meta_value看起来像一个序列化的数组。尝试反序列化
var_dump( unserialize( $row->meta_value ) );
编辑
好的，所以你想找到两个日期之间的行。这里的问题是：我们没有 MySQL 可以处理的标准化数据。
这会很慢（随着行数的增加而变慢），因为 MySQL 必须查看所有行
现在，第一步让我们尝试提取日期。不幸的是，MySQL 不提供任何正则表达式替换，因此我们必须使用字符串函数：
第 1 步：尝试提取日期
MySQL 有一组有用的字符串函数，如LOCATE、SUBSTRING和CHAR_LENGTH ，我们可以在这里使用
SET @meta_value := 'a:2:{s:10:"start_date";s:10:"05/16/2013";s:8:"end_date";s:10:"05/18/2013";}';
SET @pattern := 'start_date";s:10:"';
SELECT SUBSTRING( @meta_value, LOCATE( @pattern, @meta_value ) + CHAR_LENGTH( @pattern ), 10 ) AS `start_date`;
+------------+
| start_date |
+------------+
| 05/16/2013 |
+------------+
1 row in set (0.00 sec)
SET @pattern := 'end_date";s:10:"';
SELECT SUBSTRING( @meta_value, LOCATE( @pattern, @meta_value ) + CHAR_LENGTH( @pattern ), 10 ) AS `end_date`;
+------------+
| end_date |
+------------+
| 05/18/2013 |
+------------+
1 row in set (0.00 sec)
哇哦，看起来已经很棒了。
Step2：日期比较
现在我们要比较日期，只需尝试字符串
mysql> SELECT '05/16/2013' > CURRENT_DATE;
ERROR 1267 (HY000): Illegal mix of collations (utf8_general_ci,COERCIBLE) and (latin1_swedish_ci,NUMERIC) for operation '>'
该死，失败 - 为什么？因为我们提取了一个字符串，MySQL 比较整数，让我们做一个日期，MySQL 有一个花哨的函数STR_TO_DATE
mysql> SELECT STR_TO_DATE( '05/16/2013', '%m/%d/%Y');
+----------------------------------------+
| STR_TO_DATE( '05/16/2013', '%m/%d/%Y') |
+----------------------------------------+
| 2013-05-16 |
+----------------------------------------+
1 row in set (0.00 sec)
不错，比较？
mysql> SELECT STR_TO_DATE( '05/16/2013', '%m/%d/%Y') > CURRENT_DATE;
+-------------------------------------------------------+
| STR_TO_DATE( '05/16/2013', '%m/%d/%Y') > CURRENT_DATE |
+-------------------------------------------------------+
| 0 |
+-------------------------------------------------------+
1 row in set (0.00 sec)
mysql> SELECT STR_TO_DATE( '05/16/2013', '%m/%d/%Y') < CURRENT_DATE;
+-------------------------------------------------------+
| STR_TO_DATE( '05/16/2013', '%m/%d/%Y') < CURRENT_DATE |
+-------------------------------------------------------+
| 1 |
+-------------------------------------------------------+
1 row in set (0.00 sec)
第 3 步：包装成函数
完美..现在让我们将所有东西包装成两个函数
DELIMITER $$
CREATE FUNCTION `getStartDate`( metaValue TEXT ) RETURNS DATE
READS SQL DATA
DETERMINISTIC
SQL SECURITY INVOKER
BEGIN
DECLARE startDate DATE;
DECLARE pattern TEXT DEFAULT 'start_date";s:10:"';
SELECT STR_TO_DATE( SUBSTRING( metaValue, LOCATE( pattern, metaValue ) + CHAR_LENGTH( pattern ), 10 ), '%m/%d/%Y' ) INTO startDate;
RETURN startDate;
END$$
CREATE FUNCTION `getEndDate`( metaValue TEXT ) RETURNS DATE
READS SQL DATA
DETERMINISTIC
SQL SECURITY INVOKER
BEGIN
DECLARE endDate DATE;
DECLARE pattern TEXT DEFAULT 'end_date";s:10:"';
SELECT STR_TO_DATE( SUBSTRING( metaValue, LOCATE( pattern, metaValue ) + CHAR_LENGTH( pattern ), 10 ), '%m/%d/%Y' ) INTO endDate;
RETURN endDate;
END$$
DELIMITER ;
测试：
mysql> SELECT getStartDate( @meta_value );
+-----------------------------+
| getStartDate( @meta_value ) |
+-----------------------------+
| 2013-05-16 |
+-----------------------------+
1 row in set (0.00 sec)
mysql> SELECT getEndDate( @meta_value );
+---------------------------+
| getEndDate( @meta_value ) |
+---------------------------+
| 2013-05-18 |
+---------------------------+
1 row in set (0.00 sec)
mysql> SELECT getEndDate( @meta_value ) > '2013-05-12';
+------------------------------------------+
| getEndDate( @meta_value ) > '2013-05-12' |
+------------------------------------------+
| 1 |
+------------------------------------------+
1 row in set (0.00 sec)
mysql> SELECT getStartDate( @meta_value ) > '2013-05-12';
+--------------------------------------------+
| getStartDate( @meta_value ) > '2013-05-12' |
+--------------------------------------------+
| 1 |
+--------------------------------------------+
1 row in set (0.00 sec)
mysql> SELECT * FROM wpTest WHERE getStartDate( meta_value ) >= '2013-05-16' AND getEndDate( meta_value ) <= '2013-05-18';
+-----------------------------------------------------------------------------+
| meta_value |
+-----------------------------------------------------------------------------+
| a:2:{s:10:"start_date";s:10:"05/16/2013";s:8:"end_date";s:10:"05/18/2013";} |
+-----------------------------------------------------------------------------+
1 row in set (0.00 sec)
像魅力一样工作，但是！它会很慢！
第 4 步：规范化数据（建议）
如果您不想在应用程序中进行太多更改，因此开始日期和结束日期实际上存储在单独的日期字段中，您可以使用我们刚刚构建的两个函数和一个很酷的 MySQL 功能，称为触发器，在背景，您可以将它们用于您的选择查询。
让我们首先将这两个字段添加到您的表中（确保替换表名）
ALTER TABLE `wpTest`
ADD `end_date` DATE NOT NULL DEFAULT '0000-00-00' AFTER `meta_value`,
ADD `start_date` DATE NOT NULL DEFAULT '0000-00-00' AFTER `meta_value`;
现在我们得到了将来要用于查询的字段，让我们用我们漂亮的函数来填充它们：
mysql> UPDATE wpTest SET start_date = getStartDate( meta_value ), end_date = getEndDate( meta_value );
Query OK, 1 row affected (0.00 sec)
Rows matched: 1 Changed: 1 Warnings: 0
现在让我们关心这些字段将来会自动维护：我们需要在插入和更新之前执行此操作，因此有两个触发器：
DELIMITER $$
CREATE TRIGGER `wpTest_trg_BI` BEFORE INSERT ON `wpTest` FOR EACH ROW
BEGIN
SET NEW.`start_date` = `getStartDate`( NEW.`meta_value` );
SET NEW.`end_date` = `getEndDate`( NEW.`meta_value` );
END$$
CREATE TRIGGER `wpTest_trg_BU` BEFORE UPDATE ON `wpTest` FOR EACH ROW
BEGIN
SET NEW.`start_date` = `getStartDate`( NEW.`meta_value` );
SET NEW.`end_date` = `getEndDate`( NEW.`meta_value` );
END$$
DELIMITER ;
测试触发器：
mysql> INSERT INTO wpTest (meta_value) SELECT 'a:2:{s:10:"start_date";s:10:"05/12/2013";s:8:"end_date";s:10:"05/20/2013";}';
Query OK, 1 row affected (0.00 sec)
Records: 1 Duplicates: 0 Warnings: 0
mysql> INSERT INTO wpTest (meta_value) SELECT 'a:2:{s:10:"start_date";s:10:"05/10/2013";s:8:"end_date";s:10:"05/23/2013";}';
Query OK, 1 row affected (0.00 sec)
Records: 1 Duplicates: 0 Warnings: 0
mysql> SELECT * FROM wpTest;
+-----------------------------------------------------------------------------+------------+------------+
| meta_value | start_date | end_date |
+-----------------------------------------------------------------------------+------------+------------+
| a:2:{s:10:"start_date";s:10:"05/16/2013";s:8:"end_date";s:10:"05/18/2013";} | 2013-05-16 | 2013-05-18 |
| a:2:{s:10:"start_date";s:10:"05/12/2013";s:8:"end_date";s:10:"05/20/2013";} | 2013-05-12 | 2013-05-20 |
| a:2:{s:10:"start_date";s:10:"05/10/2013";s:8:"end_date";s:10:"05/23/2013";} | 2013-05-10 | 2013-05-23 |
+-----------------------------------------------------------------------------+------------+------------+
3 rows in set (0.00 sec)
完美，插入工作，现在测试更新：
mysql> UPDATE wpTest SET meta_value = 'a:2:{s:10:"start_date";s:10:"05/16/2014";s:8:"end_date";s:10:"05/18/2014";}';
Query OK, 3 rows affected (0.00 sec)
Rows matched: 3 Changed: 3 Warnings: 0
mysql> SELECTS * FROM wpTest;
ERROR 1064 (42000): You have an error in your SQL syntax; check the manual that corresponds to your MySQL server version for the right syntax to use near 'SELECTS * FROM wpTest' at line 1
mysql> SELECT * FROM wpTest;
+-----------------------------------------------------------------------------+------------+------------+
| meta_value | start_date | end_date |
+-----------------------------------------------------------------------------+------------+------------+
| a:2:{s:10:"start_date";s:10:"05/16/2014";s:8:"end_date";s:10:"05/18/2014";} | 2014-05-16 | 2014-05-18 |
| a:2:{s:10:"start_date";s:10:"05/16/2014";s:8:"end_date";s:10:"05/18/2014";} | 2014-05-16 | 2014-05-18 |
| a:2:{s:10:"start_date";s:10:"05/16/2014";s:8:"end_date";s:10:"05/18/2014";} | 2014-05-16 | 2014-05-18 |
+-----------------------------------------------------------------------------+------------+------------+
3 rows in set (0.00 sec)
完美，更新也能正常工作。现在我们不需要在 SELECT 查询中使用需要快速的函数：
$bookings = $wpdb->get_results( "
SELECT ID FROM $wpdb->posts WHERE
post_type = 'tvr_booking' AND
post_status IN ('publish', 'draft')
AND ID not IN (
SELECT post_id
FROM $wpdb->postmeta
WHERE
meta_key = 'stay_interval' AND
start_date >= '$start_date' AND
end_date <= '$end_date'
)
" );
希望这会有所帮助:) 玩得开心