Question

我不是一个真正的程序员，我正在尝试查询给定年份的月份报价，采购订单和发票。我遇到的一些问题（尝试专门使用查询向导）已链接上面列出的3个类别的日期。例如

InvoiceCost|InvoiceDate|QuoteCost|QuoteDate|POCost|PODate
$$$        |  Date     |  $$$    |  Date   | $$$  | Date
$$$        |  Date     |  $$$    |  Date   | $$$  | Date

（我希望有效）我想从这样的表中获得的是一个查询，它对类别进行求和并按月对它们进行排序。我得到的是行列出所有独特的日期组合，其类别总和为Via -

SELECT DISTINCTROW Format$([Table].[invoiceDate],'mmmm yyyy') AS [invoiceDate By Month],
                   Format$([Table].[PODate],'mmmm yyyy') AS [PODate By Month],  
                   Format$([Table].[quoteDate],'mmmm yyyy') AS [quoteDate By Month],
                   Sum(Table.invoiceCost) AS [Sum Of invoiceCost], 
                   Sum(Table.POCost) AS [Sum Of POCost],
                   Sum(Table.quoteCost) AS [Sum Of quoteCost]
FROM Table
GROUP BY Format$([Table].[invoiceDate],'mmmm yyyy'), 
         Format$([Table].[PODate],'mmmm yyyy'), 
         Format$([Table].[quoteDate],'mmmm yyyy'), 
         Year([Table].[invoiceDate])*12+DatePart('m',[Table].[invoiceDate])-1,       
         Year([Table].[PODate])*12+DatePart('m',[Table].[PODate])-1, 
         Year([Table].[quoteDate])*12+DatePart('m',[Table].[quoteDate])-1;

我想知道一些SQL“魔法”，它会链接表格中的日期列以避免唯一的日期组合。

修改

更具体地说，

<style type="text/css">
   table.tableizer-table {
   border: 1px solid #CCC; font-family: Arial, Helvetica, sans-serif
   font-size: 12px;} 
.tableizer-table td {
   padding: 4px;
   margin: 3px;
   border: 1px solid #ccc;}
.tableizer-table th {
   background-color: #104E8B; 
   color: #FFF;
   font-weight: bold;}
</style><table class="tableizer-table">
<tr class="tableizer-firstrow"><th>invoiceDate</th><th>QuoteDate</th><th>PODate</th><th>sumInvoice</th><th>sumQuote</th><th>sumPO</th></tr>
<tr><td>&nbsp;</td><td>17-Jul-13</td><td>01-Jul-13</td><td>$0.00</td><td>$0.00</td><td>$0.00</td></tr>
<tr><td>30-Jul-13</td><td>01-Jan-14</td><td>31-Mar-14</td><td>$100.00</td><td>$10.00</td><td>$48.00</td></tr>
<tr><td>30-Dec-13</td><td>01-Feb-14</td><td>10-Jan-14</td><td>$615.00</td><td>$50.00</td><td>$23.00</td></tr>
<tr><td>03-Jan-14</td><td>01-Jan-14</td><td>16-Mar-14</td><td>$3.00</td><td>$100.00</td><td>$615.00</td></tr>
<tr><td>16-Jan-14</td><td>01-Jan-14</td><td>20-Feb-14</td><td>$65.00</td><td>$50.00</td><td>$156.00</td></tr>
<tr><td>03-Mar-14</td><td>24-Jan-14</td><td>24-Feb-14</td><td>$0.00</td><td>$200.00</td><td>$0.00</td></tr>
<tr><td>11-Mar-14</td><td>18-Dec-13</td><td>21-Feb-14</td><td>$3,500.00</td><td>$5,000.00</td><td>$4,000.00</td></tr>
<tr><td>13-Mar-14</td><td>01-Jan-14</td><td>11-Dec-13</td><td>$15.00</td><td>$1,000.00</td><td>$231.00</td></tr>
<tr><td>15-Mar-14</td><td>30-Jan-14</td><td>19-Feb-14</td><td>$10.00</td><td>$8,000.00</td><td>$20.00</td></tr>
<tr><td>20-Mar-14</td><td>09-Jan-14</td><td>14-Feb-14</td><td>$1,000.00</td><td>$200.00</td><td>$20,000.00</td></tr>
<tr><td>21-Mar-14</td><td>01-Mar-14</td><td>20-Feb-14</td><td>$15.00</td><td>$10,000.00</td><td>$45.00</td></tr>
</table>

请注意，行中的日期会产生组合。我想看的是这样的

invoiceD|POD    |quoteD |SumOfinvoiceCost|SumOfPOCost|SumOfquoteCost
1-2014  |1-2014 |1-2014 |$68.00          |$23.00     |$9,560.00
12-2013 |12-2013|12-2013|$615.00         |$231.00    |$5,000.00
3-2014  |3-2014 |3-2014 |$4,540.00       |$663.00    |$10,000.00
7-2013  |7-2013 |7-2013 |$100.00         |$0.00      |$0.00

我已经能够通过总共4个查询来实现这一点（3个用于按月分别生成类别总和，第4个用于将日期链接在一起。）SQL是否可以在一个查询中完成所有这些操作或者我应该在访问中创建3个单独的表来存储彼此远离的类别？

如果需要更多信息，请与我们联系。感谢

Answer 1

我想知道一些链接日期列的SQL“魔法” 在表格内，以避免独特的日期组合。

要避免使用唯一日期组合来获取您无法按所有列分组的总和，您必须仅按日期进行分组：

SELECT invoiceDate, QuoteDate, PODate, month(invoiceDate) as month,
      sum(invoiceCost) as sumInvoice, 
      sum(QuoteCost) as sumQuote, 
      sum(SOCost) as sumSO
FROM Table
GROUP BY invoiceDate, QuoteDate, PODate, month(invoiceDate) as month

DISTINCT上不需要SELECT条款。

要按月分组，您必须在select和group by子句中加入该字段，但如果您没有字段month，请尝试使用month()在其中一个日期字段上运行，它具有可能出现的所有月份值。

Answer 2

我认为你想要的是：

SELECT Format$(thedate,'mmmm yyyy') as MMMYYYY,
       Sum(invoiceCost) AS [Sum Of invoiceCost], 
       Sum(POCost) AS [Sum Of POCost],
       Sum(quoteCost) AS [Sum Of quoteCost]
FROM ((select invoiceDate as thedate, invoiceCost, NULL as POCost, NULL as quoteCost
        from Table
       ) union all
       (select PODate as thedate, NULL as InvoiceCost, POCost, NULL as quoteCost
        from Table
       ) union all
       (select QuoteDate as thedate, NULL as InvoiceCost, Null as POCost, NULL as quoteCost
        from Table
       )
      ) as t
GROUP BY Format$(thedate, 'mmmm yyyy')
ORDER BY min(thedate);

编辑：

Access可能不允许子查询中允许的子查询。您也可以使用左外连接执行此操作，如果您有月份列表。所以，如果我假设所有月份都有发票，那么你可以这样做：

SELECT i.MMMYYYY,
       i.invoiceCost AS [Sum Of invoiceCost], 
       p.POCost AS [Sum Of POCost],
       q.quoteCost AS [Sum Of quoteCost]
FROM ((select Format$(invoiceDate, 'mmm yyyy') as mmmyyyy, sum(InvoiceCost) as InvoiceCost
       from Table t
       group by Format$(invoiceDate, 'mmm yyyy')
      ) i left join
      (select Format$(PODate, 'mmm yyyy') as mmmyyyy, sum(POCost) as POCost
       from Table t
       group by Format$(PODate, 'mmm yyyy')
      ) p
      on i.mmmyyyy = p.mmmyyyy
     ) left join
     (select Format$(QuoteDate, 'mmm yyyy') as mmmyyyy, sum(QuoteCost) as quoteCost
      from Table t
      group by Format$(QuoteDate, 'mmm yyyy')
     ) q
     on i.mmmyyyy = q.mmmyyyy

从同一个表链接的SQL查询

2 个答案: