使用存储数据到查询SSIS

Question

我有一个查询，我需要的数据，所以我将其存储到OLEDB连接，因为我想几次使用该数据到另一个querys

查询

SELECT * FROM (
SELECT 
TT.VOUCHER, 
TT.TAXITEMGROUP, 
TT.TAXCODE, 
TT.SOURCEBASEAMOUNTCUR, 
TT.SOURCETAXAMOUNTCUR, 
TTGJAERIVA.TAXTRANSRELATIONSHIP, 
TTGJAERIVA.GENERALJOURNALACCOUNTENTRY, 
TTGJAERIVA.LEDGERDIMENSION, 
GJAERIVA.TEXT, 
GJAERIVA.LEDGERDIMENSION AS LEDGERDIMENSIONGAE, 
GJAERIVA.POSTINGTYPE 
FROM TAXTRANS TT 
INNER MERGE JOIN TAXTRANSGENERALJOURNALACCOUNTENTRY TTGJAERIVA ON TTGJAERIVA.TAXTRANS = TT.RECID 
INNER MERGE JOIN GENERALJOURNALACCOUNTENTRY GJAERIVA ON TTGJAERIVA.GENERALJOURNALACCOUNTENTRY = GJAERIVA.RECID 
AND GJAERIVA.POSTINGTYPE IN(14,236,71,41) 
AND TT.TRANSDATE <= '2015-04-30' 
WHERE 
(TT.TAXORIGIN <> 11 AND TT.TRANSDATE BETWEEN '2015-01-01' AND '2015-04-30') OR (TT.TAXORIGIN = 11 AND TT.TRANSDATE BETWEEN '2015-01-01' AND '2015-04-30')) AS T 

然后我创建另一个OLEDB连接在那里我有另一个查询，我使用的数据来自第一查询

第二个查询（仅限于相关的代码）

LEFT MERGE JOIN 
(SELECT TTRIVA.*,TTGJAERIVA.GENERALJOURNALACCOUNTENTRY FROM TAXTRANS TTRIVA 
INNER MERGE JOIN TAXTRANSGENERALJOURNALACCOUNTENTRY TTGJAERIVA ON TTGJAERIVA.TAXTRANS = TTRIVA.RECID 
INNER MERGE JOIN GENERALJOURNALACCOUNTENTRY GJAERIVA ON TTGJAERIVA.GENERALJOURNALACCOUNTENTRY = GJAERIVA.RECID 
AND GJAERIVA.POSTINGTYPE IN(14,236,71) AND TTRIVA.TRANSDATE <[email protected]) TTRIVA ON TT.VOUCHER = TTRIVA.VOUCHER 
AND TT.SOURCERECID = TTRIVA.SOURCERECID 
AND TT.TAXITEMGROUP = TTRIVA.TAXITEMGROUP 
AND (TTRIVA.TAXCODE LIKE 'RIVA%') 
AND TTRIVA.GENERALJOURNALACCOUNTENTRY = TTGJAE.GENERALJOURNALACCOUNTENTRY 


--JOIN PARA RETENCIONES DE ISR PAGADAS 
LEFT MERGE JOIN 
(SELECT TTRISR.*,TTGJAERIVA.GENERALJOURNALACCOUNTENTRY FROM TAXTRANS TTRISR 
INNER MERGE JOIN TAXTRANSGENERALJOURNALACCOUNTENTRY TTGJAERIVA ON TTGJAERIVA.TAXTRANS = TTRISR.RECID 
INNER MERGE JOIN GENERALJOURNALACCOUNTENTRY GJAERIVA ON TTGJAERIVA.GENERALJOURNALACCOUNTENTRY = GJAERIVA.RECID 
AND GJAERIVA.POSTINGTYPE IN(14,236,71) AND TTRISR.TRANSDATE <[email protected]) TTRISR ON TT.VOUCHER = TTRISR.VOUCHER AND TT.SOURCERECID = TTRISR.SOURCERECID 
AND TT.TAXITEMGROUP = TTRISR.TAXITEMGROUP 
AND (TTRISR.TAXCODE LIKE 'RISR%') 
AND TTRISR.GENERALJOURNALACCOUNTENTRY = TTGJAE.GENERALJOURNALACCOUNTENTRY 

--JOIN PARA IEPSCUOTA 
LEFT MERGE JOIN 
(SELECT TTIEPSCUOTA.*,TTGJAERIVA.GENERALJOURNALACCOUNTENTRY FROM TAXTRANS TTIEPSCUOTA 
INNER MERGE JOIN TAXTRANSGENERALJOURNALACCOUNTENTRY TTGJAERIVA ON TTGJAERIVA.TAXTRANS = TTIEPSCUOTA.RECID 
INNER MERGE JOIN GENERALJOURNALACCOUNTENTRY GJAERIVA ON TTGJAERIVA.GENERALJOURNALACCOUNTENTRY = GJAERIVA.RECID 
AND GJAERIVA.POSTINGTYPE IN(14,236,71) AND TTIEPSCUOTA.TRANSDATE <[email protected]) TTIEPSCUOTA ON TT.VOUCHER = TTIEPSCUOTA.VOUCHER AND TT.SOURCERECID = TTIEPSCUOTA.SOURCERECID 
AND TT.TAXITEMGROUP = TTIEPSCUOTA.TAXITEMGROUP 
AND (TTIEPSCUOTA.TAXCODE LIKE 'IEPSCUOTAP') 
AND TTIEPSCUOTA.GENERALJOURNALACCOUNTENTRY = TTGJAE.GENERALJOURNALACCOUNTENTRY 

正如你可以看到我用在每个LEFT JOINFROM TAXTRANS，但我有一个存储在第一查询TAXTRANS。如何用第一个查询数据替换第二个查询？

数据流：

Answer 1

什么是需要将查询和合并它时，它可以在一组办？如果您仍然希望以单独的批处理为基础查询，我建议首先将您想要重新使用的查询插入临时表（＃表）中。使用该临时表左键连接其他查询。把它封装在一个存储过程中，然后获取数据。要进一步提高性能，请检查是否可以合并索引。

合并加入与排序是一个非常昂贵的转换，它吃性能。此外，你正在寻求的是T-SQL的改进和SSIS转换不是替代品