Yes, I agree that open protocols and published specifications are
great, but you'll want to have a good library on the client side of
the connection to make life bearable. Otherwise you are stuck parsing
every bit of data on your own. This is not HTTP and using something
like XML as the transfer protocol would be a mistake.
Ideally, code using the existing libraries would work with the server
with only minor changes. I think it will be a big disadvantage if you
have to have two separate code bases to work embedded versus
client/server.
Best,
John DeSoi, Ph.D.
>
>global valentina_server
>
> switch(theCmd)
> case "Database_SQLSelect"
> open socket to valentina_server
> write sqlCmd to socket valentina_server
> put param(2) into theSQL
> write theSQL to socket valentina_server
> read from socket valentina_server until empty
> put it into theCursor
> close socket valentina_server
> return theCursor
> break
> end switch
>end Valentina
>
>Of course I am not suggesting we trash VXCMD or V4RB, just excited
>by the possibilities of socket access.
>Also this would mean that Valentina developers can write cgi scripts
>that run on machines where they are not allowed to install add-ons
>like VXCMD. Almost all services will give you Perl, PHP, etc, but
>many will be a pain in the neck if you want to instal l something of
>your own. The net result would be that I, a Valentina developer,
>could buy hosting service anywhere with standard tools and access my
>Valentina Server elsewhere.
>
>Anyway, I hope keeping a clear, open protocol would have many
>benefits and no drawbacks for Valentina Server.
>
>Brian
©2002 John DeSoi |
