问题
RIGHT now I'm using ROW_NUMBER() in my procedure in SQL Server 2008 as follows:
WITH cars as(SELECT carid,mileage,retailprice,imageurl,model,year,
Zips.Distance AS Miles, Manufacturers.mfgName as Make,
dealers.companyname as companyname, CASE @sortby
WHEN 'D' THEN ROW_NUMBER() OVER (ORDER BY Manufacturers.mfgName)
WHEN 'P' THEN ROW_NUMBER() OVER (ORDER BY retailprice)
WHEN 'M' THEN ROW_NUMBER() OVER (ORDER BY mileage)
END as 'rownum'
FROM usedcars INNER JOIN #TempZips Zips ON Zips.ZipCode =usedcars.loczip
left join Manufacturers on Manufacturers.mfgid=usedcars.mfgid
left join dealers on dealers.dealerid = usedcars.dealerid
where usedcars.active=1 and usedcars.dealerid=@dealerid)
select @totalrecords as totalrec,* from cars
where rownum between @skip and @take
Is there an equivalent to this that works in SQL2000? (ROW_NUMBER wasn't introduced until SQL2005).
回答1:
You can create a temp table with an identity column, and insert your data into that.
Then use the temp table.
来源:https://stackoverflow.com/questions/3064788/row-number-alternative-for-sql-server-2000