一.效果图
转换后
二.sql语法
1.先是将你要的数据显示出来（当成一张表）
如 
SELECT time_test time1, STRUCTCODE, SUBSTRUCTCODE, COUNT (*) 片数, ROUND (AVG (COW_LOP1_AVG_K), 2) AS COW_LOP1_AVG_K, ROUND (AVG (COW_LOP1_AVG), 2) AS SMP_LOP1, ROUND (AVG (COW_LOP2_AVG), 2) AS SMP_LOP2, ROUND (AVG (COW_VF1_AVG), 2) SMP_VF1, ROUND (AVG (COW_WLD1_AVG), 2) SMP_WLD1, ROUND (AVG (COW_ESD04_YIELD), 2) MM400, ROUND (AVG (COW_VF4_AVG), 2) SMP_VF4, ROUND (AVG (COW_HW1_AVG), 2) SMP_HW1, ROUND (AVG (COW_IR1_YIELD), 2) SMP_IR1_YIELD, ROUND (AVG (COW_VZ1_AVG), 2) SMP_VZ1, ROUND (AVG (COW_ESD01_YIELD), 2) ESD01_YIELD, ROUND (AVG (COW_ESD03_YIELD), 2) ESD03_YIELD, ROUND (AVG (droop), 2) AS Droop FROM (SELECT DISTINCT SUBSTR (dm.COW_STARTTIME, 1, 10) time_test, SUBSTR (EPI_OUTTIME, 1, 10) time_build, SUBSTR (dm.lot, 3, 6) recipe, b.STRUCTCODE, b.SUBSTRUCTCODE, dm.COW_LOP1_AVG, dm.COW_LOP2_AVG, dm.COW_LOP1_AVG / dm.COW_LOP2_AVG AS droop, dm.COW_LOP1_AVG_K, dm.COW_VF1_AVG, dm.COW_WLD1_AVG, dm.COW_WLD2_AVG, dm.COW_VF4_AVG, dm.COW_HW1_AVG, dm.COW_IR1_YIELD, dm.COW_VZ1_AVG, dm.COW_ESD01_YIELD, dm.COW_ESD03_YIELD, dm.COW_ESD04_YIELD FROM CHIPDM.DM_SMP_PROBER dm, (SELECT lot, wo, COMPONENTID, STRUCTCODE, SUBSTRUCTCODE FROM epi.MES_EPI_WO_BOOK, epi.MES_WIP_COMP_CREATE WHERE epi.MES_EPI_WO_BOOK.lot = epi.MES_WIP_COMP_CREATE.CREATELOT) b, mes_wms_lot_nonactive wms WHERE dm.epicomponentid = wms.waferid AND dm.COW_STARTTIME IS NOT NULL AND dm.epicomponentid = b.COMPONENTID AND dm.COW_LOP2_AVG <> 0 AND dm.COW_LOP2_AVG IS NOT NULL AND SUBSTR (dm.COW_STARTTIME, 1, 10) >= '2019/12/09' AND SUBSTR (dm.COW_STARTTIME, 1, 10) <= '2020/01/02' AND SUBSTR (dm.lot, 3, 6) IN ('W1028K') AND SUBSTR (b.COMPONENTID, 12, 1) IN ('L', 'P', 'Y') AND SUBSTR (dm.LOT, 2, 1) IN ('H', 'I', 'K', 'L') AND b.STRUCTCODE IN ('W4L2D06', 'W4L2D05') AND b.SUBSTRUCTCODE IN ('01', '02', '03', '04')) GROUP BY STRUCTCODE, SUBSTRUCTCODE, time_test ORDER BY time_test
2.套语法（将你的表放入 你的sql表）
SELECT * FROM (SELECT time1,--根据时间排序 structcode || '-' || substructcode ss,--不做转换的 CASE WHEN category = '片数' THEN '片数' WHEN category = 'COW_LOP1_AVG_K' THEN 'COW_LOP1_AVG_K' WHEN category = 'SMP_LOP1' THEN 'SMP_LOP1' WHEN category = 'SMP_LOP2' THEN 'SMP_LOP2' WHEN category = 'SMP_VF1' THEN 'SMP_VF1' WHEN category = 'SMP_WLD1' THEN 'SMP_WLD1' WHEN category = 'MM400' THEN 'MM400' WHEN category = 'SMP_VF4' THEN 'SMP_VF4' WHEN category = 'SMP_HW1' THEN 'SMP_HW1' WHEN category = 'SMP_IR1_YIELD' THEN 'SMP_IR1_YIELD' WHEN category = 'SMP_VZ1' THEN 'SMP_VZ1' WHEN category = 'ESD01_YIELD' THEN 'ESD01_YIELD'-- WHEN category = 'DROOP' THEN 'DROOP'--你想转哪个字段就填哪个字段 填了的下面 UNPIVOT (VALUE FOR category IN (这里面就得填) END category,--你的SQL表里 要有字段 上面才能填 VALUE AS aa FROM ( --你的SQL表 ) UNPIVOT (VALUE FOR category IN (片数, COW_LOP1_AVG_K, SMP_LOP1, SMP_LOP2, SMP_VF1, SMP_WLD1, MM400, SMP_VF4, SMP_HW1, SMP_IR1_YIELD, SMP_VZ1, ESD01_YIELD,--和上面的字段名要相同 Droop))) PIVOT (MAX (aa) FOR time1 IN ('2019/12/09' AS "2019/12/09", '2019/12/10' AS "2019/12/10", '2019/12/11' AS "2019/12/11", '2019/12/12' AS "2019/12/12", '2019/12/13' AS "2019/12/13", '2019/12/14' AS "2019/12/14", '2019/12/15' AS "2019/12/15", '2019/12/16' AS "2019/12/16", '2019/12/17' AS "2019/12/17", '2019/12/18' AS "2019/12/18", '2019/12/19' AS "2019/12/19", '2019/12/20' AS "2019/12/20", '2019/12/21' AS "2019/12/21", '2019/12/22' AS "2019/12/22", '2019/12/23' AS "2019/12/23", '2019/12/24' AS "2019/12/24", '2019/12/25' AS "2019/12/25", '2019/12/26' AS "2019/12/26", '2019/12/27' AS "2019/12/27", '2019/12/28' AS "2019/12/28", '2019/12/29' AS "2019/12/29", '2019/12/30' AS "2019/12/30", '2019/12/31' AS "2019/12/31", '2020/01/01' AS "2020/01/01", '2020/01/02' AS "2020/01/02"))--分组列 这个我是后台sql 拼接的 动态的 ORDER BY ss,--排序不要在意 DECODE (category, '片数', 1, 'COW_LOP1_AVG_K', 2, 'SMP_LOP1', 3, 'SMP_LOP2', 4, 'SMP_VF1', 5, 'SMP_WLD1', 6, 'MM400', 7, 'SMP_VF4', 8, 'SMP_HW1', 9, 'SMP_IR1_YIELD', 10, 'SMP_VZ1', 11, 'ESD01_YIELD', 12, 'Droop', 13) ASC--自定义排序
3.合成的sql
/* Formatted on 2020/1/17 10:30:59 */ SELECT * FROM (SELECT time1, structcode || '-' || substructcode ss, CASE WHEN category = '片数' THEN '片数' WHEN category = 'COW_LOP1_AVG_K' THEN 'COW_LOP1_AVG_K' WHEN category = 'SMP_LOP1' THEN 'SMP_LOP1' WHEN category = 'SMP_LOP2' THEN 'SMP_LOP2' WHEN category = 'SMP_VF1' THEN 'SMP_VF1' WHEN category = 'SMP_WLD1' THEN 'SMP_WLD1' WHEN category = 'MM400' THEN 'MM400' WHEN category = 'SMP_VF4' THEN 'SMP_VF4' WHEN category = 'SMP_HW1' THEN 'SMP_HW1' WHEN category = 'SMP_IR1_YIELD' THEN 'SMP_IR1_YIELD' WHEN category = 'SMP_VZ1' THEN 'SMP_VZ1' WHEN category = 'ESD01_YIELD' THEN 'ESD01_YIELD' WHEN category = 'DROOP' THEN 'DROOP' END category, VALUE AS aa FROM ( SELECT time_test time1, STRUCTCODE, SUBSTRUCTCODE, COUNT (*) 片数, ROUND (AVG (COW_LOP1_AVG_K), 2) AS COW_LOP1_AVG_K, ROUND (AVG (COW_LOP1_AVG), 2) AS SMP_LOP1, ROUND (AVG (COW_LOP2_AVG), 2) AS SMP_LOP2, ROUND (AVG (COW_VF1_AVG), 2) SMP_VF1, ROUND (AVG (COW_WLD1_AVG), 2) SMP_WLD1, ROUND (AVG (COW_ESD04_YIELD), 2) MM400, ROUND (AVG (COW_VF4_AVG), 2) SMP_VF4, ROUND (AVG (COW_HW1_AVG), 2) SMP_HW1, ROUND (AVG (COW_IR1_YIELD), 2) SMP_IR1_YIELD, ROUND (AVG (COW_VZ1_AVG), 2) SMP_VZ1, ROUND (AVG (COW_ESD01_YIELD), 2) ESD01_YIELD, ROUND (AVG (COW_ESD03_YIELD), 2) ESD03_YIELD, ROUND (AVG (droop), 2) AS Droop FROM (SELECT DISTINCT SUBSTR (dm.COW_STARTTIME, 1, 10) time_test, SUBSTR (EPI_OUTTIME, 1, 10) time_build, SUBSTR (dm.lot, 3, 6) recipe, b.STRUCTCODE, b.SUBSTRUCTCODE, dm.COW_LOP1_AVG, dm.COW_LOP2_AVG, dm.COW_LOP1_AVG / dm.COW_LOP2_AVG AS droop, dm.COW_LOP1_AVG_K, dm.COW_VF1_AVG, dm.COW_WLD1_AVG, dm.COW_WLD2_AVG, dm.COW_VF4_AVG, dm.COW_HW1_AVG, dm.COW_IR1_YIELD, dm.COW_VZ1_AVG, dm.COW_ESD01_YIELD, dm.COW_ESD03_YIELD, dm.COW_ESD04_YIELD FROM CHIPDM.DM_SMP_PROBER dm, (SELECT lot, wo, COMPONENTID, STRUCTCODE, SUBSTRUCTCODE FROM epi.MES_EPI_WO_BOOK, epi.MES_WIP_COMP_CREATE WHERE epi.MES_EPI_WO_BOOK.lot = epi.MES_WIP_COMP_CREATE.CREATELOT) b, mes_wms_lot_nonactive wms WHERE dm.epicomponentid = wms.waferid AND dm.COW_STARTTIME IS NOT NULL AND dm.epicomponentid = b.COMPONENTID AND dm.COW_LOP2_AVG <> 0 AND dm.COW_LOP2_AVG IS NOT NULL AND SUBSTR (dm.COW_STARTTIME, 1, 10) >= '2019/12/09' AND SUBSTR (dm.COW_STARTTIME, 1, 10) <= '2020/01/02' AND SUBSTR (dm.lot, 3, 6) IN ('W1028K') AND SUBSTR (b.COMPONENTID, 12, 1) IN ('L', 'P', 'Y') AND SUBSTR (dm.LOT, 2, 1) IN ('H', 'I', 'K', 'L') AND b.STRUCTCODE IN ('W4L2D06', 'W4L2D05') AND b.SUBSTRUCTCODE IN ('01', '02', '03', '04')) GROUP BY STRUCTCODE, SUBSTRUCTCODE, time_test ORDER BY time_test) UNPIVOT (VALUE FOR category IN (片数, COW_LOP1_AVG_K, SMP_LOP1, SMP_LOP2, SMP_VF1, SMP_WLD1, MM400, --BS, SMP_VF4, SMP_HW1, SMP_IR1_YIELD, SMP_VZ1, ESD01_YIELD, --ESD03_YIELD, Droop))) PIVOT (MAX (aa) FOR time1 IN ('2019/12/09' AS "2019/12/09", '2019/12/10' AS "2019/12/10", '2019/12/11' AS "2019/12/11", '2019/12/12' AS "2019/12/12", '2019/12/13' AS "2019/12/13", '2019/12/14' AS "2019/12/14", '2019/12/15' AS "2019/12/15", '2019/12/16' AS "2019/12/16", '2019/12/17' AS "2019/12/17", '2019/12/18' AS "2019/12/18", '2019/12/19' AS "2019/12/19", '2019/12/20' AS "2019/12/20", '2019/12/21' AS "2019/12/21", '2019/12/22' AS "2019/12/22", '2019/12/23' AS "2019/12/23", '2019/12/24' AS "2019/12/24", '2019/12/25' AS "2019/12/25", '2019/12/26' AS "2019/12/26", '2019/12/27' AS "2019/12/27", '2019/12/28' AS "2019/12/28", '2019/12/29' AS "2019/12/29", '2019/12/30' AS "2019/12/30", '2019/12/31' AS "2019/12/31", '2020/01/01' AS "2020/01/01", '2020/01/02' AS "2020/01/02")) ORDER BY ss, DECODE (category, '片数', 1, 'COW_LOP1_AVG_K', 2, 'SMP_LOP1', 3, 'SMP_LOP2', 4, 'SMP_VF1', 5, 'SMP_WLD1', 6, 'MM400', 7, 'SMP_VF4', 8, 'SMP_HW1', 9, 'SMP_IR1_YIELD', 10, 'SMP_VZ1', 11, 'ESD01_YIELD', 12, 'Droop', 13) ASC
只要你Ctrl+C Ctrl+V 放到对应的语法中 就能实现
SELECT * FROM (SELECT time1,--根据时间排序 structcode || '-' || substructcode ss,--不做转换的 CASE WHEN category = '片数' THEN '片数' WHEN category = 'COW_LOP1_AVG_K' THEN 'COW_LOP1_AVG_K' --你想转哪个字段就填哪个字段 填了的下面 UNPIVOT (VALUE FOR category IN (这里面就得填) END category,VALUE AS aa --aa只是个自定义名称 FROM ( --放入你的表sql ) UNPIVOT (VALUE FOR category IN (片数, COW_LOP1_AVG_K -- category列的数据 字段名称数量要和case when一样 ))) PIVOT (MAX (aa) FOR time1 IN ('2019/12/09' AS "2019/12/09", '2019/12/10' AS "2019/12/10", '2019/12/11' AS "2019/12/11" ))--你的列 ORDER BY ss,--排序不要在意 DECODE (category, '片数', 1, 'COW_LOP1_AVG_K', 2) ASC--自定义排序 可有可无