Please start any new threads on our new
site at https://forums.sqlteam.com. We've got lots of great SQL Server
experts to answer whatever question you can come up with.
| Author |
Topic |
|
KidSQL
Yak Posting Veteran
88 Posts |
Posted - 2005-07-08 : 11:42:55
|
| Hello,Let's say I have a table like #mytable below:------------------------create table #mytable(date_ smalldatetime null,id varchar(3) null,val_ float null)goinsert into #mytable(date_, id, val_)values('2000-01-01', 'ABC', 4.5)insert into #mytable(date_, id, val_)values('2000-01-02', 'ABC', 4.2)insert into #mytable(date_, id, val_)values('2000-01-02', 'DEF', 8.9)insert into #mytable(date_, id, val_)values('2000-01-03', 'ABC', 4.6)insert into #mytable(date_, id, val_)values('2000-01-03', 'DEF', 9.9)insert into #mytable(date_, id, val_)values('2000-01-03', 'GHI', 22.0)goselect * from #mytable order by date_, idgodrop table #mytablego----------------------------while the results of this give me something like:date_--id--val_I would like to add another column 'count' as:date_--id--val_--countwhich would be a cumulative count of the number of items the given id has appeared in the table on or before the date of record. Thus, something like:select count(*) from #mytable where id = 'ABC' and date_ = '2000-01-03' would give me 3, 2 for the day before that, and 1 for the day before that. Does anyone have any suggestions as to how I could incorporate such a 'cumulative count' query?Any help is much appreciated. |
|
|
rohar
Starting Member
4 Posts |
Posted - 2005-07-08 : 12:54:04
|
| Something like this?:SELECT * , ( SELECT count(*) FROM #mytable m WHERE id = #mytable.id AND date_ < #mytable.date_ ) AS prevcount FROM #mytable ORDER BY date_, idROh |
 |
|
|
KidSQL
Yak Posting Veteran
88 Posts |
Posted - 2005-07-08 : 13:35:10
|
| I think so...a little change and that should do it.I appreciate your help!! |
 |
|
|
|
|
|
|
|