GWT Designer suport for GXT 3.0?

[IgyBoy]

Will there be official support for GWT Designer for GXT 3.0?
If not, is there any other GUI builder/designer tool we can use with GXT 3.0?

Asking this because of this info from Google team(Eric Clayberg on google groups):

"We (Google) have no plans to provide support
for GXT 3 on our end. Support for GXT 3 is a very big project, and now that
GWT Designer is open source along with the rest of GPE, this is a project
more appropriate for Sencha to undertake (and I encourage you to suggest to
them that they do this). In fact, our position going forward is that 3rd
party component providers should provide GWTD support on their end as they
are in the best position to do so.

While the set of GXT components is more-or-less the same, the
implementation is absolutely different and has even moved to the new
namespace "com.sencha.gxt". This means that the existing GWTD support for
GXT will absolutely not work. Supporting GXT 3 will require a new set of
wbp-component.xml descriptions, new models, new GEF parts, new wizards,
etc. GXT 3 now has some support for UiBinder, so this will almost double
the effort required for full support. The existing implementation can
certainly be used as a source of ideas (and discreet pieces of source), and
we would be happy to answer questions for anyone who wants to undertake
this project.

I'm guessing that this would be a 2-4 month project for someone with GXT
expertise as well as Eclipse/GEF experience. Picking up the needed WB/GWTD
pieces should be fairly easy using the existing GXT 2.x support as a
guideline."

[Edmund]

GWT Designer support or a GUI builder is not on the roadmap for GXT. We are, however, very focused on expanding the GXT framework capabilities.

[IgyBoy]

GWT Designer support or a GUI builder is not on the roadmap for GXT. We are, however, very focused on expanding the GXT framework capabilities.

Hi Edmund.
Thank you for taking the time to answer.
It's too bad that this is not on the roadmap. I mean, in all of out projects we heavily depend on some sort of WYSIWYG editor - i believe that is also the case with lots of other companies out there.
I guess we'll postpone the migration of some projects that we intended to switch to GXT for some time.

After all,what would our design team do?

Regards,
Igor

[gishmo]

Hi Igor,

personally I don't like any WYSIWYG-editors. The generated code is awful. Think about your architecture, code same patterns und you don't need this kind of editor.

Especially with GXT 3.0 you need less code, to write amazing application.

Frank

[IgyBoy]

Hi Igor,

personally I don't like any WYSIWYG-editors. The generated code is awful. Think about your architecture, code same patterns und you don't need this kind of editor.

Especially with GXT 3.0 you need less code, to write amazing application.

Frank

Hi Frank,
I know that WYSIWYG editors are awful code generators as such,but they are sometimes necessary - enabling you to quickly create fine looking GUIs. Especially when you have a team of designers that do not necessarily need to know what is under the hood.
I think that GXT 3.0 with support for UiBinder + WYSIWYG editor is a win-win combination.
Let the designer team worry about the looks,and developers work on things under the hood.

This is how we are working now by using Microsoft tools:
Designer has it's tools(Expression Blend for example) for creating/defining a layout(in XAML) ,and doesn't need to know anything about logic underneath.
On the other hand,we have developers working in Visual Studio in C# that do not need to worry to much about GUI.
I thought that with UiBinder there will come at least basic support for GUI editing, since that would make perfect sense.
Maybe I am missing a bigger picture here,but if so, somebody, please enlighten me.

[rgalvao]

https://groups.google.com/group/google-plugin-eclipse/browse_thread/thread/e88bead9870fbce6

So nobody is going to build it? No way I would adopt GXT without editor support. Too much complexity...

[ByReaL]

any news regarding this subject ?

[Aleksandar Krstik]

For what is worth, I also support the idea of having a GUI designer for GXT3 widgets. It's true that the code they generate isn't always optimal but at least getting it out is fast.

There were numerous times when I would just slap 10-20+ GXT2.2.5 widgets in a composite just to get things done fast, and then tweak the generated code to my liking when I had more time.

Also, visualizing how things might look without the need to compile/run the app is a big + as some projects take a long time to compile and run, and it becomes a chore just so I can check whether my button is now indeed 10px to the right as intended.

I understand why Google doesn't intent to provide support for GXT3. It's like someone asking you to make your garage bigger for someone else's car. GXT3 is a 3rd party tool for them, and if they took the time to provide support for all 3rd party tools that extend their own projects they would have A LOT of extra work in front of them.

However, I really support the notion that Sencha should at least consider placing a GUI designer tool on the roadmap. UIBinder support in GXT3 is quite crippled without this tool. Of course, the advantage of separating design stuff from program logic is there, but the ability to have a "non-coder" do the layout is necessary, especially in large companies where people have different skills and tasks.

I understand that Sencha would have to dedicate time and resources to this. Time and resources that could be spent on expanding GXT3 functionality. But I don't understand how a GUI tool can be dismissed so lightly. This is a big thing to a point that it could be easily included as a few extra bucks when buying a commercial license. And I strongly believe that serious companies wouldn't mind the extra cost.

Anyway, just my two cents on the issue. For now, I'm sticking to designing layouts using GXT2.2.5, and manually modifying the code to use GXT3 layouts/widgets afterwards. It's a pain, but at least I have the idea of what I'm designing while I'm doing it.

[danilomunoz]

Any news about this issue?

I think sencha should think about this issue once the GWT Designer support could increase the gxt commercialization. In my company, we are testing tools for the web development and this is a true point that probably will be considered in our decision.

What a pity you stop considering large companies using designer tools and developers focused on the business rules!

GXT is a great product, but start using it in an older version (GXT 2 - because its support to the designer) does not make sense.


I hope you consider these guys and my opinions...

Best regards,

[kgbroce]

At the Google I/O 2012 conference, Sencha CEO Michael Mullany spoke at the end of the "History and Future of Google Web Toolkit" session and twice mentioned upcoming support for GXT 3 in GWT Designer.

Check out the session video at: http://www.youtube.com/watch?v=VOf27ez_Hvg

At 32:18 he says: "We're also happy to be working on GWT Designer integration for GXT."

At 39:00 he says: "Better IDE support and tooling - so there's some support coming in GWT Designer for GXT 3. Our goal is to make it easier and easier to create GXT applications using standard Eclipse and standard IDEs."

Google decided that future GWT development will be controlled by a GWT Steering Committee, and Sencha has a member on the committee. This decision might be why GWT Designer integration is back on the table.

Hopefully, there will be more information soon and perhaps an updated Sencha GXT Roadmap.