Wow, someone is using trunk! Crazy, but awesome.
Unfortunately, by coincidence, I'm currently rewriting the entire
DefaultResultSetHandler -- in a serious way. It's currently in a real state
of flux.
I'll work on it this weekend some more to try to get it to a Beta 4. Your
feedback will come in handy for sure.
Clinton
On Fri, Oct 2, 2009 at 12:56 PM, Martin Ellis <mar...@ellis.name> wrote:
> Hi,
>
> I'm running into a problem using trunk, specifically with version:
> r819833 | cbegin | 2009-09-29 06:43:58 +0100 (Tue, 29 Sep 2009) | 2 lines
> Support single column mappings in primitive results
>
>
> I have a SELECT statement that uses AS to name a few columns.
> I have a resultMap that refers the AS names (not the DB column
> names), which in turn maps to property names on a bean.
>
> This all works hunky-dorey in 3.0.* releases, up to 3.0-beta-3.
> But, in the releases, loading associations seems to be broken
> in a way that I can't understand, which is why I'm trying trunk... :o)
>
> In trunk, the association loading works as I'd expect, but
> *something's* going wrong with all those renamings. There
> seems to be a case sensitivity issue in
> DefaultResultSetHandler#getRowValue().
>
> It calls loadMappedAndUnmappedColumnNames, which
> populates two lists with (uncapitalised) column names.
> This is then passed to applyPropertyMappings, which
> searches the same lists for upper-cased column names:
>
> if (propertyMapping.isCompositeResult()
> || (column != null &&
> mappedColumnNames.contains(column.toUpperCase()))) {
>
>
> I've got stuff working as I'd like with the following change:
>
> ---
> a/ibatis-3-core/src/main/java/org/apache/ibatis/executor/resultset/DefaultResultSetHandler.java
> +++
> b/ibatis-3-core/src/main/java/org/apache/ibatis/executor/resultset/DefaultResultSetHandler.java
> @@ -265,9 +265,9 @@ public class DefaultResultSetHandler implements
> ResultSetHandler {
> final String columnName = configuration.isUseColumnLabel() ?
> rsmd.getColumnLabel(i) : rsmd.getColumnName(i);
> final String upperColumnName = columnName.toUpperCase();
> if (mappedColumns.contains(upperColumnName)) {
> - mappedColumnNames.add(columnName);
> + mappedColumnNames.add(upperColumnName);
> } else {
> - unmappedColumnNames.add(columnName);
> + unmappedColumnNames.add(upperColumnName);
> }
> }
> }
>
> I wonder if perhaps someone more familiar with the code could
> check the logic there? I've been using iBATIS for all of about
> a week, so unfortunately I'm not that familiar with its workings.
>
>
> Best Regards,
>
> Martin
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: user-java-unsubscr...@ibatis.apache.org
> For additional commands, e-mail: user-java-h...@ibatis.apache.org
>
>
