> >>anyone else has seen a problem where queries of
> queries truncates some data?
>
> Queries of queries are not a panacea and are full of flaws.
> I think the problem is that in a query, the type of columns is lost,
> and when a new query
> is made on it, some assumptions about the type of columns are made by
> CF,
> probably from the way the first row looks, and not always in the
> right direction.
>
> For instance, I've seen that, if a column in a first row contains
> only digits, it is assumed
> of numeric type, and any other row containing letters will cause an
> error.
> If a column in the first row is null, it is also assumed a certain
> type (don't remember which one),
> and susequent rows may have problems.
>
Yeah, I've found QoQ marginally useful --
It is mostly impossible to:
-- sort (order by) any thing but alpha
-- compare (<, >, between) numbers, dates, etc.
-- format numbers dates
What the implementors should is get the metadata with the original
query & then preserve the metadata for any derived queries.
Seems obvious -- wonder why they didn't do it?
Dick
[Todays Threads] [This Message] [Subscription] [Fast Unsubscribe] [User Settings] [Donations and Support]
