What is the easiest / most efficient way to get the entire row inserted after an INSERT statement?
在插入语句之后插入整个行,最简单/最有效的方法是什么?
I am pretty sure I could do this as follows:
我很确定我可以这样做:
$aQuery = "INSERT into myTable (a, b, c) VALUES (1, 'Hello', 'Goodbye')";
//the IDENTITY coloumn in myTable is named id
$result = sqlsrv_query($myConn, $aQuery );
if ($result) {
$res = sqlsrv_query('SELECT LAST_INSERT_ID()');
$row = sqlsrv_fetch_array($res);
$lastInsertId = $row[0];
$subQuery = "SELECT * FROM myTable where id = {$lastInsertId}";
$subResult = sqlsrv_query($myConn, $subQuery);
if ($subResult) {
while($subrow = sqlsrv_fetch_array($subResult)) {
echo($subrow ['id'] . ', '.
$subrow ['a'] . ', '.
$subrow ['b']); //etc...
}
}
}
However, I am concerned about the possibility of another insert occurring just before my SELECT LAST_INSERT_ID()
and thus messing up my logic to boot. How can I be certain that the last inserted id is truly the INSERT I called previously, and not something happening somewhere else?
但是,我担心在我的SELECT LAST_INSERT_ID()之前可能会出现另一个插入,从而导致启动逻辑混乱。我怎么能确定最后插入的id是我之前调用的插入,而不是其他地方发生的事情呢?
Is there a more appropriate way of doing this, perhaps a complete SQL solution (such that the query returns the row automatically rather than using PHP)?
是否有一种更合适的方法来实现这一点,也许是一种完整的SQL解决方案(比如查询自动返回行,而不是使用PHP)?
UPDATE: myTable DOES have an explicitly defined (and auto-incremented) identity column, named id
.
更新:myTable确实有一个显式定义(并自动递增)的标识列,名为id。
3 个解决方案
#1
4
This will work:
这将工作:
"INSERT into myTable (a, b, c) OUTPUT Inserted.a, Inserted.b, Inserted.c VALUES (1, 'Hello', 'Goodbye')
#2
1
In Sql Server, you would use select @lastID=SCOPE_IDENTITY()
在Sql Server中,您将使用select @lastID=SCOPE_IDENTITY()
And @LastID
will have the last id inserted for the current scope; therefore, if there was another insertion in the middle, you would still get the correct record on your select.
@LastID将会为当前范围插入最后一个id;因此,如果中间有另一个插入,您仍然可以在select中获得正确的记录。
Never use @@Identity
for this or you may end up in a situation like you described.
千万不要使用@ identity,否则您可能会遇到您所描述的情况。
#3
0
If you were to use identity field (which maybe you should) there is a command called SCOPE_IDENTIY() which info you can find here: http://msdn.microsoft.com/en-us/library/ms190315.aspx
如果您要使用identity字段(可能您应该使用),有一个名为SCOPE_IDENTIY()的命令,您可以在这里找到:http://msdn.microsoft.com/en-us/library/ms190315.aspx。
Since you do not use it, you do not have to select latest data since you have it when you insert, so just use same data instead of selecting.
由于您不使用它,所以您不必选择最新的数据,因为您在插入时已经有了它,所以只需使用相同的数据,而不必选择。
#1
4
This will work:
这将工作:
"INSERT into myTable (a, b, c) OUTPUT Inserted.a, Inserted.b, Inserted.c VALUES (1, 'Hello', 'Goodbye')
#2
1
In Sql Server, you would use select @lastID=SCOPE_IDENTITY()
在Sql Server中,您将使用select @lastID=SCOPE_IDENTITY()
And @LastID
will have the last id inserted for the current scope; therefore, if there was another insertion in the middle, you would still get the correct record on your select.
@LastID将会为当前范围插入最后一个id;因此,如果中间有另一个插入,您仍然可以在select中获得正确的记录。
Never use @@Identity
for this or you may end up in a situation like you described.
千万不要使用@ identity,否则您可能会遇到您所描述的情况。
#3
0
If you were to use identity field (which maybe you should) there is a command called SCOPE_IDENTIY() which info you can find here: http://msdn.microsoft.com/en-us/library/ms190315.aspx
如果您要使用identity字段(可能您应该使用),有一个名为SCOPE_IDENTIY()的命令,您可以在这里找到:http://msdn.microsoft.com/en-us/library/ms190315.aspx。
Since you do not use it, you do not have to select latest data since you have it when you insert, so just use same data instead of selecting.
由于您不使用它,所以您不必选择最新的数据,因为您在插入时已经有了它,所以只需使用相同的数据,而不必选择。