Source is public

[GoogleCodeExporter]

Hey guys,

Ok, I've put the source into the SVN version control on here.

This was done in hopes that other coders will help out with the project and 
submit their improvements, so that the rest of the world can benefit from them 
as well.

I'll keep working on this in my spare time, but help is always nice :)

Any questions or comments, just let me know

Original issue reported on code.google.com by gareth.p...@gmail.com on 28 Nov 2011 at 9:34

[GoogleCodeExporter]

Hey,
Just got to looking at the code placed in the svn.  Is that the most up to date 
working code?  I am asking because when i went to compile the code I have many 
errors, mostly wrong identifier errors.  I did have to include the 
SkeletalViewer.vcxproj files from the Unmanaged code examples of the kinect SDK 
because the Kari solution was complaining without them, this may be the problem 
as well.  So my primary question is that if the current errors, approx 200, 
should be there.  If so I can go through and help sort and fix these, but if 
not I'll know that I should possibly redownload the SDK and I'll know the error 
is on my side.

Original comment by GoldsZi...@gmail.com on 29 Nov 2011 at 9:05

[GoogleCodeExporter]

I think I know what it is... 
I modified the Kinect SDK source so that the names of the constants made more 
sense. Their naming scheme has a lot of indescript mentality to it. Calling 
"joints" a "skeleton", but then also calling skeletons, skeletons. It got very 
annoying, so I changed things up :)

I'll commit those files into a sub-folder when I get home today.

Original comment by gareth.p...@gmail.com on 29 Nov 2011 at 9:23

[GoogleCodeExporter]

Sounds effective.  I'll take a look at the code again once it gets updated.  
Thanks again.

Original comment by GoldsZi...@gmail.com on 29 Nov 2011 at 9:44

[GoogleCodeExporter]

Ok I added the KinectBeta2 inc folder to svn, it only looks like I had done 
edits to the MSR_NuiSkeleton.h file, but I figured I might as well just commit 
them all..

for me these files are located at:
D:\Program Files\Microsoft SDKs\Kinect\v1.0 Beta2\inc

hopefully that'll fix your problems, if not, let me know and post the errors 
log.
There are some errors for me on compile, but just minor var value type notices.

Original comment by gareth.p...@gmail.com on 30 Nov 2011 at 5:31

[GoogleCodeExporter]

[deleted comment]

[GoogleCodeExporter]

For some reason the m_Log is giving me errors when it takes in even strings. I 
am wondering if you are having this problem as well?  I will take another look 
at code later and see if I can fix it.  Just wondering if these are the errors 
you have.

My Current list of Errors are:

Error   28  error C2679: binary '<<' : no operator found which takes a right-hand 
operand of type 'std::basic_string<_Elem,_Traits,_Ax>' (or there is no 
acceptable 
conversion) c:\users\ziggy\downloads\kinect-activity-reaction-interface\trunk\ka
ri\nuiimpl.cpp  1883    1   SkeletalViewer

Error   43  error C2679: binary '<<' : no operator found which takes a right-hand 
operand of type 'std::basic_string<_Elem,_Traits,_Ax>' (or there is no 
acceptable 
conversion) c:\users\ziggy\downloads\kinect-activity-reaction-interface\trunk\ka
ri\nuiimpl.cpp  2314    1   SkeletalViewer

Error   44  error C2679: binary '<<' : no operator found which takes a right-hand 
operand of type 'std::basic_string<_Elem,_Traits,_Ax>' (or there is no 
acceptable 
conversion) c:\users\ziggy\downloads\kinect-activity-reaction-interface\trunk\ka
ri\nuiimpl.cpp  2400    1   SkeletalViewer

Error   4   error C2679: binary '<<' : no operator found which takes a right-hand 
operand of type 'std::string' (or there is no acceptable 
conversion) c:\users\ziggy\downloads\kinect-activity-reaction-interface\trunk\ka
ri\nuiimpl.cpp  470 1   SkeletalViewer

Error   5   error C2679: binary '<<' : no operator found which takes a right-hand 
operand of type 'std::string' (or there is no acceptable 
conversion) c:\users\ziggy\downloads\kinect-activity-reaction-interface\trunk\ka
ri\nuiimpl.cpp  680 1   SkeletalViewer

Original comment by GoldsZi...@gmail.com on 30 Nov 2011 at 3:43

[GoogleCodeExporter]

That... is an encoding issue... I think setting your project to multibyte fixes 
those

Original comment by gareth.p...@gmail.com on 30 Nov 2011 at 5:20

[GoogleCodeExporter]

Hey,
Sorry for not responding in a while I have been very busy...  I did get 
everything to work find and was wondering what is it you are trying to add now? 
 I also would like to recommend uploading the version that this source code is 
using.  Seems to be a nice release and also works using Kinect SDK beta 2.  

Original comment by GoldsZi...@gmail.com on 7 Dec 2011 at 9:54

[GoogleCodeExporter]

oh cool, glad it's finally working for you :)
Yeah I'll upload that binary.. it's not much diff from the previous one, other 
than some speed refinements, and it's using beta2 so it's faster :)

The next feature I was trying to add was to change the execution of actions to 
be their own threads, so that it could handle more actions being executed in 
reaction to the person, and also so that actions could be given a long sequence 
of events with time delays, etc. All my attempts to get this properly working 
were thwarted...

Original comment by gareth.p...@gmail.com on 7 Dec 2011 at 10:00

[GoogleCodeExporter]

Hey while I was looking over the new release I noticed a bug and you may want 
to hold back on binary release.  I do not believe leftFootForward or 
leftFootBackward is working properly.  Have yet to look into it much but when I 
ran a test config with either option set to distanceMin = 6 for example it does 
not register and starts tracking leftLean.  I have yet to look into the code 
for it is late I have just found but I will look into it tomorrow  and see if 
it was from my possible tinkering...... And for the multithreading do you want 
each bodyaction to inherently run in its own thread or make 1 thread per [set] 
in the users config file

Original comment by GoldsZi...@gmail.com on 9 Dec 2011 at 5:16

[GoogleCodeExporter]

Ok I'll hold off on it... I did a lot of changes to it while trying to get the 
threading going, and didn't really 100% confirm the reverting of them was 
perfect :)

I was thinking 1 thread per [set]
so that a a list of [execute]'s could run.
Something like...

[set]
[require]
bodyAction = handLeftForward
[execute]
keyPress = a
[execute]
delay = 100ms
[execute]
keyRelease = a
KeyPress = b
[execute]
delay = 200ms
[execute]
keyRelease = b

that sort of thing, without blocking other actions from being able to run 
during it's long sequence

Original comment by gareth.p...@gmail.com on 9 Dec 2011 at 5:23