使用 LookupSet 计算总和和百分比

Question

我有一个 SQL 报告，用于计算字段实例的数量。我正在将其纳入总结报告中。这可以使用 Lookupset 函数来实现。但是，我想计算该列的总计并找到行的百分比。 例如：

首选雇主 = 是，首选雇主 = 否 我希望能够在计数列旁边添加一个百分比列，以显示每行的百分比。 最终结果总计为 100

计数公式： =Join(LookupSet(Fields!pscode.Value, Fields!pscode.Value, Fields!PrefCnt.Value, "PrefEmployer"), vbCrLF + vbCrLF)

总计： =RunningValue(Fields!PrefCnt.Value,Sum,"PrefEmployer")

百分比：

=IIF(RunningValue(Fields!PrefCnt.Value,Sum,"PrefEmployer") <> 0, Join(LookupSet(Fields!pscode.Value, Fields!pscode.Value, Fields!PrefCnt.Value, "PrefEmployer"), vbCrLF + vbCrLF)/RunningValue(Fields!PrefCnt.Value,Sum,"PrefEmployer")*100,0)

结果

SELECT
vt.pscode as pscode,
vt.PrefEmpl as PrefEmpl,
SUM(vt.PrefCnt) as PrefCnt
FROM
(
SELECT
  rtrim(p.sAddr1) + ' (' + rtrim(p.sCode) + ')' pscode,
  isnull(ltrim(lb.sPrefEmployer), 'No') PrefEmpl,
  count(lb.sPrefEmployer) PrefCnt
FROM
  property p (nolock)
  INNER JOIN tenant t (nolock) on p.hMy = t.hProperty
  INNER JOIN Leasebut28 lb (nolock) on t.hMyPerson = lb.hCode
  INNER JOIN tenstatus ts (nolock) ON (t.istatus = ts.istatus)
WHERE
  (
    t.dtMoveIn <= '2022-08-22'
    and (
      t.dtMoveOut > '2022-08-22'
      or t.dtMoveOut is null
    )
  )
  AND ts.Status in ('Current', 'Future', 'Notice')
  AND t.sUnitCode NOT IN ('COMAREA', 'WAIT')
  AND t.sUnitCode NOT LIKE ('NONRES%')
  AND lb.sCorp <> 'Yes'
  and p.scode = 'carcen2'
GROUP BY
  rtrim(p.sAddr1) + ' (' + rtrim(p.sCode) + ')',
  lb.sPrefEmployer
UNION ALL
SELECT
  rtrim(p.sAddr1) + ' (' + rtrim(p.sCode) + ')' pscode,
  isnull(ltrim(b.sPrefEmployer), 'No') PrefEmpl,
  count(b.sPrefEmployer) PrefCnt
FROM
  room r (nolock)
  INNER JOIN tenant t (nolock) ON (t.hmyperson = r.hmytenant)
  INNER JOIN unit u (nolock) ON (u.hmy = t.hunit)
  INNER JOIN unittype ut (nolock) on (ut.hMy = u.hUnitType)
  INNER JOIN tenstatus ts (nolock) ON (t.istatus = ts.istatus)
  INNER JOIN property p (nolock) ON (p.hmy = t.hproperty)
  LEFT OUTER JOIN RoomBut1 b (nolock) ON (r.hMyPerson = b.hCode)
WHERE
  isnull(r.dtmoveout, '01/01/2400') > Getdate()
  AND (
    t.dtMoveIn <= '2022-08-22'
    and (
      t.dtMoveOut > '2022-08-22'
      or t.dtMoveOut is null
    )
  )
  AND r.sRelationship <> 'Guarantor'
  AND ts.Status in ('Current', 'Future', 'Notice')
  AND t.sUnitCode NOT IN ('COMAREA', 'WAIT')
  AND t.sUnitCode NOT LIKE ('NONRES%')
  AND r.bOccupant <> -1
  and p.scode = 'carcen2'
 GROUP BY
  rtrim(p.sAddr1) + ' (' + rtrim(p.sCode) + ')',
  b.sPrefEmployer
) vt
GROUP BY
vt.pscode,
vt.PrefEmpl
Order by 2

数据集：

Answer 1

我可能误解了这一点，但如果您只想获得每个 pscode 的“是”与“否”百分比份额，那么您可以将查询更改为

SELECT
vt.pscode as pscode,
vt.PrefEmpl as PrefEmpl,
SUM(vt.PrefCnt) as PrefCnt
FROM
(
...
... existing subquery here
...
) vt
GROUP BY
vt.pscode,
vt.PrefEmpl
Order by 2

到

SELECT DISTINCT
    vt.pscode as pscode,
    vt.PrefEmpl as PrefEmpl,
    SUM(vt.PrefCnt) OVER(PARTITION BY vt.pscode, vt.PrefEmpl) as PrefCnt
    SUM(vt.PrefCnt) OVER(PARTITION BY vt.pscode, vt.PrefEmpl) 
       / SUM(vt.PrefCnt) OVER(PARTITION BY vt.pscode)
       as PrefCntPercent
FROM
(
...
... existing subquery here
...
) vt
Order by 2

Answer 2

我尝试了这个，我收到的输出为零。有什么建议吗？