the enviroment I am working in, requires VFP as a front end, or middle tier (to a web server) to a Mysql, server, an and HP 3000 server. both using odbc.


I did tests (speed tests) using remote view, and found that remote views worked, but were much slower than sqlexec....

however, I still use remote views sometimes, even tho they are slower. I use them for nightly processing where speed is not an issues. The reason I choose that is that it is really eaiser to program with remote views the constructing all my queries with sql (sqlexec) espically since I dont have a datamap or data dictionary for the HP3000 so to create reports, (html reports) I had to do construct a data dictionary myself. )

so long story short, for me, if I am doing reports or extracts where speed, is not an issue, and I can create my output reprots daily, I find to program that type of thing much quicker to put in place with remote views.

however, when I need to get interactive data, with vfp a front end, or using vfp as a middle tier to a web server, I usually use sqlexec() type commands. to keep the user experience where it should be.

Bob Lee

>Hi,
>
>For those of you who are using VFP code with SQL Server, Oracle, DB2 or any other DB engine different than VFP's, I have a question:
>
>Which mechanism for working with client/server data are you using and why? I personnally know of two ways: Remote views and SQL Pass through. I know that ADO is a possibility. I want to construct the best 3-tier design possible so I need to understand the more technical issues behind each choice. I don't want to get stuck in two months thinking I should have gone another way.
>
>Thank you very much in advance for all the great advice you'll provide,
>
>Stephane.