Question

如果不存在则需要插入行，如果存在则需要更新。我找到了适用于MySQL的解决方案：

INSERT INTO table (id, name, age) VALUES(1, "A", 19) ON DUPLICATE KEY UPDATE

name =“ A”，年龄= 19

但是我找不到类似的MSSQL。

Answer 1

您可以按以下顺序使用2条语句（INSERT + UPDATE）。如果不存在，则更新不会更新任何内容，如果存在，则插入内容也不会插入：

UPDATE T SET
    name = 'A',
    age = 19
FROM
    [table] AS T
WHERE
    T.id = 1

INSERT INTO [table] (
    id,
    name,
    age)
SELECT
    id = 1,
    name = 'A',
    age = 19
WHERE
    NOT EXISTS (SELECT 'not yet loaded' FROM [table] AS T WHERE T.id = 1)

或MERGE：

;WITH ValuesToMerge AS
(
    SELECT
        id = 1,
        name = 'A',
        age = 19
)
MERGE INTO 
    [table] AS T
USING
    ValuesToMerge AS V ON (T.id = V.id)
WHEN NOT MATCHED BY TARGET THEN
    INSERT (
        id,
        name,
        age)
    VALUES (
        V.id,
        V.name,
        V.age)
WHEN MATCHED THEN
    UPDATE SET
        name = V.name,
        age = V.name;

Answer 2

我更喜欢检查@@ROWCOUNT。这是一个更为紧凑的解决方案。

UPDATE table set name = 'A', age = 19 WHERE id = 1;
IF @@ROWCOUNT = 0
INSERT INTO table (id, name, age) VALUES(1, "A", 19);

MSSQL插入或更新（如果存在）

2 个答案: