the closer you are to the data, the faster it is to obtain.
SQL is a very simple language - and the text based commands are very versatile.
For instance, to get a line delimited list of all the data in the
field "Device_Name" in a database Devices_Main, one would use this:
SELECT `Device_Name` FROM `Devices_Main` ORDER BY `Device_Name` ASC
this would make a list like
BR00344
BR00345
..
SQL servers are optimized to deliver data in the fastest way and do a
bit of formatting and packaging of that data for you. So the speed
improvements at this point would be up to what you do on the Rev end.
And since the cache would be loaded on startup only, you'll have the
keys you need right away, and won't need to do that every time you
access the list.
If mySQL isn't fast enough on the server end, then Valentina would
be even faster. It doesn't matter really though. Once you get the
key, then getting any part of that record (or list of records) is
really quick, because the server only has to send what's needed.
The only issue with caching is that it must be updated when the
database is changed if your keyword list is to be up to date.
There's a lot you can do with getting your data organized just from
the way you ask for data from the server, before you even apply the
data in rev.
If you're on a mac, I'd suggest getting a copy of a good SQL client
such as the free "CocoaMySQL SBG" and play around with some SQL
calls. There are similar clients for windows and a couple that work
in a browser.
http://www.theonline.org/cocoamysql/
The hardier unix types would say 'use the commmand line' but I
learned a lot faster with such a client. Eventually you could write
your own similar client in Rev.
here's the SQL reference manual.
http://dev.mysql.com/doc/refman/5.0/en/index.html
I don't know if you've checked out these wonderful resources for
those working in SQLs in Revolution, but here's the links:
Sarah's
http://www.troz.net/Rev/tutorials/MySQL.rev.gz
http://www.troz.net/Rev/tutorials/MySQLtests.rev.gz
Trevor's
http://www.bluemangolearning.com/developer/revolution/
great stuff, they will show the way....
Stephen,
Thanks for all your great advice! This would be a Revolution
standalone app (thin client) that pulls data from a system on the internet.
Now when you say that the response is best accomplished with the data in
RAM - - - do you suggest that I have the host server package up the
keywords and pass it down to the thin-client as a 1 MB download upon
startup, or is
there a better way? (RAM on the server, for quick response?)
--
stephen barncard
s a n f r a n c i s c o
- - - - - - - - - - - -
_______________________________________________
use-revolution mailing list
use-revolution@lists.runrev.com
Please visit this url to subscribe, unsubscribe and manage your subscription
preferences:
http://lists.runrev.com/mailman/listinfo/use-revolution
