我有一个像这样的表模式。
mysql> desc material_out;
+-------------+--------------+------+-----+---------+-------+
| Field | Type | Null | Key | Default | Extra |
+-------------+--------------+------+-----+---------+-------+
| id | int(11) | NO | | 0 | |
| barcode | varchar(20) | NO | | NULL | |
| name | varchar(100) | NO | | NULL | |
| qty | double | YES | | NULL | |
| unit | varchar(20) | YES | | NULL | |
| num_letter | varchar(30) | YES | | NULL | |
| date | date | YES | | NULL | |
| destination | varchar(50) | YES | | NULL | |
+-------------+--------------+------+-----+---------+-------+
8 rows in set (0.00 sec)
mysql> desc material_in;
+-------------+--------------+------+-----+---------+-------+
| Field | Type | Null | Key | Default | Extra |
+-------------+--------------+------+-----+---------+-------+
| id | int(11) | NO | | 0 | |
| barcode | varchar(20) | NO | | NULL | |
| name | varchar(100) | NO | | NULL | |
| qty | double | YES | | NULL | |
| unit | varchar(20) | YES | | NULL | |
| num_letter | varchar(30) | YES | | NULL | |
| date | date | YES | | NULL | |
| destination | varchar(50) | YES | | NULL | |
+-------------+--------------+------+-----+---------+-------+
8 rows in set (0.00 sec)
mysql> desc goods;
+-------------+--------------+------+-----+---------+-------+
| Field | Type | Null | Key | Default | Extra |
+-------------+--------------+------+-----+---------+-------+
| id | int(11) | NO | | 0 | |
| barcode | varchar(20) | NO | | NULL | |
| name | varchar(100) | NO | | NULL | |
| unit | varchar(20) | YES | | NULL | |
| category | varchar(25) | YES | | NULL | |
| first_stok | double ) | YES | | NULL | |
+-------------+--------------+------+-----+---------+-------+
6 rows in set (0.00 sec)
在表 material_out 我有 10,000 行的数据。并拥有多达 350 种条码。在表 material_in 中,我有大约 15,000 行的数据。并拥有多达 200 种条码。
我的查询是这样的
SELECT br.barcode,
COALESCE(tNEW.total_out,0) AS total_out, COALESCE(tNEW.total_in,0) as total_in,
COALESCE(tNEW.total_in,0)-COALESCE(tNEW.total_out,0) AS result,
COALESCE(tOLD.total_out,0) AS total_out_old, COALESCE(tOLD.total_in,0) AS total_in_old
FROM (
SELECT barcode from goods where category=1
) as br
LEFT JOIN (
SELECT goods.barcode,
COALESCE(SUM(tOUT.qty),0) AS total_out,
COALESCE(SUM(tIN.qty),0) AS total_in
FROM goods
LEFT JOIN material_out AS tOUT ON tOUT.barcode=goods.barcode
LEFT JOIN material_in AS tIN ON tIN.barcode=goods.barcode
WHERE goods.category=1
AND tOUT.date >='2013-05-01' AND tOUT.date <='2013-08-31'
AND tIN.date >= '2013-05-01' AND tIN.date <= '2013-08-31'
GROUP BY goods.barcode
) AS tNEW ON tNEW.barcode=br.barcode
LEFT JOIN (
SELECT goods.barcode,
SUM(tOUT.qty) AS total_out,
SUM(tIN.qty) AS total_in
FROM goods
LEFT JOIN material_out AS tOUT ON tOUT.barcode=goods.barcode
LEFT JOIN material_in AS tIN ON tIN.kode=goods.barcode
WHERE goods.category=1
AND tOUT.date BETWEEN '2013-01-01' AND '2013-04-31'
AND tIN.date BETWEEN '2013-01-01' AND '2013-04-31'
GROUP BY goods.barcode
) AS tOLD ON tOLD.barcode=br.barcode
我用过这样的查询。结果进展顺利。但是当我尝试输入新数据时。然后我重新查询你的显示。但结果与输入的数据不符。例如,当我输入您的查询时。以下结果
+--------------+-----------+----------+------------+---------------+-------------------+
| barcode | total_out | total_in | result | total_out_old | total_in_old |
+--------------+-----------+----------+------------+---------------+-------------------+
| TNWET021 | 6195 | 15000 | 8805 | 20085 | 46200 |
| TNWET020 | 3420 | 7650 | 4230 | 4860 | 23925 |
| TNWET019 | 8370 | 25200 | 16830 | 11610 | 47175 |
| TNWET018 | 18690 | 44100 | 25410 | 13800 | 54150 |
| TNWET017 | 1140 | 3750 | 2610 | 3690 | 16200 |
| TNWET016 | 19500 | 56100 | 36600 | 31725 | 111300 |
| TNWET015 | 5145 | 18150 | 13005 | 6510 | 23400 |
| TNWET014 | 33300 | 65250 | 31950 | 96300 | 262500 |
| TNWET013 | 1170 | 5625 | 4455 | 3690 | 13200 |
| TNWET012 | 720 | 2700 | 1980 | 3870 | 13800 |
| TNWET011 | 0 | 0 | 0 | 180 | 450 |
| TNWET010 | 0 | 0 | 0 | 405 | 1125 |
| TNWET009 | 0 | 0 | 0 | 0 | 0 |
当我进行手动求和并得到这样的结果时。
mysql> select sum(qty) from material_in where barcode='TNWET021' and date BETWEEN '2013-05-01' AND '2013-08-31';
+-------------+
| sum(qty) |
+-------------+
| 750 |
+-------------+
1 row in set (0.00 sec)
mysql> select sum(qty) from material_in where barcode='TNWET020' and date BETWEEN '2013-05-01' AND '2013-08-31';
+-------------+
| sum(qty) |
+-------------+
| 450 |
+-------------+
1 row in set (0.00 sec)
当数据对于报告非常重要时,为什么结果可能会有很大不同。请帮我。
我之前使用过这个,但是执行需要很长时间。也许您可以帮助总结此查询以快速执行
SELECT COALESCE(tIN.total_in,0) + COALESCE(production.total_prod,0) AS incoming, COALESCE(tOUT.total_out,0) AS expenditure,br.barcode as barcode, br.name,br.initial_stock,br.unit,COALESCE(adj.total,0) AS adjusment,COALESCE(tIN.total_in,0) + COALESCE(production.total_prod,0) + COALESCE(adj.total,0) + COALESCE(br.intial_stock,0) - COALESCE(tOUT.total_out,0) as final_stok,so.stock_opname from (
select barcode,name,initial_stock,unit from barang where category=1
) as br
LEFT JOIN (
select (select sum(qty) from material_out where date >= '2013-05-01' AND date <='2013-08-31' and barcode=a.barcode) as total_out,a.barcode from material_out a group by a.barcode
) as tOUT
ON tOUT.barcode=br.barcode
LEFT JOIN
(
SELECT(
SELECT SUM(qty) FROM adjusment
WHERE status = '+' AND date >= '2013-05-01' AND date <= '2013-08-31'
) - (
SELECT SUM(qty) FROM adjusment
WHERE status = '-' AND date >= '2013-05-01' AND date <= '2013-08-31'
) AS total,barcode FROM adjusment
GROUP BY barcode
) AS adj
ON br.barcode = adj.barcode
LEFT JOIN (
select (select sum(qty) from material_in where date >= '2013-05-01' AND date <='2013-08-31' and barcode=a.barcode) as total_in,a.barcode,a.nama from material_in a group by a.barcode
) as tIN
ON br.barcode=tIN.barcode
LEFT JOIN (
select (select sum(qty) from view_production where date >= '$start' AND date <='$end' and kode=a.kode) as total_prod,a.barcode from view_production a group by a.barcode
) as production
ON br.barcode=production.barcode
LEFT JOIN (
select(select sum(qty) from stock_opname where date >= '2013-04-01' AND date <= '2013-05-31' AND barcode=a.barcode) as stok_opname,a.barcode from pencacahan a group by a.barcode
) as so
ON br.barcode=so.barcode