You are viewing a plain text version of this content. The canonical link for it is here.
Posted to derby-dev@db.apache.org by Kathey Marsden <km...@sbcglobal.net> on 2004/10/01 20:10:13 UTC

RE: [PATCH] retrieveMessages... true by default in ij

I was the only one who responded to this patch and will commit it unless
 I hear otherwise, but I wanted to offer one last chance for comment
about whether it should go into this release or the next, since it is a
behavior change for ij with the JCC driver.  If I don't hear anything by
end of day today, I'll check it in.


This is the original description and applicable patch.

Currently, the networkserver-related ij property
retrieveMessagesFromServerOnGetMessage is false by default.
The disadvantage of setting this property to true is that performance
might be slower, the advantage of setting it is that you get actual
message text.
It would be nice if this could be true by default (only relevant if the
driver is the universal driver).
It seems to me maximum performance may be less important with ij (as
opposed to access in an application) than getting helpful messages. And
if it is important, the property can easily be set false when starting
ij, by using the property: ij.retrieveMessagesFromServerOnGetMessage=false

Making the default value true also allows me to take out the references
to "framework", which is a property used in the (subset of IBM
Cloudscape) functionTests for Derby that I'm working on. Taking the
reference out makes a cleaner separation between product  & test.

Once this is agreed on, documentation should also change.


http://nagoya.apache.org/eyebrowse/ReadMsg?listName=derby-dev@db.apache.org&msgNo=378


Thanks

Kathey


Re: [PATCH] retrieveMessages... true by default in ij

Posted by myrnap <my...@Golux.Com>.
Kathey, Andrew,

Please do not apply this patch, please do not add it to the release.

>http://nagoya.apache.org/eyebrowse/ReadMsg?listName=derby-dev@db.apache.org&msgNo=378
>
>  
>
There are some issues with the patch when setting the property on the 
command line and then passing the same property with the url.
As it's essentially a behavior change, I should probably not have asked 
for it to go in this release.
I modified the hopefullly-soon-to-be-contributed testharness to force it 
whenever needed, so we're covered there.

Myrna