For me:
1) for 1.6.2 beta, I'd like to see a GA version with JIRA checked if there's
any bug for it.
2) an new version with 1.6.x with QueryForDataTable, Castle.DynamicProxy2 is
a bouns but not a must.
3) I don't mind the breaking changes currently in V3. ISqlMapper is designed
to be singleton-per-thread in 1.6.x, the 3.x code in repository addresses
this issue and provides a more flexible connection management model. It will
be better to keep these broken changes and I think it is easy to add a
wrapper that acts as ISqlMapper in 1.6.x.
4) I think the IMapperFactory interface in V3 can be discarded for some
reasons:
a) to keep the simplicity of iBATIS.NET
b) Visual Studio provides powerful DSL tools. IMapperFactory can be
easily complished with generative programming by user rather than in
iBATIS.NET.
5) NUnit 2.5 enables generic test, it will simplifies writing tests for
testing iBATIS on different database.
Yaojian
On Mon, Jun 22, 2009 at 9:28 PM, Michael McCurrey <mmccur...@gmail.com>wrote:
> Ok folks,
>
> I'd like to plan the next release of iBatis and would like to know "What do
> you want?" I would like to make a few small releases to bridge the 1.X
> line & the existing 3.X line due to the number of breaking changes in the
> 3.X line. Some things I want to do for 'just me' are:
>
> 1. Feature match the Java version
> 2. Fix an annoying list non lazy load bug
> 3. Rev to my Castle.DynamicProxy2 patch
> 4. Expose a QueryForDataTable
> 5. Drop .net 1.X compatibility.
> 6. Lastly, I would like something ambitious like Orin's Binsor for Castle
> on the configuration front.
>
>
> --
> Michael
>
