GetDEnabling / GetDDescription in python / c++

THE POST BELOW IS MORE THAN 5 YEARS OLD. RELATED SUPPORT INFORMATION MIGHT BE OUTDATED OR DEPRECATED

On 17/03/2012 at 09:10, xxxxxxxx wrote:

hi,

i was able to write the description states with itemdesc provided by GetDEnabling. i am not quite
sure, but this is *unwanted* useage ?

def GetDEnabling(self, node, id, t_data, flags, itemdesc) :
	itemdesc[c4d.DESC_HIDE] = True;
	return False

i should use SetBool() instead of this ? i am not quite sure if it is intended that i am messing with
itemdesc. also some other things still confuse me.

1. is intended that self is missing in the parameter list for GetDEnabling ? why is it listed for other 
overwritable methods ? when i overwrite it without the self parameter the console says "meh, got
5 parameters , but i want 6, you suck !"
2. the descID id is actually a descLevel for me.
3. t_data is alway none for me
4. and finally (doooh :D) i do not really understand where and why i should do this :

Then make sure to include a call to the parent at the end:

return NodeData.GetDEnabling(node, id, t_data, flags, itemdesc)

when i return it at the end of GetDEnabling instead of True or False for the elements enable/
disable state (which i could set with itemdesc) this would lead to an endless recursion.

thanks for your help.

happy rendering,
ferdinand

THE POST BELOW IS MORE THAN 5 YEARS OLD. RELATED SUPPORT INFORMATION MIGHT BE OUTDATED OR DEPRECATED

On 17/03/2012 at 09:22, xxxxxxxx wrote:

ok ,

is it meant this way : i should get the parent of my nodedata, and return it this way

return ParentNodeData.GetDENabling(node, id, t_data, flags, itemdesc)

however this also doesn't really make sense for me oO

THE POST BELOW IS MORE THAN 5 YEARS OLD. RELATED SUPPORT INFORMATION MIGHT BE OUTDATED OR DEPRECATED

On 17/03/2012 at 10:32, xxxxxxxx wrote:

  1. No, that's an issue of the docs. For methods, there is always a self as first parameter (as long as their non-static/classmethods)
  2. Sure? Gonna try that out. Would be an issue of the docs again if so.
  3. Don't actually know what value it is intended to be. I'm not getting much information out of the C++ SDK, too. It says >Originally posted by xxxxxxxx

Just read the passed t_data if the right id was provided, and return TRUE to enable the parameter or FALSE to disable it depending on the value. Then make sure to include a call to the parent at the end:
     return SUPER::GetDEnabling(node, id, t_data, flags, itemdesc);    _<_h5_>_Re_<_h5_>_/h5>
> Bool
>
>> TRUE if the parameter should be enabled, otherwise FALSE. It's is recommended that you include a call to the parent function as your last return.

How should I return True or False when I need to return the value returned by the parent-call?
4. >Then make sure to include a call to the parent at the end:

   return base.GetDEnabling(node, id, t_data, flags, itemdesc)  

That means you should call the parent's implementation of that method. The example line of code here is not the one of the yellow, as it does not correctly demonstrate how to do it. You can call a base-classes method by doing either

    def GetDEnabling(self, node, id, t_data, flags, itemdesc) :
    # ...
    return NodeData.GetDEnabling(self, node, id, t_data, flags, itemdesc)

or

    def GetDEnabling(self, node, id, t_data, flags, itemdesc) :
    # ...
    return super(MyNodeDataSubClass, self).GetDEnabling(node, id, t_data, flags, itemdesc)

both ways are equivalent (but the latter only applies for subclasses of the built-in object class (every class in Py4D inherits from this class, so it works. :) ).

Cheers,
-Niklas

THE POST BELOW IS MORE THAN 5 YEARS OLD. RELATED SUPPORT INFORMATION MIGHT BE OUTDATED OR DEPRECATED

On 17/03/2012 at 11:00, xxxxxxxx wrote:

2. i printed the content of the parameters to the console and id always contained something 
like this  _[MyIDRange,LargeNumber,MyPluginID].  _plus _ _ typeof returned descLevel.

4. sorry, but i still don't really get this. i understood from the documentation the placement of 
the call but i don't understand how it works in conjunction with the construction of GetDEnabling.
GetDEnabling returns True|False for the state of the passed description id. when i return the
parents GetDEnabling instead it will cascade through my object hierachy but it will never set the
state of my ids. to my understanding there can always be just one return statement in a function 
call as works similiar to the break statement.

Enabled/disabled could be set with the desciption container, but then the naming of the method 
and its description 
_<_t_<__<_t_<__<_t_>_Returns:|
True  if the parameter should be enabled, otherwise  **False _g>.

---|---
is quite misleading when the concept is to call an update the cycle as the exit statement. i can't 
get rid of the feeling that i didn't understood a basic concept here.

thanks for your great help,
ferdinand

THE POST BELOW IS MORE THAN 5 YEARS OLD. RELATED SUPPORT INFORMATION MIGHT BE OUTDATED OR DEPRECATED

On 17/03/2012 at 11:05, xxxxxxxx wrote:

As you can see in point 3, I mentioned I don't understand the C++ SDK at this point, too. ^^ I don't have any Idea how I should return True or False when I should actually return the value of the parent's method. That is logically not possible. 😵

Anyway, have you figured out any problems when not calling the parent's method and just return True or False just as you want to? I don't have time to mess around with that, sorry. :) A statement from the support would be nice.. 😠

Cheers,
-Niklas

THE POST BELOW IS MORE THAN 5 YEARS OLD. RELATED SUPPORT INFORMATION MIGHT BE OUTDATED OR DEPRECATED

On 17/03/2012 at 11:05, xxxxxxxx wrote:

t_data is representing a GeData type.
It's a variable that's used to hold data.
You can think of it as being very similar to the variable bc when you create a base container.

Get&SetParameter() methods use this GeData type as the variable that holds the parameters you find with them. Instead of the bc variable you're more used to using with base containers.

-ScottA

THE POST BELOW IS MORE THAN 5 YEARS OLD. RELATED SUPPORT INFORMATION MIGHT BE OUTDATED OR DEPRECATED

On 17/03/2012 at 12:12, xxxxxxxx wrote:

Originally posted by xxxxxxxx

As you can see in point 3, I mentioned I don't understand the C++ SDK at this point, too. ^^ I don't have any Idea how I should return True or False when I should actually return the value of the parent's method. That is logically not possible. 😵

Anyway, have you figured out any problems when not calling the parent's method and just return True or False just as you want to? I don't have time to mess around with that, sorry. :) A statement from the support would be nice.. 😠

Cheers,
-Niklas

it is working but it behaves a bit oddly sometimes. when a descriptions state is changed you 
have to acutally click or hoover something until the changes are reflected in the gui. could be 
the result of  the missing parent call,  or i have  just to use some sort of update call after i have
changed itemdesc. haven't  tested  it much yet, i have moved on to the actual plugin code 
already, as  the gui finally is at some sort of useable state.

as a side note : i know maxon bashing is generally not very appreciated ** ** in the com but the 
numerous mistakes i have already found in the documenation after ~3 month of more serious
python script and python node usage plus 2 weeks of actual python plugin usage are really a 
shame. you have literally just to poke somewhere randomly into the documenation to find 
dozens of mistakes. i can understand when some parts of the api calls aren't working due to 
the  complex system of c4d, but mixing up descriptions of get and set methods, wrong method 
names and much more is simply sloppy.
also leaving the actual support onto some users here doesn't shed a good light onto maxon.

well, enough of pointless raging, i can just thank you both for your help again and say,

happy rendering,
ferdinand

THE POST BELOW IS MORE THAN 5 YEARS OLD. RELATED SUPPORT INFORMATION MIGHT BE OUTDATED OR DEPRECATED

On 17/03/2012 at 13:00, xxxxxxxx wrote:

I don't know if this matters or not. But when I do this sort of thing in C++. I don't do it in the GetDEnabling() method.
I was under the impression that method was used for greying out things. So the user can't execute them.

I see most people using the GetDDescription() method for dynamically hiding GUI's that are using descriptions. Like this:

Bool SimpleTag::GetDDescription(GeListNode *node, Description *description,DESCFLAGS_DESC &flags)  
{      
  BaseContainer *data, *bc;  //Create two empty containers  
  const DescID *singleid;    //Create a DescID variable      
  DescID cid;                //Create another DescID variable   
  
  if(!description->LoadDescription(node->GetType()))  
      return FALSE;  
  
  data = ((BaseList2D* )node)->GetDataInstance();//Get the container for the tag  
  singleid = description->GetSingleDescID();    // <---- undocumented function call  
  
  
  
  //Check the state of the MYBOX checkbox GUI item and show or hide the slider GUI item as required  
  
  cid = DescLevel(MYSLIDER, DTYPE_REAL, 0);        //Assign the slider GUI item to this description variable      
  if(!singleid || cid.IsPartOf(*singleid, NULL))  
  {  
      bc = description->GetParameterI(cid, NULL);   //Assign the slider to this DescID variable  
      bc->SetBool(DESC_HIDE, data->GetBool(MYBOX)); //Hide the slider GUI if GetBool is true(Checkbox is enabled)  
  }  
  
  
  
  flags |= DESCFLAGS_DESC_LOADED;  
  return TRUE;  
}

I have no idea how to do this in python yet. And  I would love to see an example if you get it all working.

-ScottA

THE POST BELOW IS MORE THAN 5 YEARS OLD. RELATED SUPPORT INFORMATION MIGHT BE OUTDATED OR DEPRECATED

On 17/03/2012 at 14:13, xxxxxxxx wrote:

Originally posted by xxxxxxxx

I don't know if this matters or not. But when I do this sort of thing in C++. I don't do it in the GetDEnabling() method.
I was under the impression that method was used for greying out things. So the user can't execute them.

it is, to my understanding now GetDEnabling is meant to be used this way in python :
when you want do enable/disable an element, then you use it whith a boolen as return 
value (and you don't have to bother about the correct updates of the gui). but as soon as i 
start to manipulate the itemdesc parameter i get this odd update behaviour until i use this 
recursive parent call thing. can't say this is rock solid, because im still quite uncertain with python
and the api, but und it seems to be working for me now. will post a snippet later when i am sure 
that i am not talking bulls**t 😂.

THE POST BELOW IS MORE THAN 5 YEARS OLD. RELATED SUPPORT INFORMATION MIGHT BE OUTDATED OR DEPRECATED

On 19/03/2012 at 14:33, xxxxxxxx wrote:

two things,

1. itemdesc seems to be read only, overwriting it causes major gui problems and it doesn't seem 
to be the desciption of the current instance of your class, as changes to it effect all active instances
of your plugin. (also instances that are created after the changes).

2. i found an example for the intended use of the return value for  GetDEnabling  (at least for c++) 
maxon also just returns true or false.

Bool DoubleCircleData::GetDEnabling(GeListNode *node, const DescID &id,const GeData &t_data,DESCFLAGS_ENABLE flags,const BaseContainer *itemdesc)
{
	LONG inter;
	BaseContainer *data = ((BaseObject* )node)->GetDataInstance();
	if (!data) return FALSE;
  
	switch (id[0].id)
	{
		case SPLINEOBJECT_SUB:		
			inter=data->GetLong(SPLINEOBJECT_INTERPOLATION);
			return inter==SPLINEOBJECT_INTERPOLATION_NATURAL || inter==SPLINEOBJECT_INTERPOLATION_UNIFORM;
  
		case SPLINEOBJECT_ANGLE:	
			inter = data->GetLong(SPLINEOBJECT_INTERPOLATION);
			return inter==SPLINEOBJECT_INTERPOLATION_ADAPTIVE || inter==SPLINEOBJECT_INTERPOLATION_SUBDIV;
  
		case SPLINEOBJECT_MAXIMUMLENGTH:	
			return data->GetLong(SPLINEOBJECT_INTERPOLATION)==SPLINEOBJECT_INTERPOLATION_SUBDIV;
	}
	return TRUE;
}

nevertheless, i would really like to hear a word from maxon on the nature of the itemdesc 
parameter and if there is a way to hide description elements per instance dynamicly with 
python.

THE POST BELOW IS MORE THAN 5 YEARS OLD. RELATED SUPPORT INFORMATION MIGHT BE OUTDATED OR DEPRECATED

On 19/03/2012 at 15:21, xxxxxxxx wrote:

Lol. I can't figure out what you're trying to do.😵
Here's what you posted in your first post:
"is it possible to access the description element flags (DESC_HIDE for example) with python ?"

This indicated to me that you want to dynamically hide and unhide GUI items.
Yet you keep on insisting to use the GetDEnabling() method to do this. But that's not what this method is for.
GetDEnabling() for greying out items so the user can't click on them. The Python docs call it "Ghosting".

Here's a dynamic GUI tag plugin I made with C++ that is compiled to run under the 32 bit version of C4D:https://sites.google.com/site/scottayersmedia/TagExample.zip
Install it.. Add the tag to an object.. And try out the various GUI items to see the dynamic results.
-When you click on the checkbox. The slider disappears and the text changes in the textbox.
-When you click the first button. A new slider is created
-When you select the second combo button option. The text box disappears.
-Etc..

Is this what you're trying to do?
If not. What is the end result you're trying to achieve?

-ScottA

THE POST BELOW IS MORE THAN 5 YEARS OLD. RELATED SUPPORT INFORMATION MIGHT BE OUTDATED OR DEPRECATED

On 19/03/2012 at 16:52, xxxxxxxx wrote:

well, as i said , my goal is to dynamically hide and unhide GUI items. i also found your tag 
example plugin and microbins plugin cookbook entry on my very first google search on this
topic, but as nikklas already said, GetDDescription isn't implemented for python. he also 
pointed me to GetDEnabling. and as its description is kind of vague and the itemdesc 
parameter looked promising, i tried to abuse it to set DESC_HIDE. now i gave up, because 
you can hide and and unhide elements with itemdesc, but the gui doesn't update correctly 
and it seems to be the description of the class itself not the description of the instance of the 
class (no idea why, but when you hide parameter A, it is hidden for all instance of your plugin, 
even for instances you created after the changes to the description) . so i kind of gave up and 
i will use GetDEnabling as it seems to be intended to be used (for enabling/disabling elements)
although i would prefer to be able to acutally hide elements.

THE POST BELOW IS MORE THAN 5 YEARS OLD. RELATED SUPPORT INFORMATION MIGHT BE OUTDATED OR DEPRECATED

On 19/03/2012 at 17:35, xxxxxxxx wrote:

OK.
Just wanted to be sure we were still talking about hiding & and unhiding gizmos as the goal.

-ScottA

THE POST BELOW IS MORE THAN 5 YEARS OLD. RELATED SUPPORT INFORMATION MIGHT BE OUTDATED OR DEPRECATED

On 25/03/2012 at 10:49, xxxxxxxx wrote:

i know it is a bit crazzy , but i am on c++ now. i think i will fail miserably, but nevertheless i thought
i'll have a small chance with my minor c# knowledge. so i revisited this problem and it works fine
with the GetDDescription example from Microbins Cookbook for me.
However i have really many parameters to hide and unhide so i thought it would be good to create
a metheod to shorten things a bit. unfortunately the code isn't working as expected :) it compiles
without errors, but it doesn't hide/unhide the elements and with c4d started from VS in debugmode
it does un/hide the targeted element ( PLIGHT_LIGHT_TYPE_OBJECT ) but not under the expected 
conditions. i guess i have created some logic error i am to stupid to find or i am using pointers in a 
wrong manner.

the relevant code :

Bool Opolylight::GetDDescription(GeListNode *node, Description *description,DESCFLAGS_DESC &flags)
{    
    BaseContainer *data;
    const DescID *singleid;  
    DescID cid;
  
    if(!description->LoadDescription(node->GetType()))
        return FALSE;
  
    data = ((BaseList2D* )node)->GetDataInstance();
  singleid = description->GetSingleDescID();
  
  LONG value[] = {PLIGHT_LIGHT_TYPE_OBJECT};
  UnHideDescription(PLIGHT_SHAPE_OBJECT_LINK, DTYPE_BASELISTLINK, data->GetLong(PLIGHT_LIGHT_TYPE) , value, FALSE , singleid, description);
  
    flags |= DESCFLAGS_DESC_LOADED;
    return TRUE;
}
  
void Opolylight::UnHideDescription(LONG targetid, LONG dtype, LONG checkvalue, LONG value[], Bool isequal, const DescID *singleid, Description *description)
{
	BaseContainer *bc;
	DescID cid;
  
    cid = DescLevel(targetid, dtype, 0);
  
	if(!singleid || cid.IsPartOf(*singleid, NULL))
	{
		bc = description->GetParameterI(cid, NULL);
		if (isequal)
		{
			for (int i = 0; i < sizeof(value);i++)
			{
				if (checkvalue == value[i])
					bc->SetBool(DESC_HIDE, TRUE);
				else 
					bc->SetBool(DESC_HIDE, FALSE);
			}
		}
		else
		{
			for (int i = 0; i < sizeof(value);i++)
			{
				if (checkvalue != value[i])
					bc->SetBool(DESC_HIDE, TRUE);
				else 
					bc->SetBool(DESC_HIDE, FALSE);
			}
		}
    }
}

UnHideDescription is meant to un/hide elements if the checkvalue equals an element from the array value (or doesn't equals if isEqual is False).

thanks for your help in advance :D

THE POST BELOW IS MORE THAN 5 YEARS OLD. RELATED SUPPORT INFORMATION MIGHT BE OUTDATED OR DEPRECATED

On 26/03/2012 at 01:06, xxxxxxxx wrote:

Hi,

There's no problem at all in using a subroutine to do this. But the way it's coded is not going to work.

In GetDDescription() you create an array of LONGs - called value - with one entry. You pass that array to UnhideDescription(). But then you do this:

  
for (int i = 0; i < sizeof(value);i++)   
{   
if (checkvalue == value[i])   
bc->SetBool(DESC_HIDE, TRUE);   
else   
bc->SetBool(DESC_HIDE, FALSE);   
}   

I think what you're intending to do is iterate through all the members of the array value, but you can't use sizeof to do that. sizeof(value) returns the size in bytes of the array. A LONG will occupy 8 bytes (I think!) so with one member you'll be iterating from 0 to 7. There's no way of knowing what values you'll get from that, which probably explains why it doesn't work as expected.

Instead, you'll need to pass to UnhideDescription() the number of elements in the array and iterate with that.

I can't say that will make it work, but I'm sure it won't at the moment for that reason.

Steve

THE POST BELOW IS MORE THAN 5 YEARS OLD. RELATED SUPPORT INFORMATION MIGHT BE OUTDATED OR DEPRECATED

On 26/03/2012 at 03:37, xxxxxxxx wrote:

hi,

thanks for your help. i used this construct as replacement for a for each statement, because as i
understood for each within all mircorsoft languages uses a os specific interface (IEnumerable).
is this wrong ? however the correct approach of determing the size of an array in c++ would be :

LONG size = sizeof(value)/sizeof(LONG);
---------------------------------------
size = 4*1/4;

is this correct ? if not, could someone be so kind and post an example for how he is dealing with
for each *situations*, or in other words , how he is iterating through an array of unknown size ?

edit :

void Opolylight::UnHideDescription(LONG targetid, LONG dtype, LONG checkvalue, LONG value[], Bool isequal, const DescID *singleid, Description *description)
{
	BaseContainer *bc;
	DescID cid;
  
    cid = DescLevel(targetid, dtype, 0);
	LONG len = sizeof(value) / sizeof(LONG);
  
	if(!singleid || cid.IsPartOf(*singleid, NULL))
	{
		bc = description->GetParameterI(cid, NULL);
		if (isequal)
		{
			for (int i = 0; i < (sizeof(value) / sizeof(LONG));i++)
			{
				if (checkvalue == value[i])
					bc->SetBool(DESC_HIDE, TRUE);
				else 
					bc->SetBool(DESC_HIDE, FALSE);
			}
		}
		else
		{
			for (int i = 0; i < (sizeof(value) / sizeof(LONG));i++)
			{
				if (checkvalue != value[i])
					bc->SetBool(DESC_HIDE, TRUE);
				else 
					bc->SetBool(DESC_HIDE, FALSE);
			}
		}
    }
}

using this code, it works perfectly when cinema is started by visual studio in debug mode (things are 
hidden and unhide under given conditions). but when i start cinema alone there is no reaction at all.

THE POST BELOW IS MORE THAN 5 YEARS OLD. RELATED SUPPORT INFORMATION MIGHT BE OUTDATED OR DEPRECATED

On 26/03/2012 at 09:28, xxxxxxxx wrote:

Must admit I'm not sure why it works in the debugger but not when run alone. That seems very odd. One thing to note (which I don't actually do in my tutorial) is to return from GetDDescription with a call to the parent class equivalent - see the SDK for details.

You mention IEnumerable - that's a .NET construct, it may be available in all .NET languages including managed C++, but it won't be available in plain old C++ which is what the Cinema SDK uses.

You can get the array size that way using sizeof(array)/sizeof(element). But in the Maxon SDK if you have an array which may contain a variable number of elements, IMO it's better to maintain a counter of the number of elements in it, or to use the GeDynamicArray or GeAutoDynamicArray classes, as these have functions like GetCount() which returns the number of elements.

Steve

THE POST BELOW IS MORE THAN 5 YEARS OLD. RELATED SUPPORT INFORMATION MIGHT BE OUTDATED OR DEPRECATED

On 26/03/2012 at 10:52, xxxxxxxx wrote:

well,
after some poking arround in the code  i have found out this :

1. i added some console ouput within cinema and all code is execueted as expected (32/64 bit)
2. the problem seems to be connected to 32/64 bit versions of cinema, actually i made a mistake
and didn't set up my target application for debugging properly for the different platforms. so i was
debugging with c4d 32 bit while i was running  c4d 64 bit from my desktop. after having fixed that
the current state is following :

32 bit debug/release api - running from VS and from desktop
64 bit debug/release api - loading plugin, setting values , allocating some objects works, but this
whole UnHideDescription thing refuses to work

i guess i messed up something with my 64 bit plattforms or with compiling the 64 bit versions of 
the api libs. as the natural result to own failure is ranting about other people,  i have to say the
whole c++ api thing is quite unsatisfying for me. i cannot understand why maxon does not provide
the compiled api libs and why maxon doesn't provide at least for their recommanded IDE VS 2005
some template projects. who wants to compile the api each time (using the sdk example projects)?

so i ripped the sdk_examples appart and added the api linker links and i guess something went 
wrong there. i'm sure i'll find the mistake but this is soooo unnecessary ...

will stick with 32 bit for now ... and thanks again for your help steve

THE POST BELOW IS MORE THAN 5 YEARS OLD. RELATED SUPPORT INFORMATION MIGHT BE OUTDATED OR DEPRECATED

On 26/03/2012 at 11:58, xxxxxxxx wrote:

There are instructions on Steve's site for setting up the compiler from scratch without including the API in it that I prefer to use. Although I'm using VS 2010.
This is different from the way Maxon says to do it. But I like it better.

Being a total newbie. It was a little hard for me to follow Steve's notes. So I created my own cheat sheet that is easier for me to follow:

Instructions:  
  
-Create your plugin with the skeleton folder structure inside of it  
*** Make sure you have your main.cpp file set up. And in the source folder ***  
*** The main.cpp file is used to load the api .h files. As well as register your plugins***  
  
-Put your plugin folder with the basic file structure in the C4D plugins folder  
-Create a new emtpy C++ project(uncheck make directory. And select the C4D plugins folder(not your new plugin folder) as the source)  
-Name your project the same name as your new plugin folder so it will write the solution into that folder properly  
-Add the Debug & Release property sheets to the project(view->otherwindows->propertymanager) by right clicking them and choose: CINEMA 4D R12\resource\_api_lib\  
 Then adding these files to the proper project folders:  
   
Debug | Win 32 = DebugWin32.props   
Debug | Win 32 = ReleaseWin32.props   
  
  
-Right click the source folder and choose Add Existing Project  
-Add your main.cpp file--->VisualStudio will load the .cpp file. And it will also add all of the .h files from the api  
  
  
In configuration properties General     
  -Set Target Extension to .cdl  
  -Set Configuration type to Dynamic Library.dll  
    
  
In configuration properties Debugging  
  -Set Command to:   C:\Program Files\MAXON\CINEMA 4D R12\CINEMA 4D.exe or whichever .exe version you'll launch with the debugger  
  
  
In configuration properties C/C++  
  -In command Line. Add this to the bottom window:  /vmg /vms   
  
  
In configuration properties Linker  
  -In General->Set Output File to the same name as your project and with the extension: .cdl    
  -In General->Set Additional Library Directories to:   C:\Program Files\MAXON\CINEMA 4D R12\resource\_api_lib  
  -In Input-> Additional Dependencies. Add this to the list:  _api_Win32_Debug.lib  
  -in Advanced-> change Randomized Base Address to Disable Image Randomization (DYNAMICBASE:NO)  
  
  
DONE with 32 bit setup!!  
  
  
To set up 64 bit compiling we have more work to do:  
-From the top menu. Click the solution platforms dropdown and select Configuration Manager  
-On the right drop down list. Select "New" from the options.Then select x64  
-Next be sure to also enable the BUild option for this new platform  
-Add the Debug & Release property sheets to the project just like you did with the 32 bit platform setup above  
-This time select these files from the _api_lib folder:  
  
Debug | Win 64 = DebugWin64.props   
Debug | Win 64 = ReleaseWin64.props  
  
  
Right click on the project in the solution window and select properties  
Set up the various options the same way you set up the 32 bit options with two exceptions:  
-In Linker->General: Set Output File to the same name as your project and with the extension: .cdl64   
-In Linker Input->Additional Dependencies. Add this to the list:  _api_x64_Debug.lib  
  
  
DONE with 32&64 bit setups!!

For compiling this way in R13.
I also have to put a copy of the projectsettings.props file from the cinema4dsdk into the project's folder.
This is done instead of the macro setup procedure the SDK says to do.

-ScottA

THE POST BELOW IS MORE THAN 5 YEARS OLD. RELATED SUPPORT INFORMATION MIGHT BE OUTDATED OR DEPRECATED

On 26/03/2012 at 12:22, xxxxxxxx wrote:

Originally posted by xxxxxxxx

i have to say the whole c++ api thing is quite unsatisfying for me. i cannot understand why maxon does not provide the compiled api libs and why maxon doesn't provide at least for their recommanded IDE VS 2005 some template projects. who wants to compile the api each time (using the sdk example projects)? so i ripped the sdk_examples appart and added the api linker links and i guess something went  wrong there. i'm sure i'll find the mistake but this is soooo unnecessary ... will stick with 32 bit for now ... and thanks again for your help steve

In fairness, the fact that you have to compile the API for 32 and 64 bit does show that your compiler setup is working. IME the API and the example plugins just compile without any problems, so if that works then you know that any problems are down to you rather than the supplied code.

But of course, you only have to compile the AP once. After that you just tell the linker to use the now compiled libraries. Otherwise the whole compile-link-run (-crash-fix-compile, etc.) cycle would be intolerable. The tut on my site shows, as Scott says, how to set up a project from scratch but I agree that it isn't that straightforward.

Steve