Re: [boost] Interest in an ODBC library?

From: Phil Nash (phil.nash.lists_at_[hidden])
Date: 2005-07-27 19:38:53

On 7/27/05, Hajo Kirchhoff <mailinglists_at_[hidden]> wrote:
> Hi all,
> 
> I'd like to gauge interest in an odbc wrapper library. I am currently
> writing one and the most important difference to other libraries I've
> seen is the ease with which my library allows binding variables to odbc
> columns and parameters.

Hi,

Well I for one am keen on something like this in boost.
I was looking around for something similar a while back and toyed with
DTL (Database Template Library -
http://dtemplatelib.sourceforge.net/dtl_introduction.htm) for a while.
DTL had some nice aspects, but also some that I didn't like
(especially when I started using it for some "real" work. From first
impressions it looks you probably have most of the "nice" parts (like
easy binding), and not some of the less attractive parts (like the
non-sql interface for querying).
A few of things that you may have addressed, but which seem missing
from your examples:

1. Higher level MACRO helpers, so you can declare your struct and bind
it in the same bit of code (in addition to the two stage approach you
already showed).
2. iterator interface over "record sets" (compiant with iterator concepts).
3. As mentioned indirectly elsewhere, ability to bind to variants
(preferably one of the existing boost versions), a set of variant
tuples of various number, and an associative lookup container of
variants as the default recordset type (to fulfill the ADO-like need).

WIthout knowing more that's all I have for now - but I'll be watching
this thread with interest.
Incidentally, the project I was using DTL for previously I have now
switched to Java and JDBC, but I may end up JNI'ing back gto C++ for
ODBC stuff at a later date.

Best regards,

[)o
IhIL..