Abalaban has unfortunately declared defeat when it comes to wxWidgets (please read more than just his initial post). He has given the code to billt though, so his work won't be lost. I just hope that someone else picks up the job where he left off.
Indeed abalaban has given me his source code. It was my bounty suggestion, I want this to happen. I can't work on it myself right now, my university course (VHDL), job and family don't leave me any time for fun right now. I also am unable to use OS4 for the moment, though I very much intend to correct that as soon as I can get a netbook. (I have no place to set up my XE) Until then, if there are other programmers interested, please let the amigabounty committee know.
As I stated that's not that I declared defeat rather than admitting that I lost motivation and the fact that I don't have enough inner knowledge on BOOPSI classes to do the right choices. If someone is willing to continue I'll gladly give information, advises etc. I may even do some more code, but I won't be able to finish that myself alone and seing the bounty amount I think it's more honest to allow it to be continued rather than keeping "hostage". Also in my programmer's view Qt is better tought and implemented than wxWidget . Now that Alfkil did the Qt port I see less interest in wxWidget.
@abalaban Just upload your sources somethere on SF, so they will not die in the rest. I see you give them to Billt, but no one can be sure that he will success (and he say the same : no time and co). While if it on SF and opensoruced, anyone can help bit by bit, without asking for sources someone somethere.
I merely hold them on behalf of amigabounty group until a new taker wants them. If the bounty is closed and refunded then we'll look at posting to sourceforge, until then they wait for another bounty attempt.
I'm also waiting for ivy bridge to get a suitable pc laptop for cross compiling again. Mine died in november and I'm borrowing a cheapo laptop that is not able to run my virtual linux stuff.
I disagree that qt makes wx less interesting for a couple reasons. First, some apps I am interested in are wx. Second, wx is designed so when you run something, it appears as native to your os gui. It looks like a windows or an osx or an amiga app. Qt apps look like qt apps on any os.
If I get time and machine, ill look at continuing wx myself. If someone else doesn't want to wait for that "what if", come get it. its waiting.
see billt's answer. The project started as a bounty and many people donated to it. That would not be fair if I would just put the sources on public repository at least not without aksing them about a refund or a transfert to another bounty.
@billt
wxWidget *can* appear as native... but only if the native "wrappers" are written and provide the same level of functionnality as what wxWidget expects (else you still have the solution to create a theme which will mimic the native aspect). Now Qt *can* theoretically also be skinned to look native. But that would only be a skin which IMHO is the best solution. Anyway I can understand your POV especially if you are looking after some specific apps.
see billt's answer. The project started as a bounty and many people donated to it. That would not be fair if I would just put the sources on public repository at least not without aksing them about a refund or a transfert to another bounty.
Imho no one from donars will against of putting sources on SF, so other ones can just check it in free time and maybe help somehow. That all still better than cooking the sources somethere , where they have a chance to die without seeing the light :)
@billt Quote:
If I get time and machine, ill look at continuing wx myself. If someone else doesn't want to wait for that "what if", come get it. its waiting.
Some developers for sure will not in big interst to asking for sources someone or somethere, waiting for answers, waiting for mails, etc ,etc, while they just can go at SF and works on it (or at least, they can check it to learn, or to update something trivial until skilled coder didn't take the challenge).
That all not big deal of course, and i hope you will take a challenge someday, but just our last years prove many times, that have something opening as sources, soon or later will be taken by few developers (like SDL, like Mplayer and co). But have it closed, just make it die in silence. In end its not something like heart of the OS, which should be so protected :)
wxWidget *can* appear as native... but only if the native "wrappers" are written and provide the same level of functionnality as what wxWidget expects (else you still have the solution to create a theme which will mimic the native aspect). Now Qt *can* theoretically also be skinned to look native. But that would only be a skin which IMHO is the best solution. Anyway I can understand your POV especially if you are looking after some specific apps.
That's not accurate. The whole point of wxWidgets, is that it acts as a (relatively) thin wrapper for the native GUI system on each platform. The "built-in" widgets are only there as a filler for platforms whose native GUI is inadequate; they are not meant to be the default option.
On some platforms Qt is able to get native style information and make a pretty close approximation to the native GUI. However, when it comes to having a native looking app, nothing beats actually using the native GUI, and that's what wxWidgets does (as best it can).
Both Qt and wxWidgets have their use, and their quirks and limitations.
wxWidget was an amigabounty project so it's normal sources return to amigabounty team until someone else tries to look at it.
@Hans
yes you are right wxWidget *aims* at being a thin wrapper toward native GUI system. But fact is our GUI system is behind state of the art current GUI systems and only slowly catching up (ribbon? draggable toolbars? dockable windows? etc.) As a consequence you'll have no other choice than keeping the "universal" implementation (or write a not so thin wrapper trying to fill the defiencies), unless you are improving the native GUI system at the same time (which I can't do because of lacking inner knowkledge on Reaction).
Moreover wxWidget source code really feels its previous name (wxWindows) and it really mimic Win32/MFC API.
yes you are right wxWidget *aims* at being a thin wrapper toward native GUI system.
... And it succeeds on the major platforms.
Quote:
But fact is our GUI system is behind state of the art current GUI systems and only slowly catching up (ribbon? draggable toolbars? dockable windows? etc.) As a consequence you'll have no other choice than keeping the "universal" implementation (or write a not so thin wrapper trying to fill the defiencies), unless you are improving the native GUI system at the same time (which I can't do because of lacking inner knowkledge on Reaction).
True, but that is an AmigaOS only problem. Improving the native GUI system is something which is going to be a longer term job for multiple people. We should get there eventually.
Quote:
Moreover wxWidget source code really feels its previous name (wxWindows) and it really mimic Win32/MFC API.
Nevertheless, it is the only GUI system that enables you to use the native GUI on multiple platforms. I could heap criticism on Qt's design too, but I won't. Suffice to say that there is no single GUI system in existance that doesn't have its flaws, quirks, and issues.
True, but that is an AmigaOS only problem. Improving the native GUI system is something which is going to be a longer term job for multiple people. We should get there eventually.
Yes that's true and already some work has been done to improve that (dynamic tabs, editor gadget, etc.) but the road will be long.
Quote:
I could heap criticism on Qt's design too, but I won't. Suffice to say that there is no single GUI system in existance that doesn't have its flaws, quirks, and issues.
Yes I agree and it also probably has a great part of personnal taste. Anyway let's stop this here.