> > > > the repeatable situation that we see this happening is when we try and > set a string value that is defined as width X in the model and DB to > something greater than X. > > That should cause a validation error in EOF, not a string truncation error > in the database. Did you write that backwards? >
hmmm, come to think of it, that does look the wrong way round... > > my guess is that we must somehow be disposing of the exception somewhere > but i can't seem to figure it. > > > > any ideas ? > > If you are not catching and swallowing it in your code, my suspicions would > like in the JDBC driver or in the EOF plugin for MySQL. I have never seen > EOF swallow exceptions originating from a failure to commit. > hokey cokey, i'll start poking around from the driver up. i suspect we've introduced something daft somewhere that is slurping them up, but there is nothing startlingly obvious. might have to dig out some books and remind myself how exceptions float up through the stack :-p thanks for the hint simon ps. is WOVNG still progressing ?
_______________________________________________ Do not post admin requests to the list. They will be ignored. Webobjects-dev mailing list (Webobjects-dev@lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/webobjects-dev/archive%40mail-archive.com This email sent to arch...@mail-archive.com
