Re: Auto linking of laptop FE to proper BE when remote and when on the LAN


David W. Fenton wrote:
"yerbmeld" <embreydl@xxxxxxxxx> wrote in
news:1152896393.113067.61020@xxxxxxxxxxxxxxxxxxxxxxxxxxxx:

I use the DriveExists method and the IsReady property of the
FileSystemObject to determine if the user is connected to the
network.

What if there's no mapped drive? I don't believe in mapping drives
myself -- I think it's a stupid holdover of the old days. The only
excuse for mapping drives is to support applications that have never
been updated to reflect the realities of Windows circa 1993.


I use the UNC drive name, \\servername\root, not the mapped drive
letter. Different deparments map the drive letter our department uses
and where the replica hub resides to different folders on the same
server.

I also don't believe in introducing an outside dependency for
something that can be done without the outside dependency. Thus, I
don't use the FileSystemObject, ever.


To each his own. I prefer the FileSystemObject because it gives me the
ability to check for the existance of the network drive and whether or
the drive is ready. Just in case the drive appears in the file
structure, but has become disconnected for some reason. I went this
route because of limitations you pointed out with Dir().

I have code in the exit db button of my main form that calls the
procedure that checks for network connectivity and performs the
synch if they are connected.

Conceptually this is the same as what I suggested, though I think
you should also do it on opening the application.


I do. Check two paragraphs down from my original post.

My users always work with their replica because they are mobile.

I would use the LAN replica if they are connected the LAN, because
that eliminates the possibility of conflicts for edits done while
connected to the LAN.


In my case, it is better for us to always use the local replica. The
users my start in the building with a wireless network connection and
then take their laptop and jump in a car to run a test out of wireless
range.

. . . I use
the same procedure at start-up to check for a network connection
and then check for a newer front-end version on the server. If
there is a new version, it is copied to the user's laptop and then
their local replica is snched with the hub. If there isn't a
connection, the main form is opened and they begin working.

How do you copy the front end over top of itself?


I rename the previous version as a back up, and then copy the new
version to the local folder.

--
David W. Fenton http://www.dfenton.com/
usenet at dfenton dot com http://www.dfenton.com/DFA/

.