Re: Cached ADO

Stephen, my caching layer is based on STL too. I use maps, as they provide
much faster indexed searching on a large emount of data. As I already
described in my previous post to Bill, the basic functionality exists. But
that's not enough for me. I need to have powerful filtering capabilities,
extremely fast searching and data manipulation. The layer must be thread-safe
and must solve all collisions in multi-user environment. I would like to have
transactions, referential integrity, relationships between tables on the
caching layer too. Well, now when I think about all these things, which I
have to implement, I decide to buy an existing solution if it exists, of
course. :) If not... Then I will continue developing my own layer.


"Stephen Howe" wrote:

> > Does anyone know any libraries (free or commercial) to cache classic ADO
> > recordsets? Actually, it was already implemented in ADO.NET, thus having a
> > database in memory. Is there anything similar to it, but not a .NET based
> > solution, working with ADO and C++ on Win32 platform?
>
> I work with ADO & VC++ and know nothing like this.
> But then why work with Recordsets at all?
>
> I find it easier to work with the STL and only use ADO when it necessary to
> update the database.
> For small-medium sized read-only tables, I use server-side, forward only
> recordsets to read into std:vectors where they can easily be
> sorted/searched/manipulated.
>
> Stephen Howe
>
>
>
.