Re: svn commit: r1401254 - /ofbiz/trunk/specialpurpose/example/servicedef/secas.xml

2012-10-24 Thread Adrian Crum
I changed the comment. There is no condition to check for not empty. -Adrian On 10/24/2012 2:30 AM, Scott Gray wrote: Yeah that could be done but conditions exist to avoid the additional overhead of executing every possible ECA service. The quality of the pattern aside, my concern is more abo

Re: svn commit: r1401254 - /ofbiz/trunk/specialpurpose/example/servicedef/secas.xml

2012-10-23 Thread Scott Gray
Yeah that could be done but conditions exist to avoid the additional overhead of executing every possible ECA service. The quality of the pattern aside, my concern is more about the confusion to developers. To randomly declare a bad practice in the example app for something that is heavily use

Re: svn commit: r1401254 - /ofbiz/trunk/specialpurpose/example/servicedef/secas.xml

2012-10-23 Thread Adrian Crum
It could be fixed by removing the ECA condition and have the createExampleStatus service do a better job of checking status changes. I agree the pattern is used heavily throughout the codebase, but that doesn't make it a good pattern. -Adrian On 10/23/2012 10:23 PM, Scott Gray wrote: Hi Adr

Re: svn commit: r1401254 - /ofbiz/trunk/specialpurpose/example/servicedef/secas.xml

2012-10-23 Thread Scott Gray
Hi Adrian, This type of pattern is pretty heavily used throughout the codebase isn't it? While updating a status history table might not be the best use of it, it does still serve as an example of that pattern. Couldn't it be fixed by including an is-not-empty condition? Thanks Scott On 24/