實現 SQL 嵌套 Cursor 的最佳方法是什么

Question

請有人告訴我使用提供的數據集實現嵌套 Cursor 的最佳方法嗎？

如您所見，已經有一個 Cursor 用於查找 #TMP_CODE1 中列出的值，如果匹配，則執行 select 語句，並將結果插入 Z78E6221F6393D1356681.DB398 表中

我正在嘗試使用嵌套的 Cursor 對存儲在#TMP_CODE2 中的值復制相同的過程。

我想以與將@Letters 插入下面的#Output 表相同的方式打印@Numbers：

  PRINT @Letters;
  INSERT INTO #Output ([Dummy Column1], [Dummy Column2]) VALUES ('Letters:', @Letters)
  INSERT INTO #Output ([Dummy Column1]) VALUES ('') 

將每一行分組在一起，並在每個結果集上方顯示代碼。

如果值列在 #TMP_CODE1 和 #TMP_CODE2 臨時表中，則這些值應僅插入 Output 表中。

DROP TABLE IF EXISTS #TMP_CODE1
CREATE TABLE #TMP_CODE1(Letters varchar(3));
INSERT INTO #TMP_CODE1(Letters) VALUES 
  ('AAA'),('BBB'),('CCC'),('DDD'),('EEE'),('FFF');

DROP TABLE IF EXISTS #TMP_CODE2
CREATE TABLE #TMP_CODE2(Numbers varchar(3));
INSERT INTO #TMP_CODE2(Numbers) VALUES 
  ('111'),('222'),('333'),('444'),('555'),('666');

DROP TABLE IF EXISTS #Input
CREATE TABLE #Input (CODE1 varchar(3), [Dummy Column1] varchar (15), 
[Dummy Column2] varchar (15), CODE2 varchar(3), [Dummy Column3] varchar (15),
[Dummy Column4] varchar (15));

DROP TABLE IF EXISTS #Output
CREATE TABLE #Output (CODE1 varchar(3), [Dummy Column1] varchar (15), 
[Dummy Column2] varchar (15), CODE2 varchar(3), [Dummy Column3] varchar (15),
[Dummy Column4] varchar (15));

INSERT INTO #Input (CODE1, [Dummy Column1], [Dummy Column2], CODE2, [Dummy Column3], [Dummy Column4]) VALUES ('AAA', 'DUMMY DATA 1 1', 'DUMMY DATA 2 1', '111', 'DUMMY DATA 3 1', 'DUMMY DATA 4 1')
INSERT INTO #Input (CODE1, [Dummy Column1], [Dummy Column2], CODE2, [Dummy Column3], [Dummy Column4]) VALUES ('AAA', 'DUMMY DATA 1 2', 'DUMMY DATA 2 2', '111', 'DUMMY DATA 3 2', 'DUMMY DATA 4 2')
INSERT INTO #Input (CODE1, [Dummy Column1], [Dummy Column2], CODE2, [Dummy Column3], [Dummy Column4]) VALUES ('AAA', 'DUMMY DATA 1 3', 'DUMMY DATA 2 3', '111', 'DUMMY DATA 3 3', 'DUMMY DATA 4 3')
INSERT INTO #Input (CODE1, [Dummy Column1], [Dummy Column2], CODE2, [Dummy Column3], [Dummy Column4]) VALUES ('AAA', 'DUMMY DATA 1 4', 'DUMMY DATA 2 4', '222', 'DUMMY DATA 3 4', 'DUMMY DATA 4 4')
INSERT INTO #Input (CODE1, [Dummy Column1], [Dummy Column2], CODE2, [Dummy Column3], [Dummy Column4]) VALUES ('AAA', 'DUMMY DATA 1 5', 'DUMMY DATA 2 5', '333', 'DUMMY DATA 3 5', 'DUMMY DATA 4 5')
INSERT INTO #Input (CODE1, [Dummy Column1], [Dummy Column2], CODE2, [Dummy Column3], [Dummy Column4]) VALUES ('AAA', 'DUMMY DATA 1 6', 'DUMMY DATA 2 6', '555', 'DUMMY DATA 3 6', 'DUMMY DATA 4 6')
INSERT INTO #Input (CODE1, [Dummy Column1], [Dummy Column2], CODE2, [Dummy Column3], [Dummy Column4]) VALUES ('BBB', 'DUMMY DATA 1 7', 'DUMMY DATA 2 7', '222', 'DUMMY DATA 3 7', 'DUMMY DATA 4 7')
INSERT INTO #Input (CODE1, [Dummy Column1], [Dummy Column2], CODE2, [Dummy Column3], [Dummy Column4]) VALUES ('BBB', 'DUMMY DATA 1 8', 'DUMMY DATA 2 8', '111', 'DUMMY DATA 3 8', 'DUMMY DATA 4 8')
INSERT INTO #Input (CODE1, [Dummy Column1], [Dummy Column2], CODE2, [Dummy Column3], [Dummy Column4]) VALUES ('BBB', 'DUMMY DATA 1 9', 'DUMMY DATA 2 9', '555', 'DUMMY DATA 3 9', 'DUMMY DATA 4 9')
INSERT INTO #Input (CODE1, [Dummy Column1], [Dummy Column2], CODE2, [Dummy Column3], [Dummy Column4]) VALUES ('BBB', 'DUMMY DATA 1 10', 'DUMMY DATA 2 10', '444', 'DUMMY DATA 3 10', 'DUMMY DATA 4 10')
INSERT INTO #Input (CODE1, [Dummy Column1], [Dummy Column2], CODE2, [Dummy Column3], [Dummy Column4]) VALUES ('BBB', 'DUMMY DATA 1 11', 'DUMMY DATA 2 11', '777', 'DUMMY DATA 3 11', 'DUMMY DATA 4 11')
INSERT INTO #Input (CODE1, [Dummy Column1], [Dummy Column2], CODE2, [Dummy Column3], [Dummy Column4]) VALUES ('BBB', 'DUMMY DATA 1 12', 'DUMMY DATA 2 12', '888', 'DUMMY DATA 3 12', 'DUMMY DATA 4 12')
INSERT INTO #Input (CODE1, [Dummy Column1], [Dummy Column2], CODE2, [Dummy Column3], [Dummy Column4]) VALUES ('BBB', 'DUMMY DATA 1 13', 'DUMMY DATA 2 13', '222', 'DUMMY DATA 3 13', 'DUMMY DATA 4 13')
INSERT INTO #Input (CODE1, [Dummy Column1], [Dummy Column2], CODE2, [Dummy Column3], [Dummy Column4]) VALUES ('BBB', 'DUMMY DATA 1 14', 'DUMMY DATA 2 14', '111', 'DUMMY DATA 3 14', 'DUMMY DATA 4 14')
INSERT INTO #Input (CODE1, [Dummy Column1], [Dummy Column2], CODE2, [Dummy Column3], [Dummy Column4]) VALUES ('BBB', 'DUMMY DATA 1 15', 'DUMMY DATA 2 15', '222', 'DUMMY DATA 3 15', 'DUMMY DATA 4 15')
INSERT INTO #Input (CODE1, [Dummy Column1], [Dummy Column2], CODE2, [Dummy Column3], [Dummy Column4]) VALUES ('BBB', 'DUMMY DATA 1 16', 'DUMMY DATA 2 16', '666', 'DUMMY DATA 3 16', 'DUMMY DATA 4 16')
INSERT INTO #Input (CODE1, [Dummy Column1], [Dummy Column2], CODE2, [Dummy Column3], [Dummy Column4]) VALUES ('CCC', 'DUMMY DATA 1 17', 'DUMMY DATA 2 17', '111', 'DUMMY DATA 3 17', 'DUMMY DATA 4 17')
INSERT INTO #Input (CODE1, [Dummy Column1], [Dummy Column2], CODE2, [Dummy Column3], [Dummy Column4]) VALUES ('CCC', 'DUMMY DATA 1 18', 'DUMMY DATA 2 18', '111', 'DUMMY DATA 3 18', 'DUMMY DATA 4 18')
INSERT INTO #Input (CODE1, [Dummy Column1], [Dummy Column2], CODE2, [Dummy Column3], [Dummy Column4]) VALUES ('DDD', 'DUMMY DATA 1 19', 'DUMMY DATA 2 19', '111', 'DUMMY DATA 3 19', 'DUMMY DATA 4 19')
INSERT INTO #Input (CODE1, [Dummy Column1], [Dummy Column2], CODE2, [Dummy Column3], [Dummy Column4]) VALUES ('DDD', 'DUMMY DATA 1 20', 'DUMMY DATA 2 20', '222', 'DUMMY DATA 3 20', 'DUMMY DATA 4 20')
INSERT INTO #Input (CODE1, [Dummy Column1], [Dummy Column2], CODE2, [Dummy Column3], [Dummy Column4]) VALUES ('DDD', 'DUMMY DATA 1 21', 'DUMMY DATA 2 21', '333', 'DUMMY DATA 3 21', 'DUMMY DATA 4 21')
INSERT INTO #Input (CODE1, [Dummy Column1], [Dummy Column2], CODE2, [Dummy Column3], [Dummy Column4]) VALUES ('DDD', 'DUMMY DATA 1 22', 'DUMMY DATA 2 22', '555', 'DUMMY DATA 3 22', 'DUMMY DATA 4 22')
INSERT INTO #Input (CODE1, [Dummy Column1], [Dummy Column2], CODE2, [Dummy Column3], [Dummy Column4]) VALUES ('EEE', 'DUMMY DATA 1 23', 'DUMMY DATA 2 23', '222', 'DUMMY DATA 3 23', 'DUMMY DATA 4 23')
INSERT INTO #Input (CODE1, [Dummy Column1], [Dummy Column2], CODE2, [Dummy Column3], [Dummy Column4]) VALUES ('EEE', 'DUMMY DATA 1 24', 'DUMMY DATA 2 24', '111', 'DUMMY DATA 3 24', 'DUMMY DATA 4 24')
INSERT INTO #Input (CODE1, [Dummy Column1], [Dummy Column2], CODE2, [Dummy Column3], [Dummy Column4]) VALUES ('EEE', 'DUMMY DATA 1 25', 'DUMMY DATA 2 25', '555', 'DUMMY DATA 3 25', 'DUMMY DATA 4 25')
INSERT INTO #Input (CODE1, [Dummy Column1], [Dummy Column2], CODE2, [Dummy Column3], [Dummy Column4]) VALUES ('EEE', 'DUMMY DATA 1 26', 'DUMMY DATA 2 26', '444', 'DUMMY DATA 3 26', 'DUMMY DATA 4 26')
INSERT INTO #Input (CODE1, [Dummy Column1], [Dummy Column2], CODE2, [Dummy Column3], [Dummy Column4]) VALUES ('EEE', 'DUMMY DATA 1 27', 'DUMMY DATA 2 27', '777', 'DUMMY DATA 3 27', 'DUMMY DATA 4 27')
INSERT INTO #Input (CODE1, [Dummy Column1], [Dummy Column2], CODE2, [Dummy Column3], [Dummy Column4]) VALUES ('FFF', 'DUMMY DATA 1 28', 'DUMMY DATA 2 28', '888', 'DUMMY DATA 3 28', 'DUMMY DATA 4 28')
INSERT INTO #Input (CODE1, [Dummy Column1], [Dummy Column2], CODE2, [Dummy Column3], [Dummy Column4]) VALUES ('FFF', 'DUMMY DATA 1 29', 'DUMMY DATA 2 29', '222', 'DUMMY DATA 3 29', 'DUMMY DATA 4 29')
INSERT INTO #Input (CODE1, [Dummy Column1], [Dummy Column2], CODE2, [Dummy Column3], [Dummy Column4]) VALUES ('FFF', 'DUMMY DATA 1 30', 'DUMMY DATA 2 30', '111', 'DUMMY DATA 3 30', 'DUMMY DATA 4 30')
INSERT INTO #Input (CODE1, [Dummy Column1], [Dummy Column2], CODE2, [Dummy Column3], [Dummy Column4]) VALUES ('FFF', 'DUMMY DATA 1 31', 'DUMMY DATA 2 31', '222', 'DUMMY DATA 3 31', 'DUMMY DATA 4 31')
INSERT INTO #Input (CODE1, [Dummy Column1], [Dummy Column2], CODE2, [Dummy Column3], [Dummy Column4]) VALUES ('GGG', 'DUMMY DATA 1 32', 'DUMMY DATA 2 32', '666', 'DUMMY DATA 3 32', 'DUMMY DATA 4 32')
INSERT INTO #Input (CODE1, [Dummy Column1], [Dummy Column2], CODE2, [Dummy Column3], [Dummy Column4]) VALUES ('GGG', 'DUMMY DATA 1 33', 'DUMMY DATA 2 33', '222', 'DUMMY DATA 3 33', 'DUMMY DATA 4 33')
INSERT INTO #Input (CODE1, [Dummy Column1], [Dummy Column2], CODE2, [Dummy Column3], [Dummy Column4]) VALUES ('GGG', 'DUMMY DATA 1 34', 'DUMMY DATA 2 34', '111', 'DUMMY DATA 3 34', 'DUMMY DATA 4 34')
INSERT INTO #Input (CODE1, [Dummy Column1], [Dummy Column2], CODE2, [Dummy Column3], [Dummy Column4]) VALUES ('GGG', 'DUMMY DATA 1 35', 'DUMMY DATA 2 35', '555', 'DUMMY DATA 3 35', 'DUMMY DATA 4 35')
INSERT INTO #Input (CODE1, [Dummy Column1], [Dummy Column2], CODE2, [Dummy Column3], [Dummy Column4]) VALUES ('HHH', 'DUMMY DATA 1 36', 'DUMMY DATA 2 36', '888', 'DUMMY DATA 3 36', 'DUMMY DATA 4 36')
INSERT INTO #Input (CODE1, [Dummy Column1], [Dummy Column2], CODE2, [Dummy Column3], [Dummy Column4]) VALUES ('HHH', 'DUMMY DATA 1 37', 'DUMMY DATA 2 37', '555', 'DUMMY DATA 3 37', 'DUMMY DATA 4 37')
  
DECLARE @Letters varchar(3), @c cursor;
    
SET @c = cursor LOCAL FAST_FORWARD
    FOR SELECT Letters FROM #TMP_CODE1;

OPEN @c;
FETCH NEXT FROM @c INTO @Letters;

WHILE @@FETCH_STATUS = 0
BEGIN

  PRINT @Letters;
  INSERT INTO #Output ([Dummy Column1], [Dummy Column2]) VALUES ('Letters:', @Letters)
  INSERT INTO #Output ([Dummy Column1]) VALUES ('')

  INSERT INTO #Output (CODE1, [Dummy Column1], [Dummy Column2], CODE2, [Dummy Column3],[Dummy Column4])
  SELECT * FROM #Input WHERE CODE1 = @Letters

  FETCH NEXT FROM @c INTO @Letters;

END

SELECT * FROM #Output

所以我希望實現的結構是這樣的：

這僅適用於 AAA，我希望能夠以這種格式格式化示例代碼中提供的所有數據。

我正在使用 SQL 服務器，因此請在標准 SQL 中提供答案。

謝謝

Answer 1

我將首先說這確實應該在表示層中完成，而不是在 SQL 中。

盡管如此，您實際上可以通過仔細使用GROUPING SETS和ORDER BY以您想要的順序獲得所有想要的行：

SELECT
  CODE1           = CASE WHEN GROUPING(i.[Dummy Column1]) = 0 THEN CODE1 END,
  [Dummy Column1] = CASE WHEN GROUPING(i.CODE2) = 1 AND GROUPING(i.[Dummy Column1]) = 1
                           THEN 'Letters'
                         WHEN GROUPING(i.[Dummy Column1]) = 1
                           THEN 'Numbers'
                         ELSE i.[Dummy Column1] END,
  [Dummy Column2] = CASE WHEN GROUPING(i.CODE2) = 1 AND GROUPING(i.[Dummy Column1]) = 1
                           THEN i.CODE1
                         WHEN GROUPING(i.[Dummy Column1]) = 1
                           THEN i.CODE2
                         ELSE i.[Dummy Column1] END,
  CODE2           = CASE WHEN GROUPING(i.[Dummy Column1]) = 0 THEN CODE2 END,
  i.[Dummy Column3],
  i.[Dummy Column4]
FROM #Input i
GROUP BY GROUPING SETS (
    (i.CODE1),
    (i.CODE1, i.CODE2),
    (i.CODE1, i.CODE2, i.[Dummy Column1], i.[Dummy Column2], i.[Dummy Column3], i.[Dummy Column4])
)
ORDER BY
  i.CODE1,
  GROUPING(i.CODE2) DESC,
  i.CODE2,
  GROUPING(i.[Dummy Column1]) DESC;

db<>小提琴

GROUPING SETS可以返回小計和總計。 它是ROLLUP更靈活的替代方案。

GROUPING() function 告訴您列是否已在該特定行中分組。 您可以在ORDER BY中使用它來獲取正確 position 中的總行數。 然后您可以在SELECT中使用它來區分這些行。

• 注意你的數據類型。 某些列可能彼此不兼容，因此您可能需要強制轉換。
• 請注意ORDER BY ，如果您未在列上指定表別名，則它將按最終SELECT值排序，而不是實際列值