How to group on records but . . .(one last try)
Well, this is a very "manual" solution, but if you don't come up with
anything else, you could step through your table as a DAO recordset, and
write your results to a second table. For each record, check if the
pertinent fields have changed. If not, update your current record; if so,
add a new record.
HTH
- Turtle
Quote:
> I want a query to group records on 3 or 4 fields but in one field (PO)
only if
> they are consecutive records. For example, let's say I have:
> PO----------------Date-------Category---------Group---------FOBExt
> 10000------------4/15--------Sleepwear-------Regular-------1000
> 10001------------4/15--------Sleepwear-------Regular-------1500
> 10003A----------4/15--------Sleepwear-------Regular-------1400
> 10004------------10/15-------Daywr--- --------Fash-----------1600
> 10005------------10/15-------Daywr---- -------Fash-----------2000
> 10006------------4/15--------Sleepwear-------Regular-------1800
> I would want the following reduction grouping on the PO, Date, Category,
Group
> and Summing FOBExt:
FirstPO----------LastPO----------Date-------Category---------Group--------
Quote:
> -SumFOBExt
10000------------10003A----------4/15--------Sleepwear-------Regular-------
3
900
10004------------10005----------10/15--------Daywr-------------Fash-------
Quote:
> ---3600
10006------------10006------------4/15--------Sleepwear-------Regular------1
800
Quote:
> Notice it groups 10000 and 10001 and 10003A on one line eventhough they
are not
> consecutive numbers (but they are consecutive fields). (Also the field
PO is 5
> numbers with an optional Letter.)
> Notice also that PO 10006 is not included in this group because, though
it has
> the same criteria on the Group By, it is not consecutive. I do group by
> queries all the time but they would compact this down to two lines with a
false
> 10000 to 10006 record. So the question is how do I separate the odd
fish?
> How to groupby with only consecutive records?
> Thanks in advance,
> Doyle60
