As always, you hear it here, first; expect the official announcement to come monday.
Our new August releases of the entire product suite are available as scheduled, now. this includes trial builds, as well as releases for our licensed users and the free products. a lot of cool new things (and literally hundreds of fixes and minor enhancements) went into this round of releases, some highlights follow:
Oxygene now ships with the Service Pack 1 version Visual Studio 2008 in the box, at a largely reduced download size. it also introduces support for Silverlight Beta 2 and language enhancements such as a “soft” using clause, an extended “yield” keyword that can now pass back a whole nested sequence in one step, and new “parallel sequence” and “queryable sequence” types to make jt easier to work with DLinq/DA LINQ and PFX, respectively. Plus of course lots more.
Both RemObjects SDK and Data Abstract for .NET now provide support for writing rich web clients using Silverlight, as well. this was no small feat, as the Silverlight has some tight restrictions and required a bit of rearchitecting. As a result, RO/.NET also gains general support for Async method calls – as all remote calls in Silverlight need to be async.
On the database side, we have full DA LINQ support in Silverlight (in fact, there’s no DataSet in Silverlight, so that’s the only way to get data). DA is the only framework to support LINQ client-side; Microsoft’s own solutions constrain your LINQ usage to the middle tier.
Of course DA LINQ itself is new in .33 as well (we shipped an unsupported prototype in .31), and if i may say so, it rocks. i’ve been using it extensively in my wok on the new (and upcoming) remobjects.com website, and believe me, it makes data retrieval easier than ever.
Also new is DA SQL, which forms the underpinning of the LINQ support, but can also be used separately. Basically, DA SQL allows you to write plain SQL code client side, to specify how data is queried. But this code is not passed on to the back-end (that’d be a big multi-tier no-no and bypass all security, business logic, not to mention your schema abstraction). No, the SQL code is interpreted on the middle-tier and applied to the tables and field names that you have defined in your schema. In essence, you can now treaty your middle tier server like a SQL database, while still benefiting of all the multi-tier architecure and security.
DA SQL is implemented for .NET servers only, in .33 – we wanted to let the code mature and be well-tested before we port it to Delphi, which will happen next. You can consume DA SQL from .NET and Delphi clients of course (i even use it from the Mac an iPhone in my DA/OSX alpha testing).
Expect to read a lot more about DA LINQ and DA SQL, soon.
On the Delphi side, both RemObjects SDK and Data Abstract gain IDE integration into Lazarus, as well as full support for Delphi 2009 aka Tiburon. This is no small feat, as it took a lot of effort to make sure RO and DA fully leverage the new Unicode support without giving up wire and code compatibility with both older versions or RO/DA itself and of Delphi.
Unfortunately we cannot ship pre-build packages (or a trial version) for Delphi 2009 yet, because the final build was not available at the time we built our releases. But the licensed user versions of RO, DA and Hydra come with package sources that allow you to easily build and install in Tiburon. Of course we’ll rectify this as soon as the final version is out.
There’s also a new database driver (DOA) and a boat-load of new samples in both RO and DA.
You can find the full list of changes in all products at http://remobjects.com/changelogs.