Hello again,
I have another question for which I didn't find a proper answer:
Is it possible to draw part of the scene with logical operations
enabled?
Something like this
glPushAttrib(GL_ENABLE_BIT);
glEnable(GL_COLOR_LOGIC_OP);
glLogicOp(GL_XOR);
glDisable(GL_LIGHTING);
Hi Dirk,
Could you please outline the task I have to do for writing a special
StateChunk for the logical ops?
More specifically:
- Is there a good example which I can use for guidiance?
- I do have the impression that there is some code generation
machinery at work for building
Hello Carsten and Dirk,
the LogicOpChunk implementation is working like a charm. Thank you for
taking the time.
And if that's the case,
can I use it from within my 1.8
(daily build: source) installation?
just drop the files in OpenSG/Source/System/State and run make/scons
from the
Hi,
I'm looking for a convenient way for line picking. Is there something
like a polytope intersector in OpenSG?
Is it possible to use the standard OpenGL picking mechanism?
Best,
Johannes
-
This SF.net email is sponsored
Hi Dirk,
Dirk Reiners wrote:
Johannes Brunen wrote:
Is there something
like a polytope intersector in OpenSG?
Not a general Polytope, but if you really just want to do picking the
Frustum should be enough.
Yes!
Is it possible to use the standard OpenGL picking mechanism?
In theory yes
Hi Dirk,
Dirk Reiners wrote:
Johannes Brunen wrote:
Below you can find what we are currently doing with our none OpenSG
GL
app. I would like to see something along that line for OpenSG.
However,
currently I can't really estimate all of the implications in the
OpenSG context.
The basic idea
Hi Dirk,
Dirk Reiners wrote:
Well, not really. It was never supported in hardware, so it's just running
through the software emulation layer and is not all that fast. It will also
completely go away in OpenGL 3.
Oh, I didn't know that. Actually, I thought that it was completely done
by the
Hi Dirk,
Sorry, second try, but with proper subject line...
Dirk Reiners wrote:
Well, not really. It was never supported in hardware, so it's just running
through the software emulation layer and is not all that fast. It will also
completely go away in OpenGL 3.
Oh, I didn't know that.
Hi Dirk,
Dirk Reiners wrote:
Next try: Really what you need is object-object intersection, aka
collision detection. As it so happens somebody posted a link to a
pretty
nice collision detection library a few days ago. ;)
Does that sound like a solution?
Sorry, but no. Collision detection
in my code?
Best regards,
Johannes Brunen
-
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems? Stop.
Now Search log events and configuration files using AJAX and a browser
Hi Carsten,
Carsten Neumann wrote:
well, the NewAction stuff was an experiment to reimplement the
traversal infrastructure,
but instead of making things easier it brought a whole new wave of
complexity and thus
is now bitrotting in experimental.
What a pity; I really liked the NewAction
Hi,
may I report a little bug I have found in the assignment operator of
class FrustumVolume (OSGFrustumVolume.cpp, ~360):
At first the implementation fails because of the wrong number of
iterations
const FrustumVolume FrustumVolume::operator =(const FrustumVolume b1)
{
=for(Int32 i = 0; i
Hi,
I'm looking for Python 2.5.1 bindings of OpenSG (1.8 or 2.0) on windows.
I would like to use it for rapid prototyping and testing.
Is there anything usable in existence?
Where can I find the necessary resources?
I have read about pyOpenSG. Is it still supported and in sync with the
OpenSG
Hi,
since I have trouble posting to the list I try a simple one.
Please forgive me the impoliteness.
Greetings
Johannes
-
This SF.net email is sponsored by: Microsoft
Defy all challenges. Microsoft(R) Visual Studio 2008.
Hi,
after some time spending on other task, I have now restarted my
project using the OpenSG library.
As you can see below I have spend some time into the OpenSG framework
and did reach a lot.
But now I'm stuck and need some input from outside.
I have written some new classes in order
Hi,
Sorry, but I send it again so that leading '' are gone.
after some time spending on other task, I have now restarted my project
using the OpenSG library.
As you can see below I have spend some time into the OpenSG framework
and did reach a lot.
But now I'm stuck and need some input from
Hi Carsten,
thank you for your promptly response.
I did found the problem with the frozen windows. It was not a problem
with
the OpenSG framework but with my usage of the windows timer
functionality.
I did implement a tick timer for each window redrawing my scene after
some
period of time.
Hi,
I'm facing the following problem. I would like to use glShadeModel(GL_FLAT) and
glShadeModel(GL_SMOOTH) in my application.
However, I can't find a proper mapping of the glShadeModel command to a OpenSG
primitive.
Would it be a good idea to create a new OSGShadeModelChunk for that?
Hello Carsten,
thank you very much for the immediat help. I do have two follow-ups.
1. I'm not proficient with the patch utility. Can just give me a short
hand for extracting
and/or applying the content of the patch file? For now I just copied
the content
by hand and removed all the '+'
Hi,
I'm using the class PassiveViewport in my application. The function
'activateSize' of PassiveViewport
calls the parent's 'activateSize' function. This is unfortunate because
this function calls
'glScissor(pl, pb, pw, ph)'. My application however, handles the scissor
region independently.
Hi,
today I would like to contribute a new group core node which allows the
user to
select none, some or all children nodes for rendering. Basically, it is
very (very)
similar to the already existing Switch core but with a slightly altered
semantic.
Following I will give a simple usage example:
Hello Carsten,
Carsten Neumann wrote:
I'd be happy to add such a contribution, however I'm wondering if it
would be better to add a MFInt32 _mfChoices; to such a type and allow
selection of an arbitrary number of children ?
The only advantage I see in my solution is simplicity and the saving
Hi,
I have a problem of translating some native OpenGL code into OpenSG
jargon.
Can anyone give me some hints how I should represent the following
OpenGL code into OpenSG primitves? Additionally, am I going to have
problems, if I later on try to do some shadowing of my scene, because of
using
Hello Carsten,
Carsten Neumann wrote:
yes, to create new FieldContainers or modify existing ones you
basically
write or edit a .fcd file and have FCEdit generate the
SomeContainerBase.{h,cpp,inl} and (for a new container)
SomeContainer.{h,cpp,inl} as well. Then you add functions/code to
Hi,
Marcus Lindblom wrote:
You could perhaps add a convenience function allowing one to select a
subset of the first 32 (or 64) nodes using bit values, which
internally
still uses the mfchoices approach?
Done.
Doesn't Node have an getIndexOfChild() or something similar?
Didn't know that.
Hi,
I have a couple of questions about the OpenSG framework and the 2.0
branch especially:
1. Is the semantic part of the light cores implemented? I.e. does a
light core carrying node only influence its childrens graphs? In OpenSG
1.8 lights seems to be always global, which is quite unfortunate.
Hello Carsten,
Carsten Neumann wrote:
Johannes Brunen wrote:
Would it be possible to implement the PassiveViewport without calling
the glScissor function or
with calling it optionally, in order to not break any existing code
relying on the current behavior?
I'm a bit more concerned about
Hello Carsten,
Carsten Neumann wrote:
we are still a bit unsure if we should use a new chunk or put it into
e.g. PolygonChunk. Once that is decided I'll make the same change in
2.0
as well.
Did you have any news on that?
Best,
Johannes
Virus checked by G DATA AntiVirusKit
Hello Stephane,
Stephane Routelous wrote:
for me, type=both works with 1.8, and i'm almost sure I changed
nothing.
I did forget about that. Thanks for reminding...
Best
Johannes
Virus checked by G DATA AntiVirusKit
Version: AVK 19.12 from 21.08.2008
Virus news:
Hello Carsten,
Carsten Neumann wrote:
Johannes Brunen wrote:
In OpenSG 1.8 lights seems to be always global, which is quite
unfortunate.
have you tried RenderAction::setLocalLights(true) ?
No, I didn't know this option. I will try, thanks for pointing.
In 2 there is a ChunkOverrideGroup
Hello Dirk,
Dirk Reiners wrote:
Johannes Brunen wrote:
6. Does anyone have thougth about a LOD node which does perform
database
paging of the branches currently not rendered?
We're in the process of writing something like that specialized for
point clouds, but I'm hoping we can make
Hi,
can anyone tell me how I correctly setup the light model states in
OpenSG 2.0:
GL_LIGHT_MODEL_AMBIENT
GL_LIGHT_MODEL_LOCAL_VIEWER
GL_LIGHT_MODEL_COLOR_CONTROL
Regards,
Johannes
Virus checked by G DATA AntiVirusKit
Version: AVK 19.106 from 25.08.2008
Virus news:
Hi,
I'm on the way of testing the OpenSG 2.0 framework. Currently, I'm
building the necessary dependency tree for OpenSG 2.0.
However, I'm uncertain which collada version I have to compile and to
install. Did anyone has successfully compiled and/or installed collada
on the windows platform?
Hi,
today I have to report a problem I encountered on storing a scene into
the 'bin' file format and subsequentially reading it two times. Below
you can find a very simple program showing the problem. Running this
program crashes on my vc-7.1 windows box. Actually, the problem has
something to do
Hi Andreas,
It must be _SCL_SECURE=0. This is a spelling error.
hmm don't believe this google gave me 3540 hits for _SECURE_SCL and
only
100 for _SCL_SECURE
Ok, you are absolutely right. I did it wrong on all my libs. Pardon for
bothering...
well optdbg would be a better name for this
Hello Carsten,
today I reviewed the code you checked into the trunk with respect to the
multiswitch core. You did change the MultiSwitch::changed() implementation,
but I do not understand what was wrong with my initial proposal. Actually,
I'm more fond of my original expression.
Additionally,
Hello Carsten,
Carsten Neumann wrote:
Johannes Brunen wrote:
did you do this for convenience or did it not work otherwise. Using
python Tools/scons-local/scons.py to invoke scons worked fine for me
(python has to be in the PATH, of course).
No, only for convenience. Both files just map
Hello Carsten,
Johannes Brunen wrote:
Carsten Neumann wrote:
Johannes Brunen wrote:
The first problem I did encounter was that the boost filesystem library
could not be found. After some investigations I came to the conclusion
that
there is an error in the scons implementation in file
Hello Gerrit and Marcus,
Gerrit Voss wrote:
I haven't had time to look through all the details, but I'm sure
2.x with VS2008 builds on my xp system (without changes and minus
collada).
I can't confirm that. With msdev .net 2008 sp1 little tweaking of the scons
system is necessary.
Maybe, it
Hello Gerrit,
I have tried on two different machines. On the first one I have vc7.1 and
vc9. On the second one I have only vc9 active, but a properly deinstalled
vc7.1.
Best,
Johannnes
-
This SF.Net email is sponsored
Hi,
my 2 cents...
Go with a recent boost version and adapt for more exotic platforms.
Best,
Johannes
-
This SF.Net email is sponsored by the Moblin Your Move Developer's challenge
Build the coolest Linux based
Hello Andreas,
Andreas Halm [EMAIL PROTECTED] schrieb im
Newsbeitrag news:[EMAIL PROTECTED]
However, I asked myself, as boost is a prerequisite to OpenSG and
compiling
that is very easy, why OpenSG is not using jam, too. However, I really
don't
know anything about build systems, this is
Hello Carsten,
may I send you an example of the clip plane capping problem as a
contribution to the OpenSG tutorials. Probably it needs some clean up but I
think it is worth the effort. Could you please take a look. Comments are
welcome. Currently it is only OpenSG 1.8. I did not have had time
Hello Carsten,
there are some quirks however. Attached you can find a slightly modified
version which allows movement of two clip planes by keystrokes.
keys:
a = enable/disable clip plane 0
s = enable/disable clip plane 1
y = move clip plane 0 on positive normal direction
x = move clip plane 0
Hello Carsten,
you are right. I did overlook that. Unfortunately, I have still problems now
on moving the geometry. I did change the example a little so that I'm i)
able to create two objects and ii) to move them around. Possibly, that
problem is of similar nature as the former one.
If you
Hello Carsten,
I did check my mails. The last mail from the opensg user list does contain
my attachment and I could open it without any problem. So the problem might
be found on your side.(?)
Perhaps some window/unix conversion thing.
Best regards,
Johannes
Hello Carsten,
now I have understood what was wrong.
How can I force rendering to draw the geometries independently from each
other? I.e. to draw each branch with the proper stencil settings independent
of the settings of other geometry branches?
Doing this should, imho resolve the
Hello Carsten,
hm, I know Gerrit has most/all his projects running off 2.0 and the
things we are working on over here also use it. In general my impression
is that it works decent and the increased strictness wrt. to reference
counts and object lifetime have turned up some nasty bugs in
Hi,
today I ask for some guidance with the following problem:
I have written a pick framework which allows fast frustum picking of polygons,
lines and points. In principle this works fine until recently. Problems show up
the moment I have introduced the first 'InverseTransform' core into the
Hi,
Of course, I could write different handler functions for each
class, like 'enterAccumulateInverseTransform', but that is
cumbersome and does not scale very well.
No this is also not possible. 'InverseTransform::accumulateMatrix' is also
protected. So, I can't build my
Hi,
So if there is not another way, please make the functionality public.
In particular...
System/FieldContainer/OSGNodeCore.h
System/NodeCores/Groups/Misc/OSGBillboard.h
System/NodeCores/Groups/Misc/OSGInverseTransform.h
System/NodeCores/Groups/Misc/OSGScreenGroup.h
Best,
Johannes
Hello Carsten,
I'll make accumulateMatrix public, but will keep adjustVolume protected
unless you have a pressing need for it (the code you posted did not
attempt to use it?).
Yes, I only use function accumulateMatrix. I was a little bit fast to assume
a future need of adjustVolume.
Thanks
Hello Carsten,
I have found the following problem on compiling a fresh CVS OpenSG 1.8
on windows:
Source\System\NodeCores\Groups\Misc\OSGMultiSwitch.cpp : error C4335:
Mac file format detected: please convert the source file to either DOS
or UNIX format
If have solved the problem locally with
Hello Carsten,
I'm not sure if OpenSG 2.0 is also affected.
yes, it has a similar problem, but the TextureChunk is deprecated there
and replaced with TextureObjChunk and TextureEnvChunk. Anyways I'll
change the SFImage to be a child pointer field, then it should handle
this automagically.
Hello Dirk,
I would also go for a clear self explaining naming scheme.
However, I'm a little puzzeled about the variants you are going to build on
default. Normally, I use the following variants:
DEBUG- /MDd /Od
DEBUG_OPTIMIZED- /MDd /O2
RELEASE- /MD /O2
This gives me the
Hello Marcus,
Ok, I understand. But then it is desirable that the DEBUG_OPTIMIZED variant
is added too. This one is typically used (at least by me) for application
development.
Regards,
Johannes
Marcus Lindblom ma...@yar.nu schrieb im
Newsbeitrag news:49de04e5.3080...@yar.nu...
Johannes
Hello,
short question. Did you specify the /Oi compiler option for the usage of
intrinsics on all targets correctly?
Best
Johannes
--
Stay on top of everything new and different, both inside and
around Java (TM)
Hello Carsten,
Carsten Neumann carsten_neum...@gmx.net schrieb im
Newsbeitrag news:4a3e7c0a.9030...@gmx.net...
first off I believe the fact that it prints the right versions is likely
a red herring, I don't think cmake 2.6.4 has support for boost 1.39,
i.e. it is not on the list of versions
Hello Gerrit,
Gerrit Voß vo...@vossg.org schrieb im
Newsbeitrag
On Thu, 2009-06-25 at 08:37 +0200, Johannes Brunen wrote:
Debug - /MDd /Od as usually found on windows: suffix 'D'
DebugOpt - /MDd /O2 /Ob2 this one is perfect for application
development with consistent debug runtime
Hello Gerrit,
Debug - (short for DebugDbg) debug runtime + debug compiler
DebugRel - debug runtime + release compiler (was DebugOpt)
ReleaseDbg - release runtime + debug compiler
Release- (short for ReleaseRel) release runtime + release compiler
the lib suffixes than would be:
Hello Gerrit,
Gerrit Voß vo...@vossg.org schrieb im
Newsbeitrag news:1245983834.3311.65.ca...@feynman.camtech.ntu.edu.sg...
As I said I really don't have a preference as long as it is consistent
in some form. So your solution would be:
Debug- (short for DebugNoOpt) debug runtimr +
Hello Carsten,
Carsten Neumann carsten_neum...@gmx.net schrieb im
Newsbeitrag news:49e74194.3000...@gmx.net...
Here are some of my questions:
1. How can I adapt the build process with respect to the following
parts:
- special compiler flags: I build all my libraries with the
follwing
Hello Gerrit,
Gerrit Voß vo...@vossg.org schrieb im
Newsbeitrag news:1246004014.13150.14.ca...@feynman.camtech.ntu.edu.sg...
hmm, on windows cmake comes with a user interface to change the settings
directly at least for me start-programs-cmake-cmake brings up one.
Is it a problem to
Hello Gerrit,
Gerrit Voß vo...@vossg.org schrieb im
Newsbeitrag news:1246007112.13150.31.ca...@feynman.camtech.ntu.edu.sg...
based libraries and would actually find a generic solution.
well if you use cygwin already some sed or awk should work to
postprocess the cmakecache file and than rerun
Hello Gerrit,
Gerrit Voß vo...@vossg.org schrieb im
Newsbeitrag news:1246422106.3445.21.ca...@feynman.camtech.ntu.edu.sg...
ok, I did the changes, could you check if they make sense for you.
Small warning of which you most likely are aware, you have to start
with a blank/empty cache to get
Hello Lars,
Lars Bilke lars.bi...@ufz.de schrieb im
Newsbeitrag news:4a4b198b.1060...@ufz.de...
Johannes,
it would be really cool to provide all these libraries for everyone. Is
it possible to upload them to the OpenSG wiki or at least all the
binaries to the subversion repository?
If there
Hi Gerrit,
I did send a message with an attachment. I got the following answer:
Your message to Opensg-users awaits moderator approval
Will the message be approved or should I resend it (maybe privately)?
Best,
Johannes
Hi Marcus,
Marcus Lindblom ma...@yar.nu schrieb im
Newsbeitrag news:h2hp5u$gs...@ger.gmane.org...
We/I need to reduce the #include-hogging for that.
I didn't understand that!?
Note that for third-party libs, we don't really need all those variants
there. Just something built against
Sorry Gerrit,
but could you provide me than you e-mail address?
Best,
Johannes
--
___
Opensg-users mailing list
Opensg-users@lists.sourceforge.net
Hello Gerrit,
I'm still thinking of how to do that most efficiently on windows.
Unfortunately it is such an empty system, and I don't like to make
cygwin a prerequisite. One way is to go the VTK way and include
all the sources needed, which is ok for the small stuff like
image libs but with
Hello Gerrit,
Gerrit Voß vo...@vossg.org schrieb im
Newsbeitrag news:1246609102.3442.40.ca...@feynman.camtech.ntu.edu.sg...
hmm, fascinating I'll have a look. I thought I tested it but it might be
that I screwed something in between.
Be aware that I could only take a tiny glance at the
Hello Georg,
OpenSG 1.8 does set the locale in osgInit(). I did solve problems with
respect to the locale settings by resetting them 'correctly' after
osgInit().
IMHO OpenSG should not set locale at all. It is the business of the
application not of a library component.
Only my 2 cents.
Best,
Hello Gerrit and Georg,
short question what is the correct setting so that floats always
go through any kind of ascii IO as 1.34234 ? I don't really want to deal
with cases where files can't be read because floats are expected to be
1,2343 or written as 1,24143.
We use the following settings
Hi Dominik,
Dominik Rau dominik_...@gmx.de schrieb im
Newsbeitrag news:4a4f572f.9050...@gmx.de...
I discovered this issue in OSG::SimpleSHLChunk.inl
...
I think getValue() != NULL would be a better choice.
Yes.
There are at least three other files which contain the same problematic
line:
Hello,
currently I'm trying to switch to the OpenSG 2.0 codebase. In this
context I have a couple of questions.
1. Can I safely assume the support of a boolean context from the XRecPtr
smart pointer types, i.e. can I write something like the following?
NodeRecPtr node = ...
if
Hello,
I have the following problem. Currently I'm porting an application from
OpenSG 1.8 to OpenSG 2.0. Simuntanously, I'm bringing it into a 64Bit
compatible state.
Now I have encountered the case, that I have once written an Attachement
derived class which just holds an multi field of type
Hi,
I'm still not completly sure about the correct smart pointer usage.
Please allow me to outline a usage scenario I found in my code:
I have a home grown scene manager class which does manage the scene.
class scene_manager
{
public:
Node* getRoot() { return _node; }
void
Hi,
I have found a little difference between 1.8 and 2.0 which is
problematic. I'm using PassiveWindow and have implemented a custom scene
manager which basically calls the PassiveWindow 'render' method. Now,
this one is not implemented in PassiveWindow and the base implementation
of Window is
Hello Carsten,
Thank you, actually I do not monitor the commits. How can I do that? I would
be interested.
Best,
Johannes
--
Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day
trial.
.
With best regards
Johannes Brunen
P.S.: I hope that the model image is more or less self explaining.
Virus checked by G DATA AntiVirusKit
Version: AVF 19.509 from 25.08.2009
Virus news: www.antiviruslab.com
--
Let
Hello Carsten,
Carsten Neumann carsten_neum...@gmx.net schrieb im
Newsbeitrag news:4a954859.5090...@gmx.net...
In OpenSG 1.8 I first used the StateSortingGroup class. After some
experimenting I discovered that the usage of this node is not necessary
at all.
hm? why not, i.e. what does
Hello Carsten,
I have some additional questions about the GroupingStage class.
I'm a little bit confused about the core aspect of the GroupingStage. Is it
kind of a core at all and especially is it kind of a grouping core. I.e. am
I allowed to add multiple childrens to a node carrying this
Hello Carsten,
I'm still not able to get correct cap rendering in case of multiple defined
clipplanes. For that I did write my scene into an OSB file (test3.osb) and
loaded it with the 10loading.exe example. I couldn't see any geometry. Ok, I
thought it might be better to start with a simpler
Hello Carsten,
I did some more investigations and found that I have had an error in my
SceneManager.cpp file. I did not store the clip plane beacon node
separately. I thought that the setBeacon call would deliver ownership to the
clip plane chunk. But this is not the case. I have to record the
Hello Gerrit,
could you please check if the following line in CMake/UpdateCompiler.cmake
ADD_DEFINITIONS(-DCRT_SECURE_NO_DEPRECATE)
is correct. I think that is should be
ADD_DEFINITIONS(-D_CRT_SECURE_NO_DEPRECATE)
Best,
Johannes
Gerrit Voß vo...@vossg.org schrieb im
Newsbeitrag
Hello Carsten,
Carsten Neumann carsten_neum...@gmx.net schrieb im
Newsbeitrag news:20090831171934.68...@gmx.net...
sorry for the delay in getting back to you on this.
No problem...
Ok, IIUC you've solved the problem that your beacon nodes
were lost by keeping pointers to them in
Hello Carsten,
I did change the clipplanecaps.cpp example:
- move the stage node down the graph
- added container collection attachment
- added key event for file storing
- added some preloadSharedObject statements
- minor docu changes
I did upload the file to
Hello,
I have a problem with the conversion of an OSB file saved with OpenSG
1.8 and loaded with OpenSG 2.0. The file contains a ChunkMaterial with a
TextureChunk applied to it. After reading in OpenSG 2.0 I only found a
TextureObjChunk but no TextureEnvChunk. I looked into the
Hello Gerrit,
I do have still a problem. Now I have two TextureEnvChunks in my scene graph
according to the two model entities of my root.osb model. Unfortunately, I
do see both geometries rendered with the texture environment mode
GL_REPLACE. However, the two TextureEnvChunks do carry
Hi Gerrit,
After further looking into the file OSGOSBChunkMaterialElement.cpp I think
that the problem comes from the line
setContainer(ChunkMaterialUnrecPtr(ChunkMaterial::create()));
of the read function in this file. Here the actual type of the element is
lost. Later on in the code the
Hi,
I'm facing an additional small problem with the CMake build system with
respect to the specific image file types. Let me start by saying that
I'm not using the support libs OpenSG provides. After some
instrumentation I got the following output on running cmake (only JPEG
on the compass)
--
Hello Gerrit and Carsten,
sorry but I will be out of business until Monday morning. Therefore I can't
answer your questions for now. A nive weekend with some friend in the German
Eifel is lying ahead of me.
Best,
Johannes
Hello Gerrit and Carsten,
I'm back and have tried to get answers to all of your questions. So here it
goes...
it should be set in build_dir/Source/Base/Base/OSGConfigured.h which
is included by OSGConfig.h which in turn should be included be the
jpeg file handler before checking for
Hello Gerrit,
I do have a new piece of information:
The call
_pHandle = LoadLibrary(libName);
does fail for the OSGImageFileIO dynamic link library, i.e. returns NULL
Do you know a good code place where I should further debug. The library is
actually in the search path for the loader. As
Hello Gerrit,
Gerrit Voß vo...@vossg.org schrieb im
Newsbeitrag news:1252318865.3203.24.ca...@feynman.camtech.ntu.edu.sg...
hmm, it could also mean that one of the dependent libs is not found.
...
Another thing to try would be depends.exe
which shows which libraries/symbols are unresolved.
Hello Gerrit,
The OpenExr library still does cause problems for me. After spotting the
dynamic link library problem I did switch to static linking but still uses
my own build system setup. The only notably difference of the support
libraries provided by the OpenSG setup is the naming scheme.
Hello Gerrit,
Gerrit Voß vo...@vossg.org schrieb im
Newsbeitrag news:1252569216.3192.20.ca...@feynman.camtech.ntu.edu.sg...
bad news, looks like I shredded some of my .net installation and now
nearly nothing, in particular VS, works properly.
Sorry for that...
So it might take
a second
Hello Gerrit,
Gerrit Voß vo...@vossg.org schrieb im
Newsbeitrag news:1251888126.3345.8.ca...@feynman.camtech.ntu.edu.sg...
catches all Materials but the OSBChunkMaterialElement only registers
for ChunkMaterial itself, not for SimpleMaterial which you seem to use.
So the whole thing should
Hello Gerrit,
I did take a preview of the correction you have committed to svn.
// No chunks loaded
if(_chunksPtrFieldIt == PtrFieldListIt())
return;
This one does unfortunately not work for the Visual Studio platform at least
not for the case that the
_HAS_ITERATOR_DEBUGGING
Hello Gerrit and Marcus,
Just to chip in: I think not. A default-initialized iterator is not
valid for comparison with anything at all. You aren't allowed to do
anything with it except overwrite it.
flawed hmm, anyway I just got lured in by the behavior that was in there
since the early
1 - 100 of 295 matches
Mail list logo
