You are not logged in.

#51 2011-09-06 14:01:42

zester
Member
From: Wilkes Barre Pa
Registered: 2011-08-13
Posts: 156
Website

Re: Quantum Project (KDE4/Qt4 Fork)

Teho wrote:
zester wrote:

There is a lot to say on the topic of Qt5, Project Lighthouse & Wayland, KDE5 and how a perfect
storm of all hell is about to break loose. I dont want to make this post to long, but if you use or play any type
of application/game that uses or needs hardware acceleration graphics. Your not going to like what Qt5
has in store for you. Any hope of AAA games is dead. And the Nokia/Qt - KDE5 Devs are not only aware of this
but they don't care.

Could you please explain how Qt5 is going to ruin all apps that need hardware accerlation?

zester wrote:

Svg graphics are dead also. Unless that is you want your desktop in a web browser. I have to find
the links to the conversations me and some other developers had with Nokia/Qt devs prepair to get sick to your stomach.

QtSvg is going to remain(?) in Qt5 so I don't see the problem here. Even if it were to be removed most likely KDE developers would adopt it or qtwebkit would have matured to the point where QtSvg is no longer needed.

QtSvg is one of the modules that will not see any further development, And is being offloaded to the community. Depending on how bad binary compatibility is broken in Qt5 and
if the QtSvg module isnt updated the QtSvg module won't work with Qt5. When this was announced around 200 Qt developers expressed there concern. Nokia/Qt's
comment was ether A. you can maintain it or you self or B. you can use webkit but we are not going to be developing it any further.   

The backend windowing system code is being moved to lighthouse, and the frontend is being moved to OpenGL, what backend Qt devs that I personal know. Have hinted that X compatibility
for the time being will remain but once Wayland is viable it to will become a community supported module. Same goes for the raster engine. They plan to move to an all OpenGL/Shader/OpenCL architecture. Now at first glance this seams like a great idea.


Keep in mind we are talking about future AAA game development.

1. XLIB(Raster), Qt5(OpenGL), Ogre3D
    Its bad enough to run KWin, Compiz and Ogre3D but the situation gits even worse if everything is offloaded to
    the GPU. The Ogre3D Hydrax Demo won't even work on a non- Ati/Nvidia card.

2. Wayland(OpenGL), Qt5(OpenGL), Ogre3D
    First off without an Intel Card you will have no hardware acceleration for your graphics.
    The general community seams to think that in time they Nvidia/Ati binary blobs will be made to work with wayland.
    but this isn't necessarily true. The Nvidia/Ati drivers will have to directly link with KMS in the linux Kernel. If they do
    this they will need to open source the Nvidia/Ati drivers in order to not be in direct violation of the GPL the situation
    here is different than that of Xlib. Not even Linus can provided an exception here. So this would lead you to think that
    Ati/Nvidia will be forced to open source there drivers. But thats not possible eather. Both of those drivers/companys licenses half
    of the tech in there cards from third party patent holders. And Microsoft owns 30% of Nvidia.  Well maybe these patent holders will
    allow those parts to be open sourced? Nope thats not going to happen. Those patent holders ownly care about one thing MONEY
    and they make there MONEY from the licensing of the TECH used in those drivers. Well maybe NVIDIA will rewrite those sections?
    Thats not going to happen eather. Open Source OS's only make up 1 to 1.3% of there revenue. Ok well there is the opensource Nvidia/Ati
    drivers, Mesa, Gallium3D. This would be possible but falls under the same situation as the Nvidia/Ati drivers, patents & how long will it take
    these drivers to reach there Binary counterparts. Not to mention All four of those projects have had to re-invent things like texture compression
    because of the patent issue.

    Now when I express my concerns on Wayland (Nvidia/Ati) to the proper Nokia/Qt devs. They informed me that we will just have to use a
    compatible Intel card. And that there were other chipsets besides Nvidia/Ati with hardware acceleration.

    If you spend anytime on Qt DevNet then you will know that there main focus is on Mobile platforms.
    So I did my research on those other chipsets, and behold there mobile chipsets for Arm platforms lol.
    There talking about SGX drivers.


Hence chances are that if something extraordinary doesn't happen then Qt5 and AAA games will be ohhh such a sad situation.

And if you want to see what QtOpenGL/QtWebkit looks like
http://zester.googlecode.com/files/qt-l … viewer.zip
http://developer.qt.nokia.com/forums/viewthread/6151/


If you do a search in regards to Qt (Ogre3D, Irrlicht, Horde3D, Panda3D) youl will find that most of the real community work was done by ME lol wink

Last edited by zester (2011-09-06 14:17:44)

Offline

#52 2011-09-06 14:39:49

zester
Member
From: Wilkes Barre Pa
Registered: 2011-08-13
Posts: 156
Website

Re: Quantum Project (KDE4/Qt4 Fork)

You can take a look at QDesktop now https://github.com/zester/Chipara Note: This isn't fully functional and doesn't reflect what I have on my desktop
I am trying to work out an issue with File/Directory monitoring and what would be the correct way to update the desktop. So it is what it is a partial
preview.

In mainwindow.cpp change this line wallpaper.load("/home/steven/Picture/Arch_Linux_Wallpaper_by_james66.jpg"); to the path to the wallpaper you wish
to use. Its hard coded while I get monitoring of the config file working properly.

The issue I am having with monitoring the config file for changes to update the desktop. It works until the application you are using to edit the config
file, be it a text editor or gui app. Looses focus then it just doesn't work anymore until you restart the QDesktop app.

I want it so that others can build there own desktop config applications, and I dont want to force them to have to use a shared memory or dbus approach
but I will have to see.

The code base is very small and needs to stay that way a true KISS approach. All setting will eventually be moved to the config file, you will be able to change the
view mode from Icon to List just by changing a config option in the config file.

Last edited by zester (2011-09-06 14:49:09)

Offline

#53 2011-09-06 18:43:36

zester
Member
From: Wilkes Barre Pa
Registered: 2011-08-13
Posts: 156
Website

Re: Quantum Project (KDE4/Qt4 Fork)

I forked  LemonBoy's l(ightweight) d(evice) m(ounter) https://github.com/zester/ldm on github
Going to test it and see about integrating the code so when you mount devices they will auto appear on the desktop
and in the file manager.

QtMobilitys QtSystemInfo Module should be what I need to make this work with out directly making any changes to
ldm.

For those that are not aware QtMobility is freaking awesome. It handels everything from NetworkMangment down
to E-Mail, SIP, Everything hardware related almost. If your monitor or computer had a tilt sensor there's an Api for that also.

Eggwm also has a sister project called touchegg https://code.google.com/p/touchegg/ its for gestures and it works with any
window manager.

Manjaro Linux (Arch Linux fork)
https://github.com/ManjaroLinux has a couple of Qt4 projects going like a Login Manager and Hardware Detection Application/Library
something to keep an eye on.

So many possibilities.


Before anyone makes any comments about forking. Thats what github is all about, you see something you like you fork it.

Last edited by zester (2011-09-06 19:21:44)

Offline

#54 2011-09-06 20:14:24

zester
Member
From: Wilkes Barre Pa
Registered: 2011-08-13
Posts: 156
Website

Re: Quantum Project (KDE4/Qt4 Fork)

Any thoughts on this Audio Framework I am putting together?

QAF is being created for various reasons: to create a simple Qt style audio API, that can be used to create a range of applications from audio players,3d audio for games to audio production applications.

24n1q3r.png
Larger View

Features
Note: Just to name a few.

2D & 3D audio
Flash support
Supports encoding, decoding, muxing and demuxing most audio formats
Supports mono, stereo, quad, 5.1, 7.1 and multichannel audio devices
Allows applications to share the same "real" audio device regardless of what format is requested by the application.
Supports recording and full duplex in addition to playback.
Ability to mix stereo and multichannel audio streams up to 7.1/200Khz/32bit.
Supports full 24 bit range without loss of precision during internal computations.
Each application has its own independent volume controls.
Supports loop back recording.
Real-time, low latency connections for both audio and MIDI data between applications
Audio synthesis framework
Networked audio across the board
Optional easy to use C++ API for developer's



All of the interfaces are optional and the api will have a enumerator like phonon does that you would use depending on what type of
application your building, and depending on the enumerator you choose will depend on what interfaces(librarys) are used.

Started building this because PulseAudio is total crap. and if you say other wise it shows how much you know about audio tongue

Also I hold 1000 shares of Alcatel-Lucent stock I contacted them to see if being that I am a stock holder and my application is free/open source if I could be exempt from the mp3 patent/royalty fee's. Not that I care about mp3, but that would be a nice little bonus to slap on to this framework to be able to say "Yes using this framework you can legally process mp3's". There response was we will have to get back to you Mr Starr lol. Regardless ill stick with my Wav, Flac, Vorbis, Speex codecs wink

Notice the order in which I listed those codecs you will be quizzed later on when to use which one.

Last edited by zester (2011-09-06 21:11:35)

Offline

#55 2011-09-06 20:23:23

Lothium
Member
Registered: 2009-10-10
Posts: 192

Re: Quantum Project (KDE4/Qt4 Fork)

Oh wow... I must say, this all sounds very interesting! I use KDE for a long time, but if you go on developing Quantum Project at this speed and with this quality, I look at it in the near future. KDE is very heavy and a lightweight desktop in Qt sounds great. So, thanks for your work and please go on!

Best regards!

Offline

#56 2011-09-06 20:25:18

fsckd
Forum Fellow
Registered: 2009-06-15
Posts: 4,173

Re: Quantum Project (KDE4/Qt4 Fork)

(Off-topic posts on future of QT and KDE split to its own thread. Please continue discussion on that topic there. Thanks.)


aur S & M :: forum rules :: Community Ethos
Resources for Women, POC, LGBT*, and allies

Offline

#57 2011-09-06 21:26:31

zester
Member
From: Wilkes Barre Pa
Registered: 2011-08-13
Posts: 156
Website

Re: Quantum Project (KDE4/Qt4 Fork)

Ok Audio Codecs.

Wav - Wav is an archival format if you have the option this is the format
you want to have when converting to other formats or when your editing,
creating new sounds with a synth.
It's Uncompress Lossless (Uncompressed and Doesn't lose quality when converted)

Flac - Flac is generally used when you want to do audio production
it's tag-able so basically if your makeing your own techno song or
if you need really high quality audio with reduced size then flac is what you want.
It's Compressed Lossless (Compressed and Doesnt lose quality when converted)

Vorbis - Vorbis is a general format like mp3 it has good quality and a small size
it's simply a playback format and should never be used as a edible format. Each time
its converted it loses quality.
It's Compressed Lossy (Compressed and will lose quality when converted)

Speex - Speex is the same as vorbis but its designed for voice. Its used in voip applications
and just voice applications in general
Each time it's converted it loses quality
It's Compressed Lossy  (Compressed and will lose quality when converted)

Then you have

OGG and Matroska there container formats and
are used to generaly combine Audio with Video
but they can be used for just audio also.

These codecs form the bases for they above audio framework and if used properly will
give us an incredible audio experience.

This will be important to know because these codecs combine with the audio framework will put us
in the same class as some of the best audio frameworks there are. NSound combined with OpenAL-Soft
and  Libavcodec/Libavformat will destroy FMOD for Game Audio


I should also put this in the wiki because far to many people ask stupid questions like how do I convert my mp3 music collection to ogg  the answer is you dont! If all you already have is an mp3 then just leave it as is because you will just destroy your music collection.
You dont convert a file any file thats in a lossy format unless you want to lose quality.

Just dont buy mp3's

When ripping from cd rip to wav or flac or buy from a music store like 7digital that offers flac as a download option for your music. Also never buy music in the ogg format eather unless you never plan on doing anything with it other than just listening to it.

Last edited by zester (2011-09-06 22:28:15)

Offline

#58 2011-09-06 22:56:13

stealthy
Member
Registered: 2011-05-02
Posts: 67

Re: Quantum Project (KDE4/Qt4 Fork)

Eh, I would like to learn some Qt. Is it possible to use Compiz/Fusion-Icon as the WM, I don't really like anything else. IMO it should be lightweight but very sexy.


clipodder-git A small simple cron-friendly podcast downloader, with support for arbitrary user defined media types (pdf, html, etc...)

Offline

#59 2011-09-07 01:43:37

zester
Member
From: Wilkes Barre Pa
Registered: 2011-08-13
Posts: 156
Website

Re: Quantum Project (KDE4/Qt4 Fork)

stealthy wrote:

Eh, I would like to learn some Qt. Is it possible to use Compiz/Fusion-Icon as the WM, I don't really like anything else. IMO it should be lightweight but very sexy.

You can use any window manager that you would like this project is about breaking everything up so you can use what ever combination your heart desires.

Offline

#60 2011-09-07 06:28:18

stealthy
Member
Registered: 2011-05-02
Posts: 67

Re: Quantum Project (KDE4/Qt4 Fork)

Ok cool. We should not follow in Gnome/Xfces steps when it comes to additional programs (Xfburn, Xarchiver). We should simply make the DE (panel, plugins), with a nice configuration editor. We should keep KISS,    beauty, and functionality/flexibility in mind.


clipodder-git A small simple cron-friendly podcast downloader, with support for arbitrary user defined media types (pdf, html, etc...)

Offline

#61 2011-09-07 12:20:25

zester
Member
From: Wilkes Barre Pa
Registered: 2011-08-13
Posts: 156
Website

Re: Quantum Project (KDE4/Qt4 Fork)

stealthy wrote:

Ok cool. We should not follow in Gnome/Xfces steps when it comes to additional programs (Xfburn, Xarchiver). We should simply make the DE (panel, plugins), with a nice configuration editor. We should keep KISS,    beauty, and functionality/flexibility in mind.

Not sure what you mean but I have plans for a bunch of optional applications. But no application will depend on another. Even the config dialog for the desktop is optional.

Offline

#62 2011-09-07 13:25:33

goran'agar
Member
From: Nothern Italy
Registered: 2009-05-19
Posts: 171

Re: Quantum Project (KDE4/Qt4 Fork)

Hello zester. Do you plan to develop a file viewer (ps, pdf, etc.)? It seems there aren't qt-only pdf viewers. Or maybe I just can't find them.
btw, I like very much your idea of having no application depending on another.


Sony Vaio VPCM13M1E  - Arch Linux - LXDE

Offline

#63 2011-09-07 13:52:23

zester
Member
From: Wilkes Barre Pa
Registered: 2011-08-13
Posts: 156
Website

Re: Quantum Project (KDE4/Qt4 Fork)

goran'agar wrote:

Hello zester. Do you plan to develop a file viewer (ps, pdf, etc.)? It seems there aren't qt-only pdf viewers. Or maybe I just can't find them.
btw, I like very much your idea of having no application depending on another.

Yes I have plans for a pdf, ... viewer

Here is what I am playing with in regards to document viewers.

Standard
ht5jcy.png

and

Fancy wink
2afl7ux.png

I am also playing with a couple of different pdf librarys for editing and composing pdf's
kind of like scribus does.


And yahh all this stuff is being made to not only standalone in regards to any other application but also
I am trying to keep the code base as small and readable as possible. So experienced and non-experienced
programmers can just pick them up, hack on them, fork them or do as they please with them.

The licensing is also done the same way you can choose to sub-license under any GPL compatible licenses
be is BSD, MIT, Apache or Public Domain.

Last edited by zester (2011-09-07 14:09:54)

Offline

#64 2011-09-07 14:21:14

goran'agar
Member
From: Nothern Italy
Registered: 2009-05-19
Posts: 171

Re: Quantum Project (KDE4/Qt4 Fork)

zester wrote:
goran'agar wrote:

Hello zester. Do you plan to develop a file viewer (ps, pdf, etc.)? It seems there aren't qt-only pdf viewers. Or maybe I just can't find them.
btw, I like very much your idea of having no application depending on another.

Yes I have plans for a pdf, ... viewer
<snip>

Eh eh, I imagined that :-)
Please let us know when it's ready for testing.


Sony Vaio VPCM13M1E  - Arch Linux - LXDE

Offline

#65 2011-09-07 14:29:54

zester
Member
From: Wilkes Barre Pa
Registered: 2011-08-13
Posts: 156
Website

Re: Quantum Project (KDE4/Qt4 Fork)

goran'agar wrote:
zester wrote:
goran'agar wrote:

Hello zester. Do you plan to develop a file viewer (ps, pdf, etc.)? It seems there aren't qt-only pdf viewers. Or maybe I just can't find them.
btw, I like very much your idea of having no application depending on another.

Yes I have plans for a pdf, ... viewer
<snip>

Eh eh, I imagined that :-)
Please let us know when it's ready for testing.

I will you can also click on the website link below my name to check my github repo. And my wife and a few friends that work on the LFS project are going to
take the applications that I have break there code up and write a LDFS (Linux Desktop From Scratch) book. Lots of people have emailed me about helping out but
sadly not very many are c++ programmers which is what I need right now. But regardless this project WILL happen lol.

Soo the next year will be an interesting one. smile

Last edited by zester (2011-09-07 14:32:55)

Offline

#66 2011-09-07 16:35:28

tatsh
Member
Registered: 2011-09-07
Posts: 3

Re: Quantum Project (KDE4/Qt4 Fork)

I have used KDE since 3.5.x and I'm on Gentoo.

I am interested in your project. I've done a bit of Qt, but never with KDE dependencies. I've been highly critical of KDE doing things like:
- Akonadi works rarely, and it sucks
- Nepomuk sucks
- The licensing scheme for KDE is not very commercial friendly (all must be GPL) so we're never going to see things like Skype integrate in
- Messing with KDE PIM so that now it REQUIRES mysql and Akonadi to play well
- Kate on OS X lacks some very important features
- Kate is impossible to get on its own (this is true for all KDE apps) and it's my favourite text editor
- Early release issues and regressions therefore (like this bug: https://bugs.kde.org/show_bug.cgi?id=278891 )
- Amarok 2 is HORRIBLE!!!!!! Horrible design, horrible layout, nothing like Amarok 1, SLOW.

I've been thinking to start a new project altogether not based on KDE at all (as in borrow NOTHING from KDE), but definitely based on Qt. Qt is so nice and modular and well-designed, and almost like its own dialect of C++ that doesn't require a high knowledge of all the quirks of the language.

Try the music player Clementine. It's based on KDE 3.5's Amarok 1.4 and I love using it (although its memory footprint is not exactly where I want). The number one thing about it for me is that it doesn't rely upon KDE at all and looks and runs great. And it doesn't suck like Amarok 2.

Qt IMO only needs a lightweight desktop, nothing so fancy as the KDE project has added. Qt 3 of course was a lot smaller, so the footprint of KDE on top wasn't really felt even on older machines. Plus there was no expectation of eye candy whatsoever.

Contact me: https://github.com/tatsh
My latest work is making a toolchain for cross-compiling to OS X. Certainly could try testing that, but unfortunately I haven't yet built Qt on it (the precompiled libs on their own should work however).

Last edited by tatsh (2011-09-07 16:35:54)

Offline

#67 2011-09-07 17:01:11

zester
Member
From: Wilkes Barre Pa
Registered: 2011-08-13
Posts: 156
Website

Re: Quantum Project (KDE4/Qt4 Fork)

tatsh wrote:

I have used KDE since 3.5.x and I'm on Gentoo.

I am interested in your project. I've done a bit of Qt, but never with KDE dependencies. I've been highly critical of KDE doing things like:
- Akonadi works rarely, and it sucks
- Nepomuk sucks
- The licensing scheme for KDE is not very commercial friendly (all must be GPL) so we're never going to see things like Skype integrate in
- Messing with KDE PIM so that now it REQUIRES mysql and Akonadi to play well
- Kate on OS X lacks some very important features
- Kate is impossible to get on its own (this is true for all KDE apps) and it's my favourite text editor
- Early release issues and regressions therefore (like this bug: https://bugs.kde.org/show_bug.cgi?id=278891 )
- Amarok 2 is HORRIBLE!!!!!! Horrible design, horrible layout, nothing like Amarok 1, SLOW.

I've been thinking to start a new project altogether not based on KDE at all (as in borrow NOTHING from KDE), but definitely based on Qt. Qt is so nice and modular and well-designed, and almost like its own dialect of C++ that doesn't require a high knowledge of all the quirks of the language.

Try the music player Clementine. It's based on KDE 3.5's Amarok 1.4 and I love using it (although its memory footprint is not exactly where I want). The number one thing about it for me is that it doesn't rely upon KDE at all and looks and runs great. And it doesn't suck like Amarok 2.

Qt IMO only needs a lightweight desktop, nothing so fancy as the KDE project has added. Qt 3 of course was a lot smaller, so the footprint of KDE on top wasn't really felt even on older machines. Plus there was no expectation of eye candy whatsoever.

Contact me: https://github.com/tatsh
My latest work is making a toolchain for cross-compiling to OS X. Certainly could try testing that, but unfortunately I haven't yet built Qt on it (the precompiled libs on their own should work however).

Great have your tried LLVM(Clang, Clang++) it should build Qt and if I am not mistaken the soon to be if not already Native compiler suit for Mac. I am trying to move away from GPL code
it's not as free as it claims. My definition of free is you can do whatever you want with the code.

And yah if I can avoid KDE code then I will for the main project but I am also trying to liberate various KDE projects for those that do want them. It's really a lets see what happens kind of thing.


But I think that once this project really gets going everyone will be happy. smile Well I hope lol

Last edited by zester (2011-09-07 17:12:58)

Offline

#68 2011-09-07 17:14:44

diegoxter
Member
From: Valencia, Venezuela
Registered: 2011-02-19
Posts: 11
Website

Re: Quantum Project (KDE4/Qt4 Fork)

zester wrote:

Great have your tried LLVM(Clang, Clang++) it should build Qt and if I am not mistaken the soon to be if not already Native compiler suit for Mac. I am trying to move away from GPL code
it's not as free as it claims. My definition of free is you can do whatever you want with the code.

And yah if I can avoid KDE code then I will for the main project but I am also trying to liberate various KDE projects for those that do want them. It's really a lets see what happens kind of thing.

Just asking: Do you plan on giving some kde-free love to blogilo? Its a great blog app, but it depends a lot on kde... And there arent any other blog tools written in qt these days... (i just know about blogilo and qtm, which didnt worked for me)...


"In brightest days, In darkest nights, no evil shall escape my sight!" Sheldon Cooper cuoting Green Lantern.

Offline

#69 2011-09-07 19:13:15

tatsh
Member
Registered: 2011-09-07
Posts: 3

Re: Quantum Project (KDE4/Qt4 Fork)

zester wrote:
tatsh wrote:

I have used KDE since 3.5.x and I'm on Gentoo.

I am interested in your project. I've done a bit of Qt, but never with KDE dependencies. I've been highly critical of KDE doing things like:
- Akonadi works rarely, and it sucks
- Nepomuk sucks
- The licensing scheme for KDE is not very commercial friendly (all must be GPL) so we're never going to see things like Skype integrate in
- Messing with KDE PIM so that now it REQUIRES mysql and Akonadi to play well
- Kate on OS X lacks some very important features
- Kate is impossible to get on its own (this is true for all KDE apps) and it's my favourite text editor
- Early release issues and regressions therefore (like this bug: https://bugs.kde.org/show_bug.cgi?id=278891 )
- Amarok 2 is HORRIBLE!!!!!! Horrible design, horrible layout, nothing like Amarok 1, SLOW.

I've been thinking to start a new project altogether not based on KDE at all (as in borrow NOTHING from KDE), but definitely based on Qt. Qt is so nice and modular and well-designed, and almost like its own dialect of C++ that doesn't require a high knowledge of all the quirks of the language.

Try the music player Clementine. It's based on KDE 3.5's Amarok 1.4 and I love using it (although its memory footprint is not exactly where I want). The number one thing about it for me is that it doesn't rely upon KDE at all and looks and runs great. And it doesn't suck like Amarok 2.

Qt IMO only needs a lightweight desktop, nothing so fancy as the KDE project has added. Qt 3 of course was a lot smaller, so the footprint of KDE on top wasn't really felt even on older machines. Plus there was no expectation of eye candy whatsoever.

Contact me: https://github.com/tatsh
My latest work is making a toolchain for cross-compiling to OS X. Certainly could try testing that, but unfortunately I haven't yet built Qt on it (the precompiled libs on their own should work however).

Great have your tried LLVM(Clang, Clang++) it should build Qt and if I am not mistaken the soon to be if not already Native compiler suit for Mac. I am trying to move away from GPL code
it's not as free as it claims. My definition of free is you can do whatever you want with the code.

And yah if I can avoid KDE code then I will for the main project but I am also trying to liberate various KDE projects for those that do want them. It's really a lets see what happens kind of thing.


But I think that once this project really gets going everyone will be happy. smile Well I hope lol

Yes, I have llvm-gcc and gcc working (for x86, x86-64, and arm (iPhone)). Have not yet got Clang working.

I've already tried Qt as of today and it can get somewhat far but needs a lot of patches before it would ever cross-compile a typical app properly. The issue is their custom build system, and the tools like uic and moc.

uic, rcc, and moc all need to be compiled by the system gcc so that they would be for the building system (in ./configure, the --build argument). This is because they are platform specific, yet should be able to run on another platform without issue. This is currently not how Qt's make system expects things to be. As a result, my system's moc generates C++ that will not compile because the minimal differences in what moc does between Mac and my machine's moc results in different assembly output (which may not be correct some of the time) from collect2.
The real goal here would be: $CPU-apple-moc, $CPU-apple-darwin11-rcc, $CPU-apple-uic (in my case, compiled for x86_64 Linux but targeting x86_64-apple-darwin11) but I have not got this yet.

The easier way right now is to copy their libs over from OS X and then build moc, rcc, and uic separately. They will probably need patches. The .pro files can specify the paths to these tools so it would be safe.

Simple (especially C-only) libraries like libpng do build fine. However, note that I haven't yet got a multilib compiler. All of them are separate. So if you build libpng, you would only have it for that one target and not all that the compiler would have available (on Mac: i686, x86_64, and possibly armv6 and armv7). To build a multi-arch binary at the moment would require building all libs for that arch as well, then using lipo (which does work on a non-Mac). What you can't do and probably will never be able to: `codesign`. codesign_allocate is available but codesign is closed source for obvious reasons. If you're serious about selling on the Mac app store, buy a Mac then :-P

Offline

#70 2011-09-07 21:41:41

zester
Member
From: Wilkes Barre Pa
Registered: 2011-08-13
Posts: 156
Website

Re: Quantum Project (KDE4/Qt4 Fork)

Finally clicking on desktop icons execute applications as detached processes.

Sooooo lets see what is working now.

Basic Features
--------------------------------
Desktop Wallpaper (Done)
Desktop Icons (Done)
.desktop and mimetype support (Done)
Right Click Desktop and Desktop Icon Menu (Mostly Done)   <--- Have to code the action functions
Executing .desktop Icons as detached processes (Done)
Opening desktop files with the preferred application (Not Done yet)
Desktop Settings (Mostly Done)
Desktop Setting and Desktop Dir Monitoring (Done....But not working right)

Need to strip out some code thats not needed.

Getting really close now smile

Last edited by zester (2011-09-07 21:44:40)

Offline

#71 2011-09-08 00:27:27

zester
Member
From: Wilkes Barre Pa
Registered: 2011-08-13
Posts: 156
Website

Re: Quantum Project (KDE4/Qt4 Fork)

Desktop Directory Monitoring is now working perfectly. So anything added to your Desktop directory
will automatically update's the desktop.

Offline

#72 2011-09-08 16:10:54

zester
Member
From: Wilkes Barre Pa
Registered: 2011-08-13
Posts: 156
Website

Re: Quantum Project (KDE4/Qt4 Fork)

Started working on the Filetype(Mimetype) Association application its a clone of KDE4's version.

While working on this application I looked into how not only KDE handles Mimetype's but also how Gnome and Xfce4
handle Mimetypes I noticed alot of inconsistencies between the different desktop environments.

I also discovered some issues with a third party mimetype parser library I am using for Qt and the desktop, but this
is a good thing because not only am I now aware of the issue I am also aware on how I can eventually
replace this third party mimetype parser with a a better in house design.

The Filetype(Mimetype) Association application is not required but the xml file it produces will be.

There are actually alot of places that mimetype association will and can be used. Particularly for the
package manager and multimedia player.

Once this applications is completed I can finish the desktop. And once the desktop is
completed I will be able to  put together a file manager that is on par with dolphin in a matter of
days.   

oqwz1e.png
Larger View

Offline

#73 2011-09-08 16:22:07

Lothium
Member
Registered: 2009-10-10
Posts: 192

Re: Quantum Project (KDE4/Qt4 Fork)

Its incredible how fast you are! Very good work!

Offline

#74 2011-09-08 16:28:05

zester
Member
From: Wilkes Barre Pa
Registered: 2011-08-13
Posts: 156
Website

Re: Quantum Project (KDE4/Qt4 Fork)

Lothium wrote:

Its incredible how fast you are! Very good work!

Well I have been using Gnu/Linux and programming for 15 years and have worked on both Gnome/GtK and KDE/Qt
so I have a lot of experiences lol but thanks.

Offline

#75 2011-09-08 16:39:43

zester
Member
From: Wilkes Barre Pa
Registered: 2011-08-13
Posts: 156
Website

Re: Quantum Project (KDE4/Qt4 Fork)

Son of a B%^& the city just came by a told me I have to evacuate my house.
http://www.time.com/time/nation/article … 03,00.html

Crapppppppp!!!!!!! This sucks soooo bad.

Well ill take my computer to a friends house maybe ill get to work on this some more if not ill
see you all in a couple of days.

Offline

Board footer

Powered by FluxBB