Hi Eric,
Yes, this change has caused some migration headaches. The main reason
for the change was because of inheritence issues. If the more specific
type is used in the method signature, then there was issues with
subclasses and a bug in the SourceGenerator/SourceFactory that didn't
correctly r
Another source code change that impacted my
team when we upgraded from 0.9.3.9 to 0.9.5.3 was the signature of the unmarshal
method changed from the type'd return to java.lang.Object. Now I suppose it
could be argued that this is the proper signature, and I will certainly agree
with that, bu
This one time, at band camp, Mickael Guessant said:
MG>When debugging Castor mapping file, the SQL statements issued by Castor
MG>can be incorrect. As everything is hidden in the internals of the
MG>library, it can get tricky to determine what is wrong.
MG>
MG>In such cases, having the SQL statem
When debugging Castor mapping file, the SQL statements issued by Castor
can be incorrect. As everything is hidden in the internals of the
library, it can get tricky to determine what is wrong.
In such cases, having the SQL statement in the log just before the
exception can really help...
Here is
