In this show we discuss the ins and outs of using XML as data type for SQL server. We talk about how are lives can be reconstructed off the internet. We talk a little about the validation framework Hyrule. We will be discussing this more in detail next week when we bring Dan Vega on the show. Dave talks about the security patch released for ColdFusion and how it broke some ColdFusion installations. Then the topic of Flex comes up and we talk about weather or not to Flex. Then we talk about our, currently, limited experiences with FuseGuard. Dave then talks about his failed computer and the loss of a project due to a dead hard drive. Dave also makes a failed attempt at a StarTrek reference.
Tool of the Week:
Backupify - http://www.backupify.com/
SpinRight - http://www.grc.com/sr/spinrite.htm
Show Topic Links:
Base64 Image Converter
http://base64imageconverter.riaforge.org/
Dan Vega - Hyrule
http://www.danvega.org/blog/index.cfm/Hyrule
FuseGuard 2.0 - Web Application Firewall for ColdFusion
http://foundeo.com/security/
ColdFusion : Security Bulletin APSB10-11
http://kb2.adobe.com/cps/841/cpsid_84102.html
May 14, 2010 at 6:49 PM Enjoyed the FuseGuard discussion, thanks guys!
BTW for the 8K Row Data Length limits in SQL Server, checkout using varchar(max) datatype: http://www.petefreitag.com/item/734.cfm
May 17, 2010 at 9:21 PM What're the benefits of XML batch insert over multiple INSERT statements constructed by cfloop and send over in single cfquery?
May 17, 2010 at 9:25 PM I also asked here the same question here: http://stackoverflow.com/questions/2844729/bulk-insert-to-sql-server-insert-statements-vs-one-insert-xml-statement
May 17, 2010 at 10:19 PM @Henry Personally I would not do that. If one of the insert queries were to fail it would break the entire batch. Also, there is no way of knowing the results of each insert unless you do a select after the insert looking for the data. Doing it via XML into a stored procedure gives your programatic control over each insert and the results. This would allow you to return the results of each insert as well as preventing one insert from killing the rest.
These are just a couple reasons off the top of my head.
--Dave
May 17, 2010 at 10:32 PM @Dave Thanks you. We'll look into that.