On 08/03/2009, Dan Fabulich <d...@fabulich.com> wrote: > > My first attempt at releasing a commons project; please be gentle. :-) > > Compatibility warning: This version is mostly a bugfix release, but to fix > DBUTILS-31 we had to upgrade the JVM dependency from JDK 1.3 to JDK 1.4. > Except for that, it is backwards compatible with DbUtils 1.1. > > PLEASE TEST THIS RELEASE WITH A REAL DATABASE! > > Although this project has reasonable unit tests, it has no integration > tests with any actual databases; it is quite possible that the fix for > DBUTILS-31 has broken something on Oracle, MS SQL Server, Derby, or your > favorite database.
The Unit test ResultSetIteratorTest looks rather odd, in that it assigns and checks the row[] array to be non-null within the iterator loop, but checks row[0] etc outside the loop. Is it really intended to only check the last row[] array? If so, it ought to check for null first. Similar comments apply to: ArrayListHandlerTest BeanListHandlerTest MapListHandlerTest BasicRowProcessorTest BeanProcessorTest Also, ProxyFactoryTest uses "instanceof" checks against methods that are declared to return the class being tested - these should surely check for null instead? > To verify DBUTILS-31, use QueryRunner to put a null value in a field, e.g. > with QueryRunner.update. Ideally it would be good to verify putting nulls > in fields of various types: char, varchar, int, boolean, date, etc. > > -- > > Tag: > > https://svn.apache.org/repos/asf/commons/proper/dbutils/tags/DBUTILS_1_2 DOAP ought to have an AL header. > Site: > > http://people.apache.org/builds/commons/dbutils/1.2/RC1/site/index.html > "...and relies only on a standard Java 1.3 or later JRE. " That should be 1.4. Given that it is now using 1.4, Junit could be updated to 3.8.2 > Binaries: > > http://people.apache.org/builds/commons/dbutils/1.2/RC1/staged/commons-dbutils/commons-dbutils/1.2/ Sigs and hashes OK N&L files look OK. Source archive agrees with SVN tag. > [ ] +1 release it > [ ] +0 go ahead I don't care > [ ] -1 no, do not release it because -0.5 because the unit tests seem wrong. > > > > --------------------------------------------------------------------- > To unsubscribe, e-mail: dev-unsubscr...@commons.apache.org > For additional commands, e-mail: dev-h...@commons.apache.org > > --------------------------------------------------------------------- To unsubscribe, e-mail: dev-unsubscr...@commons.apache.org For additional commands, e-mail: dev-h...@commons.apache.org
