_Member_Email], [g_Member_QQ], [g_Member_Wangwang], [g_Member_Msn], [g_Member_Note], [g_Member_JoinTime], [FK_User_AttnUserID], [g_push], [g_Member_Order], [gmt_last_modified]) VALUES (2, 2, 4, 4, 40, N'錦鯉', 30, N'', N'', N'測試', N'總', N'陳', N'M', N'2013-11-31', 0, 11, N'20140324/2014032409Xva2Ix.jpg', N'', N'13900000000', N'', N'', N'', N'', N'', N'', NULL, NULL, NULL, NULL, CAST(0x0000A2F80095DF52 AS DateTime))

轉換以後的結果保存在 F:\\Members_mysql.sql 中，相似於：工具

INSERT [Members] ([g_MemberID], [FK_GroupId], [FK_UserId], [FK_C_UserId], [g_Member_State], [g_Member_Jobtitle], [g_Member_Open], [g_Member_Nickname], [g_Member_FirstnamePY], [g_Member_Firstname], [g_Member_LastnamePY], [g_Member_Lastname], [g_Member_sex], [g_Member_Birthday], [g_Member_MaritalStatus], [FK_City], [g_Member_FacePath], [g_Member_Address], [g_Member_Mobile], [g_Member_hometel], [g_Member_Email], [g_Member_QQ], [g_Member_Wangwang], [g_Member_Msn], [g_Member_Note], [g_Member_JoinTime], [FK_User_AttnUserID], [g_push], [g_Member_Order], [gmt_last_modified]) VALUES (1, 1, 4, 4, 40, NULL, 30, N'大手機', N'', N'', N'總', N'總', N'M', N'2013-11-31', NULL, 11, N'20131231/2013123115MF0DD3.jpg', N'', N'13900000000', N'', N'', N'', N'', N'', N'', NULL, NULL, NULL, NULL,  '2013-12-31 15:18:09' )
INSERT [Members] ([g_MemberID], [FK_GroupId], [FK_UserId], [FK_C_UserId], [g_Member_State], [g_Member_Jobtitle], [g_Member_Open], [g_Member_Nickname], [g_Member_FirstnamePY], [g_Member_Firstname], [g_Member_LastnamePY], [g_Member_Lastname], [g_Member_sex], [g_Member_Birthday], [g_Member_MaritalStatus], [FK_City], [g_Member_FacePath], [g_Member_Address], [g_Member_Mobile], [g_Member_hometel], [g_Member_Email], [g_Member_QQ], [g_Member_Wangwang], [g_Member_Msn], [g_Member_Note], [g_Member_JoinTime], [FK_User_AttnUserID], [g_push], [g_Member_Order], [gmt_last_modified]) VALUES (2, 2, 4, 4, 40, N'錦鯉', 30, N'', N'', N'測試', N'總', N'陳', N'M', N'2013-11-31', 0, 11, N'20140324/2014032409Xva2Ix.jpg', N'', N'13900000000', N'', N'', N'', N'', N'', N'', NULL, NULL, NULL, NULL,  '2014-03-24 09:05:40.358' )

能夠看到oop

CAST(0x0000A2A500FC2E4F AS DateTime) 被轉換成了：'2013-12-31 15:18:09' 測試

CAST(0x0000A2F80095DF52 AS DateTime) 被轉換成了：'2014-03-24 09:05:40'編碼

固然要達到直接運行插入mysql數據庫中，還要去掉 [g_MemberID] 中的 [] 中括號，還有每一行的末尾要加上逗號 , 再者 INSERT [Members] ([g_MemberID], [FK_GroupId], ...) 語句只須要出現一次，採用 insert into tab(col_a,col_b,col_c) values (1,2,3), (4,5,6) ... 這樣的插入形式，能夠明顯提升插入速度。

另外，若是sql server導出時，若是選擇導出到excel時，datetime字段的值是不須要進行轉換的。在數據量很小時，能夠選擇這種方式。

2. 處理 sql server 導出的 decimal 類型字段和 insert 語句

sql server在導出 decimal 類型的字段時，導出的結果是相似於：CAST(5.00 AS Decimal(6, 2)) 這樣的形式，咱們須要將其轉換成 5.00，還有上面說到的要去掉[g_MemberID] 中的 [] 中括號, 每一行的末尾要加上逗號，還有要轉換成 insert into tab(col_a,col_b,col_c) values (1,2,3), (4,5,6), ....; 形式的結果。下面給出一個相對完整的處理代碼：

import java.io.BufferedReader;
import java.io.BufferedWriter;
import java.nio.charset.Charset;
import java.nio.charset.StandardCharsets;
import java.nio.file.Files;
import java.nio.file.Path;
import java.nio.file.Paths;
import java.util.regex.Matcher;
import java.util.regex.Pattern;

/**
 * sql server數據庫在向mysql等數據庫遷移數據時，會遇到幾個麻煩
 * 1) datetime 類型字段的導出結果須要轉換；
 * 2) decimal 類型的字段的導出結果須要轉換；
 * 3) insert 語句須要去掉 [和 ], 去掉多餘的 insert, 最後變成  insert into user(id,name) values(1,'aa'), (2,'bb'); 的形式
 * @author digdeep@126.com
 */
public class SqlServerConverter {
    private static final int insertSQLcount = 10;
    private static String reg = ".*((?i)CAST\\(0x[0-9-a-f-A-F]+ AS DateTime\\)).*"; // CAST(0x00009E0E0095524F AS DateTime)
    private static Pattern pattern = Pattern.compile(reg);
    
    private static String sqlInertReg = ".*((?i)INSERT .* \\(.*\\) VALUES )\\(.*\\).*"; //insert member (id,name) values (xxx)
    private static Pattern sqlInertPattern = Pattern.compile(sqlInertReg);
    
    private static String decimalReg = ".*((?i)CAST\\(.* AS Decimal\\(.*\\)\\)).*";;    //CAST(5.00 AS Decimal(6, 2))
    private static Pattern decimalPattern = Pattern.compile(decimalReg); 
    
    /**
     * 將每一條insert語句中的 "CAST(0x00009E0E0095524F AS DateTime)" 轉換成對應的：'2010-10-13 09:03:39'
     * @param line
     * @return
     */
    public static String dealWithDateTime(String line){
        Matcher matcher = null; 
        String matcherStr = null;
        
        while (line.matches(reg)) {
            matcher = pattern.matcher(line);
            matcherStr = null;
            if (matcher.find()) {
                matcherStr = matcher.group(1);    // matcherStr = "CAST(0x00009E0E0095524F AS DateTime)"
                if (matcherStr != null) {
                    String mysqlStr = SqlServerDateTimeUtils.convertRawDateTimeToStr(matcherStr);    
                    if(mysqlStr != null)    
                        line = line.replace(matcherStr, " '"+mysqlStr+"' "); // mysqlStr = '2010-10-13 09:03:39'
                }
            } else {
                break;    // break inner while loop
            }
        }
        return line;
    }
    
    /**
     * 將每一條insert語句中的 "CAST(5.00 AS Decimal(6, 2))" 轉換成對應的：5.00
     * @param line
     * @return
     */
    public static String dealWithDecimal(String line){
        Matcher matcher = null; 
        String matcherStr = null;
        
        while (line.matches(decimalReg)) {
            matcher = decimalPattern.matcher(line);
            matcherStr = null;
            if (matcher.find()) {
                matcherStr = matcher.group(1);    // matcherStr = CAST(5.00 AS Decimal(6, 2))
                if (matcherStr != null) {
                    String result = matcherStr.substring("CAST(".length(), 
                                                matcherStr.toUpperCase().indexOf(" AS DECIMAL("));
                    line = line.replace(matcherStr, result);
                }
            } else {
                break;    // break inner while loop
            }
        }
        return line;
    }
    
    /**
     * @param sqlServerPath 輸入文件
     * @param sqlCharset 輸入文件的編碼
     * @param mysqlPath 輸出文件，以utf-8格式輸出
     * @return
     */
    public static boolean converterSqlServerToMySQL(String sqlServerFile, Charset sqlServerCharset, String mysqlFile){
        Path inPath = Paths.get(sqlServerFile);
        Path outPath = Paths.get(mysqlFile);
        
        // writer 採用 StandardCharsets.UTF_8
        try(BufferedReader reader = Files.newBufferedReader(inPath, sqlServerCharset);
             BufferedWriter writer = Files.newBufferedWriter(outPath, StandardCharsets.UTF_8);){
            
            String line = null; 
            Matcher matcher = null; 
            String matcherStr = null;
            
            int i = 0;
            while ((line = reader.readLine()) != null) {
                // 1. 轉換 datetime: CAST(0x00009E0E0095524F AS DateTime) 轉換成 '2010-10-13 09:03:39'
                line = dealWithDateTime(line);
                
                // 2. 轉換decimal: CAST(5.00 AS Decimal(6, 2)) 轉換成 5.00
                line = dealWithDecimal(line);
                
                // 3. 去掉 [ 和 ], 去掉多餘的 insert, 最後變成  insert into user(id,name) values(1,'aa'), (2,'bb'); 的形式
                matcher = sqlInertPattern.matcher(line);
                if(matcher.find()){
                    i++;
                    matcherStr = matcher.group(1);    // matcherStr ==  insert [user]([id], [name]) values (1,'aa')
                    
                    if(i % insertSQLcount == 1){    // 每 insertSQLcount 條 insert 做爲一組進行提交
                        if(i > 1){
                            writer.write(";");    // 加入 ; 進行提交
                            writer.newLine();
                        }
                        line = line.replace(matcherStr, " ");    // line ==  (1,'aa')
                        matcherStr = matcherStr.replace("[", "");
                        matcherStr = matcherStr.replace("]", "");
                        
                        matcherStr = matcherStr.toLowerCase().replace("insert", "insert into ");
                        line = matcherStr + "\n" + line;    // line == insert into user(id,name) \n values(1,'aa')
                    }else{
                        line = line.replace(matcherStr, ",");    // line == ,values(2,'aa')
                    }
                    writer.write(line);
                    writer.newLine();
                }
            }
            
            if(i % insertSQLcount != 1 && i > 0) // 當末尾沒有封號(;)， 時才添加一個封號(;)，避免在末尾出現兩個封號(;;)
                writer.write(";");
            
            writer.newLine();
            writer.flush();        // 注意這裏必定要 flush ，否則文檔的最後面的最後一兩條數據不會完整寫到文件中的！！！
        } catch (Exception e) {
            e.printStackTrace();
            return false;
        }
        
        System.out.println("OK.");
        return true;
    }
    
    public static void main(String[] args) {
        // 這裏的字符集通常是 StandardCharsets.UTF_16 或者 StandardCharsets.UTF_8，具體看你導出時採用的是那種字符集
        // 若是sql server導出時選擇了 unicode，那麼這裏就應該使用 StandardCharsets.UTF_16
        SqlServerConverter.converterSqlServerToMySQL("F:\\Members.sql", StandardCharsets.UTF_16, 
                                                    "F:\\Members_mysql.sql");
        
        System.out.println("-------------------------------------------------");
        
        SqlServerConverter.converterSqlServerToMySQL("F:\\model_choise.sql", StandardCharsets.UTF_16, 
                                                        "F:\\model_choise_mysql.sql");
        System.out.println("done.");
    }
    
}

轉換以前 F:\\model_choise.sql 的文件內容以下所示：

SET IDENTITY_INSERT [model_choise] ON
INSERT [model_choise] ([id], [gid], [member_id], [user_id], [model_id], [status], [choise_time], [item_id], [pay_amount], [accuracy], [g_name], [g_mobile], [reation_id]) VALUES (992, NULL, NULL, 736, 481, 1, CAST(0x0000A49500C0DE5F AS DateTime), NULL, NULL, NULL, NULL, NULL, NULL)
INSERT [model_choise] ([id], [gid], [member_id], [user_id], [model_id], [status], [choise_time], [item_id], [pay_amount], [accuracy], [g_name], [g_mobile], [reation_id]) VALUES (993, NULL, NULL, 737, 559, 1, CAST(0x0000A49501420847 AS DateTime), NULL, NULL, NULL, NULL, NULL, NULL)
INSERT [model_choise] ([id], [gid], [member_id], [user_id], [model_id], [status], [choise_time], [item_id], [pay_amount], [accuracy], [g_name], [g_mobile], [reation_id]) VALUES (1009, NULL, NULL, 9, 528, 1, CAST(0x0000A496013F1119 AS DateTime), NULL, NULL, CAST(5.00 AS Decimal(6, 2)), NULL, NULL, NULL)
INSERT [model_choise] ([id], [gid], [member_id], [user_id], [model_id], [status], [choise_time], [item_id], [pay_amount], [accuracy], [g_name], [g_mobile], [reation_id]) VALUES (1010, NULL, NULL, 4, 528, 1, CAST(0x0000A496013F9F76 AS DateTime), NULL, NULL, CAST(5.00 AS Decimal(6, 2)), NULL, NULL, NULL)
// ... ...

轉換以後 F:\\model_choise_mysql.sql 內容以下所示：

insert into  model_choise (id, gid, member_id, user_id, model_id, status, choise_time, item_id, pay_amount, accuracy, g_name, g_mobile, reation_id) values 
 (992, NULL, NULL, 736, 481, 1,  '2015-05-11 11:42:12' , NULL, NULL, NULL, NULL, NULL, NULL)
,(993, NULL, NULL, 737, 559, 1,  '2015-05-11 19:32:28' , NULL, NULL, NULL, NULL, NULL, NULL)
,(1009, NULL, NULL, 9, 528, 1,  '2015-05-12 19:21:41' , NULL, NULL, 5.00, NULL, NULL, NULL)
,(1010, NULL, NULL, 4, 528, 1,  '2015-05-12 19:23:42' , NULL, NULL, 5.00, NULL, NULL, NULL)
,(1011, NULL, NULL, 4, 495, 1,  '2015-05-12 19:24:38' , NULL, NULL, NULL, NULL, NULL, NULL)
,(1012, NULL, NULL, 4, 554, 1,  '2015-05-12 19:25:23' , NULL, NULL, NULL, NULL, NULL, NULL)
,(1013, NULL, NULL, 566, 528, -1,  '2015-05-12 20:33:24' , NULL, NULL, 5.00, NULL, NULL, NULL)
,(1014, NULL, NULL, 8, 495, -1,  '2015-05-12 21:03:45' , NULL, NULL, 5.00, NULL, NULL, NULL)
,(1015, NULL, NULL, 8, 498, -1,  '2015-05-12 21:04:15' , NULL, NULL, 5.00, NULL, NULL, NULL)
,(1016, NULL, NULL, 8, 558, -1,  '2015-05-12 21:04:39' , NULL, NULL, 5.00, NULL, NULL, NULL)
;
insert into  model_choise (id, gid, member_id, user_id, model_id, status, choise_time, item_id, pay_amount, accuracy, g_name, g_mobile, reation_id) values 
 (1017, NULL, NULL, 8, 417, -1,  '2015-05-12 21:05:29' , NULL, NULL, 5.00, NULL, NULL, NULL)
,(1018, NULL, NULL, 8, 494, -1,  '2015-05-12 21:29:20' , NULL, NULL, 4.00, NULL, NULL, NULL)
,(1019, NULL, NULL, 8, 528, -1,  '2015-05-12 21:31:41' , NULL, NULL, 5.00, NULL, NULL, NULL)
,(1026, NULL, NULL, 566, 493, 1,  '2015-05-13 02:14:09' , NULL, NULL, 5.00, NULL, NULL, NULL)
,(1027, NULL, NULL, 566, 532, 1,  '2015-05-13 02:14:17' , NULL, NULL, 5.00, NULL, NULL, NULL)
,(1028, NULL, NULL, 566, 459, 1,  '2015-05-13 09:16:44' , NULL, NULL, 5.00, NULL, NULL, NULL)
,(1029, NULL, NULL, 566, 421, -1,  '2015-05-13 10:03:52' , NULL, NULL, 5.00, NULL, NULL, NULL)
,(1032, NULL, NULL, 9, 495, 1,  '2015-05-13 13:27:32' , NULL, NULL, 5.00, NULL, NULL, NULL)
,(1033, NULL, NULL, 9, 554, 1,  '2015-05-13 13:28:21' , NULL, NULL, 5.00, NULL, NULL, NULL)
,(1034, NULL, NULL, 9, 560, -1,  '2015-05-13 17:10:21' , NULL, NULL, NULL, NULL, NULL, NULL)
;
// ... ...

能夠看到 CAST(0x0000A496013F1119 AS DateTime), 和 CAST(5.00 AS Decimal(6, 2)) 都被正確的轉換了。insert 語句的轉換也是徹底正確的。因此轉換以後的文件 F:\\model_choise_mysql.sql 是能夠以下使用 source 命令來直接插入mysql數據庫中:

mysql> source F:\\model_choise_mysql.sql
Query OK, 10 rows affected (0.03 sec)
Records: 10  Duplicates: 0  Warnings: 0

Query OK, 10 rows affected (0.05 sec)
Records: 10  Duplicates: 0  Warnings: 0

Query OK, 10 rows affected (0.04 sec)
Records: 10  Duplicates: 0  Warnings: 0

Query OK, 10 rows affected (0.04 sec)
Records: 10  Duplicates: 0  Warnings: 0

Query OK, 10 rows affected (0.05 sec)
Records: 10  Duplicates: 0  Warnings: 0

Query OK, 10 rows affected (0.02 sec)
Records: 10  Duplicates: 0  Warnings: 0

Query OK, 10 rows affected (0.02 sec)
Records: 10  Duplicates: 0  Warnings: 0

Query OK, 10 rows affected (0.02 sec)
Records: 10  Duplicates: 0  Warnings: 0

Query OK, 8 rows affected (0.03 sec)
Records: 8  Duplicates: 0  Warnings: 0

mysql> select count(*) from model_choise;
+----------+
| count(*) |
+----------+
|       88 |
+----------+
1 row in set (0.00 sec)

實際使用時，private static final int insertSQLcount = 10; insertSQLcount 應該調大一點，幾百，上千都是能夠的。

另外上面使用了 Java 1.7 中新引入的 try-with-recourse 的自動關閉資源的使用方法:

try(BufferedReader reader = Files.newBufferedReader(inPath, StandardCharsets.UTF_16);
             BufferedWriter writer = Files.newBufferedWriter(outPath, StandardCharsets.UTF_8);)