Reverse Intersect

[Bob Sneidar]

Hi Mark. 

That would work too I suppose. I thought about it, but remember that I want this method to be portable, that is I want it to work no matter what two databases are used, and no matter what the amount of data is. Importing a HUGE mySQL database every sync cycle would not be efficient. And once the sync occurs for the first time, the actual amount of data I will need to import is minimal. 

Trevor found an issue with sqlYoga where named database objects were reloading the database schema every time a call was made using the object. He has fixed that so speeds using sqlYoga are not noticeably any slower than the native Rev API calls. Thanks again Trevor for getting right on that. 

Because of that my paging method is working really well in conjunction with sqlYoga. My next big hurdle is issuing one query that will update multiple records at a time. I know with SQL I can separate multiple commands with the semi-colon. But I think I read somewhere that I cannot pass a semi-colon to the Rev API, which will mean I will have to update records one at a time. I hope that is not the case, because that will make any method I use with Rev unscalable for large databases. I will have to experiment on that today. 

Bob


On May 27, 2010, at 8:47 AM, Mark Wieder wrote:

> Bob-
> 
> Here's another idea: IIRC you've got a remote MySQL database and a
> local SQLite database. How about converting the local one to MySQL and
> then you could issue a single SQL query joining the two databases?
> Seems like that would simplify things, cut down on the amount of data
> transferred, and let you work with smaller recordsets...
> 
> -- 
> -Mark Wieder
> mwieder at ahsoftware.net
> 
> _______________________________________________
> use-revolution mailing list
> use-revolution at lists.runrev.com
> Please visit this url to subscribe, unsubscribe and manage your subscription preferences:
> http://lists.runrev.com/mailman/listinfo/use-revolution