My only suggestion is to dump it to a boring old text file instead of a
WDDX string then use the bulk insert (assuming you're using SQL 7) like
this:
BULK INSERT page_logs
FROM '#request.dbroot#\newlogs.txt'
WITH
(FIELDTERMINATOR = '|',
ROWTERMINATOR = '~')
where #request.dbroot# is a drive on the SQL 7 box. like:
<cfset request.dbroot="\\sqlserver\c$\temp">
The reason for doing this is that you could dump 100 or a 1000 records
in like 20-30 milliseconds instead of looping over a query. Or storing
the raw WDDX.
Steve Nelson
Dan Haley wrote:
>
> Depending on your resources and access to the CF server, etc.
>
> - You could have a scheduled page that runs every few minutes and takes the
> content of the application structure and processes it - takes up a thread
> but no hit to an individual
> - Or, when you hit the "100" mark, dump the application structure into a
> WDDX string and throw it in the database, then process these outside of CF
> with something like a VB solution (pretty simple to work with WDDX in VB)
> - Or, if the per page hit isn't too big from throwing raw data into the
> database, have a scheduled job inside the database process the data instead
> of doing any in-page processing (scheduled jobs are really easy to do in
> Oracle - not available in Access)
>
> Hopefully these get you going in the right direction - I'm sure there will
> be more ideas - probably even better!
>
> Dan
>
> -----Original Message-----
> From: Ben Lowndes [mailto:[EMAIL PROTECTED]]
> Sent: Tuesday, November 21, 2000 8:13 AM
> To: Fusebox
> Subject: Fusebox internal logging.
>
> Hello everybody.
>
> I remember a couple of months ago a discussion on this list concerning
> performing logging within Fusebox... I'm about to embark on a similar thing
> and was wondering if anyone made any progress with this???
>
> I'm not actually doing full logging of every user as I don't believe that
> would be efficent, but just on a section of my site. Here's what I was
> thinking:
>
> When a user hits the page I load all the data on them into a request scoped
> structure. I then copy this into an app level struct holding many users
> details... When this app level struct contains a certain number of users
> details, e.g 100, I do a bulk insert into the Db of all these details.
>
> A few questions about this technique...
>
> What's the best way to do a Db insert from a large struct??? Do I just loop
> over the struct inside a SQL query?
> When the 100th user (or what ever I set the limit to on the struct to be)
> hits the page, the loading of their page will be considerably slowed by the
> SQL insert, is there anyway to "spawn" another process, so it wont affect
> any users??
>
> Thanks
>
> Ben.
>
> ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
> Structure your ColdFusion code with Fusebox. Get the official book at
> http://www.fusionauthority.com/bkinfo.cfm
>
> Archives: http://www.mail-archive.com/[email protected]/
> Unsubscribe: http://www.houseoffusion.com/index.cfm?sidebar=lists
>
> The information contained in this message is privileged and confidential. It is
>intended solely for the use of the named recipient. If you are not the intended
>recipient, you are hereby notified that any disclosure, copying, distribution, or use
>of the contents of this transmission is strictly prohibited. If you receive this
>message in error, please notify the sender immediately. Thank you.
> ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
> Structure your ColdFusion code with Fusebox. Get the official book at
>http://www.fusionauthority.com/bkinfo.cfm
>
> Archives: http://www.mail-archive.com/[email protected]/
> Unsubscribe: http://www.houseoffusion.com/index.cfm?sidebar=lists
--
Steve Nelson
http://www.SecretAgents.com
Tools for Fusebox Developers
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Structure your ColdFusion code with Fusebox. Get the official book at
http://www.fusionauthority.com/bkinfo.cfm
Archives: http://www.mail-archive.com/[email protected]/
Unsubscribe: http://www.houseoffusion.com/index.cfm?sidebar=lists
