Making cegui more mobile - oriented.

[CrazyEddie]

Thanks for the vote

[Tiresias]

critical_mass++;

[DtD]

Because multi-touch is a application specific thing, it might be a better idea to allow multiple click events at multiple positions to occur, but nothing else fancy for the most part. Things like the "scroll inertia" should probably be implemented by the application.

As for an iPhone port, I support. It shouldn't be hard at all actually. Irrlicht already has a branch for the iPhone, all that might need to be done for CEGUI working in its current state would be to compile it for the iPhone.

~DtD

[Tiresias]

since android just released the new ndk with openGl es support was wondering if you have a better view on this topic?

[CrazyEddie]

Wow! It's been, what, two days? - I must get on this immediately (that's a joke, btw, I'm not just being a dick ).

In all seriousness, I am aware of the news as regards to the release of NDK 3, and upon reading the news, I did think of you guys who are interested in this. We also now have the GLES renderer in trunk (though without build system support at the moment), so many of the pieces are now in place. The main obstacle at the moment is manpower, it would be much cooler all round if someone else was prepared to take this ball and run with it. While this is something that I myself will inevitably get to eventually, I am unable to commit myself to any definite course of action at this time; I am just one person and I can not do everything, and can not be all things to all people all the time

Presently I can just about cope with maintenance work on the v0-7 code; FYI there has been no real 'new' development at all on CEGUI since December last year. On top of this situation, I am also committed to writing a skin editor tool which is also pretty much stalled.

CE.

[Tiresias]

i know its hard task but if may people are interested then u can get funded and find help,
but well it was just to have some news
let us know if there are some changes on this,
cheers

[CrazyEddie]

Yeah, sure, and I'll keep you guys updated.

This, and all the other things I've said we'll do, are very important to me and to the project. It's just somewhat difficult when I'm the only person working on the code. This is something we're going to look to address, by trying to bring more people in, though obviously it's better to get people involved directly to an area that's important to them (as opposed to giving people tasks to complete, which is not quite the same when people are volunteering We are not, and likely will never be, in any position to pay people to work on the project, that's not a direction I'm interested in heading regardless of 'funding' - developers on CEGUI must work voluntary).

The idea of open source - after all - is so that others can get involved and help out, not so that I can sit here and do all the work Or to put it another way, it will get done a lot sooner if people helped out - also, it's not necessarily up to me to 'find help' - you guys here on the forum are the 'pool of interested parties' that form the likely source of contributors - it's no use going 'outside' because people who have no interest in the project, by definition, will not be interested

CE.

[Tiresias]

actually i have some questions about this topic:

technically:
1. could you at least provide some first evaluation of the amount of work needed to compile the CeGUI lib on Android NDK, i.e:
11. Not openGL part
12. OpenGL rewrite into OpenGL ES + other changes "mobile interactiv screen" related for ex etc

2. if some company (or group) decides to do the job (providing cegui for android ndk), how would it work ? (i mean you are the creator of all of this so strange that some group does the job without you being involved or you approving the code)

[CrazyEddie]

I think I already said somewhere - if not in this thread, then another - that it's very difficult to give an estimate of the work involved because I have zero experience of the task; it would just be me sitting here pulling numbers out of my arse the air

The 'required' dependencies are basically ANSI C compliant, and CEGUI itself is standard C++, so for general porting, if the target platform has tools that support standard C and C++ - including the standard libraries - then largely the code would compile as-is (maybe not the dynamic module loading code, though if you went for a static build, that's not used anyway). As I said, whether this is the case or not, I have no clue - you would have more idea about this than I currently do

As far as OpenGL ES goes, as I already mentioned, we have a working GLES (1.1 IIRC) renderer in trunk, it just lacks build systems. With regards to input and such, again, I have absolutely no idea - if I were attempting such a thing I would be coming at it as a complete n00b; I have zero experience in this field.

There are two ways another group could go about this; first, they could do the work with a view to having any required changes merged back in, in which case things continue as normal, or alternatively they might decide to fork the code, in which case whatever they produce ceases to be CEGUI

CE.

[Tiresias]

yes u already answered this, and i think i already asked ...

ok thats what i thought about the code, c c++ ansi more or less,
which is good news,

i guess that until someone try a real port we will not know precisely what time of work will be needed, but being standard c++ and having opengl es port already tends to show that it *should* not be that difficult ... at least it worsts the try, especially because there STL is ported on NDK (STLPort),

ok thanks for input, please if u remember any big constraint or whatever would be great to post here then if anyone thinks about doing the port they can take this into account!

[mukik182]

In tho Ogre forums there's a thread in which they are trying to port ogre to android (http://www.ogre3d.org/forums/viewtopic.php?f=5&t=37608). IIRC Android NDK did not support exceptions and that would need to be compiled apart. Every problem on porting and the found workarounds are stated there.

HTH

[CrazyEddie]

Thanks for that link. Kinda sucks that it can't use those two basic standard C++ features (exceptions and rtti), and makes it pretty worthless IMO As in Ogre, our main concern is for exceptions. I recently macro-ised our exception use within CEGUI, though actually pulling off a 'removal' of exception usage within CEGUI is nigh-on impossible and if it could be done, CEGUI would suddenly become very, very fragile or even unstable.

CE.

[Mako_energy]

I apologize if I am necro'ing this post inappropriately. I registered on these forums just to express my interest in this. Specifically for the iPhone. My team is working on a game and we're nearing a point where we have to start work on the UI. We use Ogre for our graphics and that is how we heard about this library. It seems odd that you'd want to tie in with Ogre enough to change your license for this library to stay their primary UI library, but not support all their platforms(Ogre supports the iPhone last I looked, please correct me if I am wrong). Anyway, the iPhone is a platform we plan to port to and CEGUI being able to operate on it could make or break our use of your library. So I'd like to cast my vote for getting an iPhone port of CEGUI.

[CrazyEddie]

We did not change our license to suit Ogre; CEGUI moved to the MIT license many, many years before Ogre did. I'm not quite sure how you got the impression that CEGUI being licensed under MIT has anything to do with the fact that Ogre recently changed to MIT (perhaps we could claim that Ogre copied us? Since we were MIT for around four years before Ogre changed ).

I understand there are a lot of guys who would like to see this, and I can appreciate that. However, I will ask that you guys appreciate the situation that I am in. I am the only active developer on CEGUI, which means that it is very, very hard to get all the requests done, and / or done in a timely manner.

You will be aware, for example, that Sinbad was not the person who did the main work to make Ogre run on iPhone (it was, I believe, masterfalcon. And I additionally believe masterfalcon was made a full team member only after he had done this work in order that he could support and maintain it). And so the situation is the same in CEGUI. It's high time you guys realised that if you want some of this major work done, you're going to have to start contributing to the development effort, because I'm afraid one guy can't do it all

CE.

[Mako_energy]

I got that impression from here (in your FAQ):

http://www.cegui.org.uk/wiki/index.php/ ... icensed.3F

I apologize if I am pushing any buttons, I don't really expect one person to automagically will a port into existence. Nor am I sure if you were to start now that it would be ready by the time we actually needed it. I just wanted to show there was more interest rather then silently go somewhere else, so you could make a more informed decision about your development. I would like to offer help in this regard, but I'm not sure I can offer much more then research and pointing you in the right direction with some things. My team is also low on people and we're trying to create a game engine, so our hands are full as it is.