开发者

T-SQL Help - Pivot Like function to merge up some data

开发者 https://www.devze.com 2023-03-09 20:46 出处:网络
I have a T-SQL Question to ask, normally SQL is my forte, but for some reason I am stumped on the following. So here it goes.

I have a T-SQL Question to ask, normally SQL is my forte, but for some reason I am stumped on the following. So here it goes.

I have a table with the following format:

CREATE TABLE [tmp_LoadDataWithAutoID](
    [DataID] int PRIMARY KEY IDENTITY(1,1),
    [SiteType] [varchar](800) NULL,
    [RegionId] [varchar](800) NULL,
    [RegionName] [varchar](800) NULL,
    [Site] [varchar](800) NULL,
    [SiteName] [varchar](800) NULL,
    [PluCode] [varchar](800) NULL,
    [Quantity] [varchar](800) NULL,
    [SalesValue] [varchar](800) NULL,
) ON [PRIMARY]

This is loaded in from a TXT Fule using BCP.

Th problem with the file is that the Store information is not given on every row, and each record does not tell me what the store information is, it just automatcially the row above that has the information in it.

1   F   2 OR    AUCKLAND 100021 KAITAIA 1   1   11.95   SHH LIMITED 2038    1688    732
2                       10  626.95  4620.16 
3                       2   71278.466   121049.29   NULL
4                       2807    12  -3.48   NULL
5                       29470000105 68  217.6

So what I have done is ran query that grabs the store information that I have with its DataID

 DataID SiteType    RegionId    RegionName  SiteSite    SiteName    PluCode Quantity    SalesincGST Dealer  FuelOnly    ShopOnly    Both
1   F   2 OR    AUCKLAND    100021  KAITAIA 1   1   11.95   SHIRBERT LIMITED    2038    1688    732
628 F   2 OR    AUCKLAND    100062  KAIKOHE 2   78382.017   130780.28   STAGE 3 LIMITED 4097    1761    1159
1301    F   2 OR    AUCKLAND    100074  PAIHIA  2   51332.505   86250.69    STAGE 3 LIMITED 1860    996 709
1859    F   7 OR    BAY OF PLENTY/CENTRAL PLATEAU   101232  OPOTIKI 2   28869.128   48474.4 D & L INVESTMENTS LIMITED   1660    1558    606

So I have these ID's 1 through to 627, 628 - 1301 AND 1302 - 1858.

The question is, how can iIpivot and join this table back onto itself to work out what is the min and max DataID for each store?

***EDIT ****

To make it easier I will describe a smaller Subset of the table.

DataID  SiteSite    SiteName
1   100021  STORE KAITAIA
628 100062  STORE KAIKOHE
1301    100074  STORE PAIHIA
1859    101232  STORE OPOTIKI
2383    101250  STORE TE KUMI
3135    101359  STORE MIDWAY
4055    101381  STORE WAIROA
4988    101429  STORE TARADALE
5540    101448  STORE TAMATEA
6062    101460  STORE HAVELOCK NORTH
6556    101481  STORE KARAMU ROAD
6971    101500  STORE PARKVALE
7423    101605  STORE SEAVIEW ROAD
7812    101624  STORE MERRILANDS
8334    101646  STORE INGLEWOOD

I need to create query that inverts the table something like this:

SiteSite,  SiteName,           MinDataID, MaxDataID
100021     STORE KAITAIA       1           627
100062     STORE KAIKOHE       628         1300
100074     STORE PAIHIA 开发者_运维百科       1301        1858 


Once you have Site information for each row . You can use the below query to get the Minimum and Maximum Data ID for each store.

First part to incorporate the row numbers

Select ROW_NUMBER() Over(Order by dataId) as RowNo, * into #temp1 from [tmp_LoadDataWithAutoID]

Second part will get the minimum dataid

Select Min(dataid) as [MinDataID]
into #temp2 From [tmp_LoadDataWithAutoID]
Group By Site 

Here you can get the min and max DataId

Select t2.[MinDataID] MinDataID, 
(Case When 
    (Select DataID - 1 from #temp1 Where RowNo = t1.RowNo + 1) is Null Then t2.[MinDataID] 
Else 
    (Select DataID - 1 from #temp1 Where RowNo = t1.RowNo + 1) 
End) MaxDataID
From #temp1 t1
Inner Join #temp2 t2 on t1.DataID = t2.[MinDataID]

Finally drop the temporary tables

drop table #temp1
drop table #temp2


Worked it out finally, Thanks for looking at it though.

Answer was:

SELECT DISTINCT DMRA.SiteSite, DMRA.SiteName, MaxDataID = DMRA.DataID, MinDataID =( MIN(DMRB.DataID))
FROM #tmp_DataMarketRecords DMRA
LEFT JOIN #tmp_DataMarketRecords DMRB ON  DMRA.DataID < DMRB.DataID
GROUP BY DMRA.DataID, DMRA.SiteSite, DMRA.SiteName
ORDER BY MaxDataID
0

精彩评论

暂无评论...
验证码 换一张
取 消

关注公众号