If it does indeed map to a GUID - which I do not personally know one
way or the other - the OP should read the following article, before he
plans on using it: http://www.trigeminal.com/usenet/usenet011.asp
OT, but, why does he say that a double is subject to floating point
issues? A double is an exact representation, afaik. No fp issues
involved, at all ...
A double is a floating point type and as such can only provide exact
representations of an infinitesimal subset of real numbers. All the
usual FP issues apply.
OT, but, why does he say that a double is subject to floating point
issues? A double is an exact representation, afaik. No fp issues
involved, at all ...
which was the primary reason for using a 'time_stamp'
field on SQL Server tables used with Access.
Generally, doubles are a bad choice for primary key
because, unless you are familiar with the platform,
you can't tell if they will work at all. So any
general db advice has to advise against that.
Holy cow! I always assumed they were "long long"s. I hope I haven't
used them anywhere! I certainly concur that a real/float data type
should never be used as a primary key. Thanks both, for the info.
One of the DateTime types in SQL Server is LongLong.
It maps to an Access DateTime field, and is the
field type used by the Access upsizing wizard for
Access DateTime fields.
So you always get floating point conversion on
DateTime fields when you connect Access to SQL
Server.
(david)
Ask a Question
Want to reply to this thread or ask your own question?
You'll need to choose a username for the site, which only take a couple of moments. After that, you can post your question and our members will help you out.