我在Python中看到的东西绝对让我大吃一惊！
当我运行以下脚本时，我得到的结果与数据库中的结果不匹配：
import pyodbc
conn = pyodbc.connect('DSN=example;Description=example;UID=example;Trusted_Connection=Yes;APP=Microsoft Office 2010;WSID=example;Database = example')
cursor = conn.cursor()
QS = """select
cct.product_group,
cct.website,
SUM(cct.sales_value) as sales,
SUM(cct.sales_value) - SUM(cct.cogs) + SUM(cct.carriage_reclaim) - SUM(cct.carriage_costs) - SUM(cct.carriage_costs_estimate) - SUM(cct.metapack_costs) - SUM(cct.mktg_costs) - SUM(cct.mktg_costs_estimate) - SUM(cct.finance_costs) - SUM(cct.royalty) - SUM(cct.scrap_costs) as cavo,
sum(cct.budget_sales_value) as budget_sales,
sum(cct.budget_sales_phased) as budget_sales_phased,
sum(budget_sales_value - budget_cogs + budget_carriage_reclaim - budget_carriage_costs - budget_metapack_costs - budget_mktg_costs - budget_affiliate_costs - budget_offline_costs - budget_royalty - budget_finance_costs - budget_scrap_costs) as budget_cavo
from
Phocas_TG.dbo.cavo_cube_trans cct
where
cct.date between '2016-02-01' and '2016-02-01'
and
cct.product_group in ('BATH', 'HEAT', 'SOLA')
group by
cct.product_group,
cct.website
"""
cursor.execute(QS)
row = 1
while row:
row = cursor.fetchone()
print row
结果（错误）：
(u'SOLA', u'Beam LED', 0.0, 0.0, 162742.97946200587, 0.0, 18193.81407224502)
(u'BATH', u'Best Bathrooms', 0.0, 0.0, 41916.63833219, 0.0, 8616.793372278)
(u'HEAT', u'Best Bathrooms', 0.0, 0.0, 6608.969775007, 0.0, 2156.7834558070012)
(u'HEAT', u'Best Heating', 0.0, 0.0, 1221384.659527106, 0.0, 377841.5168052811)
(u'HEAT', u'Best Heating DE', 0.0, 0.0, 29173.095597537005, 0.0, 8475.738008001004)
(u'HEAT', u'Best Heating IE', 0.0, 0.0, 22200.070480626007, 0.0, 7635.299995903)
(u'BATH', u'Big Bathroom Shop', 0.0, 0.0, 194083.00040568004, 0.0, 24954.13424082598)
(u'HEAT', u'Big Bathroom Shop', 0.0, 0.0, 22149.000000066, 0.0, 8705.787840001001)
(u'SOLA', u'Big Bathroom Shop', 0.0, 0.0, 5100.480000258, 0.0, 1083.4538447940001)
(u'BATH', u'Cheap Suites', 0.0, 0.0, 96801.99990245397, 0.0, 14238.596937037)
(u'HEAT', u'Cheap Suites', 0.0, 0.0, 106569.44176593704, 0.0, 33754.97526726799)
(u'BATH', u'Hudson Reed CA', 0.0, 0.0, 37082.100495228, 0.0, 12856.460342323007)
(u'HEAT', u'Hudson Reed CA', 0.0, 0.0, 9353.568286208001, 0.0, 880.9490233900004)
(u'BATH', u'Hudson Reed DE', 0.0, 0.0, 101559.039541116, 0.0, 23292.634339761993)
(u'HEAT', u'Hudson Reed DE', 0.0, 0.0, 113215.76626921304, 0.0, 36837.99376355698)
(u'SOLA', u'Hudson Reed DE', 0.0, 0.0, 4854.377637117, 0.0, 1737.9875217079998)
(u'BATH', u'Hudson Reed ES', 0.0, 0.0, 19604.31999891699, 0.0, 4734.427979470002)
(u'HEAT', u'Hudson Reed ES', 0.0, 0.0, 34249.141983546, 0.0, 10209.710455291004)
(u'SOLA', u'Hudson Reed ES', 0.0, 0.0, 289.80000006, 0.0, 116.75621339399999)
(u'BATH', u'Hudson Reed FR', 0.0, 0.0, 347181.18108432164, 0.0, 69657.60552098301)
(u'HEAT', u'Hudson Reed FR', 0.0, 0.0, 226974.89734840707, 0.0, 66110.17134250299)
(u'SOLA', u'Hudson Reed FR', 0.0, 0.0, 4854.377637116998, 0.0, 1737.7115216809996)
(u'BATH', u'Hudson Reed IT', 0.0, 0.0, 16166.474983498987, 0.0, 3377.9398536559997)
(u'HEAT', u'Hudson Reed IT', 0.0, 0.0, 44038.997226871994, 0.0, 10705.928342997002)
(u'SOLA', u'Hudson Reed IT', 0.0, 0.0, 289.80000006, 0.0, 116.75621339399996)
(u'BATH', u'Hudson Reed NL', 0.0, 0.0, 109605.65601838099, 0.0, 21186.901077737002)
(u'HEAT', u'Hudson Reed NL', 0.0, 0.0, 111566.249953006, 0.0, 25728.021958158002)
(u'SOLA', u'Hudson Reed NL', 0.0, 0.0, 289.80000006, 0.0, 116.75621338999997)
(u'BATH', u'Hudson Reed US', 0.0, 0.0, 183865.59700310798, 0.0, 71254.06636748096)
(u'HEAT', u'Hudson Reed US', 0.0, 0.0, 45660.18595451701, 0.0, 14384.518580477998)
(u'BATH', u'Trueshopping', 0.0, 0.0, 79736.07721019002, 0.0, 17654.34511784)
(u'HEAT', u'Trueshopping', 0.0, 0.0, 74418.77915825398, 0.0, 21116.580455338993)
None
但是，如果我运行SAME脚本，但是稍微改变了sql字符串（在其中一行中添加了一点空格），我得到了正确的结果：
import pyodbc
conn = pyodbc.connect('DSN=example;Description=example;UID=example;Trusted_Connection=Yes;APP=Microsoft Office 2010;WSID=example;Database = example')
cursor = conn.cursor()
QS = """select
cct.product_group,
cct.website,
SUM(cct.sales_value) as sales,
SUM(cct.sales_value) - SUM(cct.cogs) + SUM(cct.carriage_reclaim) - SUM(cct.carriage_costs) - SUM(cct.carriage_costs_estimate) - SUM(cct.metapack_costs) - SUM(cct.mktg_costs) - SUM(cct.mktg_costs_estimate) - SUM(cct.finance_costs) - SUM(cct.royalty) - SUM(cct.scrap_costs) as cavo,
sum(cct.budget_sales_value) as budget_sales,
sum(cct.budget_sales_phased) as budget_sales_phased,
sum(budget_sales_value - budget_cogs + budget_carriage_reclaim - budget_carriage_costs - budget_metapack_costs - budget_mktg_costs - budget_affiliate_costs - budget_offline_costs - budget_royalty - budget_finance_costs - budget_scrap_costs) as budget_cavo
from
Phocas_TG.dbo.cavo_cube_trans cct
where
cct.date between '2016-02-01' and '2016-02-01'
and
cct.product_group in ('BATH', 'HEAT', 'SOLA')
group by
cct.product_group,
cct.website
"""
cursor.execute(QS)
row = 1
while row:
row = cursor.fetchone()
print row
结果（正确）：
(u'SOLA', u'Beam LED', 5065.771499999995, 96.61683876775501, 162742.97946200587, 10194.260000000007, 18193.814072245008)
(u'BATH', u'Best Bathrooms', 2953.044100000001, 92.97357616908354, 41916.63833219, 2635.92, 8616.793372277998)
(u'HEAT', u'Best Bathrooms', 714.1080000000002, 242.6900400000002, 6608.969775007001, 411.96000000000004, 2156.7834558070003)
(u'SOLA', u'Best Eco Shop', 476.55, 90.5512, 0.0, 0.0, 0.0)
(u'HEAT', u'Best Heating', 67855.21000000008, 20803.223713668747, 1221384.6595271053, 75820.06999999998, 377841.516805281)
(u'HEAT', u'Best Heating DE', 764.414853, -74.59615568499949, 29173.095597537005, 1804.9900000000002, 8475.738008001008)
(u'HEAT', u'Best Heating IE', 511.684638, 145.72208938499998, 22200.070480626004, 1384.869999999999, 7635.299995903002)
(u'BATH', u'Beste Badezimmer', 605.427752, 136.0132520000001, 0.0, 0.0, 0.0)
(u'BATH', u'Big Bathroom Shop', 9055.997500000003, 1090.7121487611153, 194083.00040568007, 12434.420000000002, 24954.134240825995)
(u'HEAT', u'Big Bathroom Shop', 1637.3744000000004, 624.0700689273556, 22149.000000066, 1391.72, 8705.787840001)
(u'SOLA', u'Big Bathroom Shop', 0.0, 0.0, 5100.480000257999, 319.48999999999995, 1083.453844794)
(u'BATH', u'Cheap Suites', 6708.027599999998, 1218.3929213940442, 96801.99990245402, 6082.849999999999, 14238.596937036995)
(u'HEAT', u'Cheap Suites', 7877.586600000003, 3094.981148824446, 106569.44176593701, 6610.709999999997, 33754.975267268004)
(u'BATH', u'Hudson Reed CA', 2450.690545, 520.74435478, 37082.100495228, 2355.16, 12856.460342323011)
(u'HEAT', u'Hudson Reed CA', 198.955982, 43.108281999999996, 9353.568286207998, 593.02, 880.94902339)
(u'BATH', u'Hudson Reed DE', 7626.790740999998, 1737.584896192042, 101559.039541116, 6444.300000000002, 23292.634339761993)
(u'HEAT', u'Hudson Reed DE', 10397.459475000007, 3731.6650875108417, 113215.76626921303, 7008.19, 36837.993763557)
(u'SOLA', u'Hudson Reed DE', 0.0, 0.0, 4854.377637117, 304.08000000000004, 1737.9875217079998)
(u'BATH', u'Hudson Reed ES', 2748.9035340000005, 965.6333561828495, 19604.319998916995, 1216.6799999999998, 4734.42797947)
(u'HEAT', u'Hudson Reed ES', 1702.1610810000002, 589.3538634675374, 34249.141983546, 2118.4200000000005, 10209.710455291004)
(u'SOLA', u'Hudson Reed ES', 0.0, 0.0, 289.80000006, 18.16, 116.75621339399999)
(u'BATH', u'Hudson Reed FR', 47273.83856200002, 14384.879203607652, 347181.18108432164, 21603.31999999999, 69657.60552098295)
(u'HEAT', u'Hudson Reed FR', 19740.58251500001, 6704.351543201098, 226974.89734840702, 13947.230000000001, 66110.171342503)
(u'SOLA', u'Hudson Reed FR', 0.0, 0.0, 4854.377637116998, 304.08000000000004, 1737.7115216809996)
(u'BATH', u'Hudson Reed IT', 1520.260361, 221.10746836602254, 16166.474983499003, 1003.3399999999998, 3377.939853656)
(u'HEAT', u'Hudson Reed IT', 3123.910959, 667.0734018041572, 44038.997226871994, 2706.38, 10705.928342997002)
(u'SOLA', u'Hudson Reed IT', 0.0, 0.0, 289.80000005999995, 18.16, 116.75621339399997)
(u'BATH', u'Hudson Reed NL', 16583.111918, 5610.879630999159, 109605.65601838099, 6872.79, 21186.901077737002)
(u'HEAT', u'Hudson Reed NL', 10411.469125, 3712.277327632342, 111566.249953006, 6818.089999999997, 25728.021958158024)
(u'SOLA', u'Hudson Reed NL', 0.0, 0.0, 289.80000006, 18.16, 116.75621338999997)
(u'BATH', u'Hudson Reed US', 8481.261439000004, 3675.9046331535123, 183865.59700310792, 11516.219999999996, 71254.06636748099)
(u'HEAT', u'Hudson Reed US', 2097.8916030000005, 938.4614114663307, 45660.18595451701, 2833.7799999999993, 14384.518580478)
(u'BATH', u'Trueshopping', 5265.242654999999, 583.9883320369803, 79736.07721019002, 5047.579999999999, 17654.345117839996)
(u'HEAT', u'Trueshopping', 5519.389900000001, 1508.0805920000003, 74418.77915825396, 4634.490000000001, 21116.580455338986)
(u'SOLA', u'Trueshopping', 777.320599, 268.97614415250007, 0.0, 0.0, 0.0)
None
如果有帮助，数据库本质上是一个每晚重建的数据立方体。第一个查询的结果似乎与昨天的数据库相匹配 - 这可能是某种缓存系统吗？哪些结果存储在已运行的查询中？
我很困惑！有人见过这个吗？