A Few SDK Questions

I have been using the SDK ok yet after trying certain things have run into a few problems, i tested with various things but did not get them working. If anyone knows about any of the following questions that would be a big help.

---

Is it possible to make multiple copy's of a object act differently when in a level together?

For example i have a simple counter with 2 added in the level but if i add 10 to only one of them both get the value 10. I want just one to have 10 and the other whatever value that was using etc, how do you make any objects added react as separate objects?

----

It is not important but why does Menu.h grey out the syntax after the first condition menu in visual C++ 2008 express? They seem to work ok still but i am not sure why it does that though.

---

Ints and static floats work fine but how do you set a float through a action and then get this returned as the float in a expression?

For example if i do this in the action -

NeededValue = Param(TYPE_FLOAT);

and this in the expression -

ReturnFloat(NeededValue);

It works like a int still at the action then makes it float so if in the events you set 10.5 it gives 10.0 back. I want it to return 10.5 though but cannot get this to work correctly.

---

How do you add/attach together strings that have already been set or one through a action and the other static like this for example -

char w1 = (params input string here);
char w2[] = {'b','c'};

char out[] = (w1,w2);

ReturnString(out);

The code is of course broken but should give the basic idea of what i mean. If i am just setting a string to send through or merge 2 static set strings that works fine but i cannot get this to work.

---

Also with strings how do you combine values and strings so if you had something like "90 percent" with the 90 as a int or float etc and the "percent" added on and returned as a string?

Quote:

---

Originally Posted by Atom

Is it possible to make multiple copy's of a object act differently when in a level together?

---

Store the value in RUNDATA (aka rdPtr).

Quote:

---

Originally Posted by Atom

It is not important but why does Menu.h grey out the syntax after the first condition menu in visual C++ 2008 express? They seem to work ok still but i am not sure why it does that though.

---

VC++ greys out text if it's in a #if condition and it doesn't think it will pass that condition when it compiles. It can be wrong.

Quote:

---

Originally Posted by Atom

Ints and static floats work fine but how do you set a float through a action and then get this returned as the float in a expression?

For example if i do this in the action -

NeededValue = Param(TYPE_FLOAT);

and this in the expression -

ReturnFloat(NeededValue);

It works like a int still at the action then makes it float so if in the events you set 10.5 it gives 10.0 back. I want it to return 10.5 though but cannot get this to work correctly.

---

In the action line, is "NeededValue" a float or int variable?

Quote:

---

Originally Posted by Atom

How do you add/attach together strings that have already been set or one through a action and the other static like this for example -

char w1 = (params input string here);
char w2[] = {'b','c'};

char out[] = (w1,w2);

ReturnString(out);

The code is of course broken but should give the basic idea of what i mean. If i am just setting a string to send through or merge 2 static set strings that works fine but i cannot get this to work.

---

I'd look at the string manipulation functions in the VC++ help. Note that you'll need to learn about memory allocation at some point to deal with strings properly in C/C++, it's not hidden like in other languages.

Quote:

---

Originally Posted by Atom

Also with strings how do you combine values and strings so if you had something like "90 percent" with the 90 as a int or float etc and the "percent" added on and returned as a string?

---

Look up sprintf(). You'll need a string allocated large enough to hold the result (sprintf will kindly tell you how big a buffer you need if you don't give it one).

I have a few questions as well, but I can only remember one at this point (I'll post a new thread for more)

---

How do you set IDs of values etc based on the event number? I can't seem to find a way to ID a value without a parameter that the player has to input, and that can be quite tedious. Is there a way to dynamically set IDs so that it can be called upon without needing an extra "ID" parameter?

By the way, the ID parameter points to the cell number in an array in rdPtr

Thanks Dynasoft,

The menu was as i guess it seems. I am still confused with most of this though. For the RUNDATA im not sure what to change still as i am fairly new to the SDK. I was hoping someone would post some basic snippets or steps so i am lost still with that bit right now. :(

The float thing is as said before i just need a value i pass into it with a action in the event editor this value would then need to be a float/decimals (if thats what was entered in) and i would do maths etc in the code, then it's passed to the codes expressions where the user returns it in mmf for a counter or whatever. The value needs to be able to be like 150.2574 etc if i entered that in though yet it always turns it to int even when i set it to a float type.

It's fine if i stick to int values but i want to do more advanced maths now and i haven't seen any source code do this as there is not much available so i have no idea how to do it. I was hoping for a simple code snippet to show the method of doing it. For this i tried loads of stuff like changing the types to float, long etc and various other stuff as well but nothing worked. I am guessing it's the most basic of things when you know how also yet no luck.

The string stuff i can do if it is the same thing like a static known string and combining these. The problem is when i mix a input string from a action or expression with this. I have already searched online and did not find anything useful for that as it's not based on the SDK but i will look at the VC++ help but im still not fully sure what i would be searching for.

For values with strings sprintf() should work hopefully thanks i will try this.

So yeah i am still wondering about 3 of the previous questions. For this i spend about 6-8 hours trying things with no luck but im guessing most of these will have a simple solution when i know how to code it. Since the code in the examples i gave is broken it makes it hard to describe what i mean fully so sorry if my questions are slightly confusing but hopefully you know what i mean. Thanks for the help.

I don't do this much, i was hoping someone would help sort these though so... Bump

---

I was mainly still needing the use of floats/decimal numbers in actions and to send to and return these in expressions although i am still not sure why it doesn't work.

Why does this code for the action -

Code:

---

ACTION(
        /* ID */                        0,
        /* Name */                        "Set The Counter",
        /* Flags */                        0,
        /* Params */                (1,PARAM_NUMBER,"Set The Counter")
)
{
        LastSetValue = Param(TYPE_FLOAT);
        CounterValue = LastSetValue;
}

---

and this the expression -

Code:

---

EXPRESSION(
        /* ID */                        0,
        /* Name */                        "Get Counter Value(",
        /* Flags */                        0,
        /* Params */                (0)
) {
    ReturnFloat(CounterValue);
}

---

Give a int without the decimals when built?

You're supposed to set EXPFLAG_DOUBLE instead of 0 in the expression flags. I think it works even if you don't set it though.

The problem is (unless they fixed it) that Param(TYPE_FLOAT) returns an integer. I wrote some macros to get around this problem when I made my Box2D extension:

Code:

---

//Float Param Help Functions
inline float gfParam(LPRDATA rdPtr, int param1) {int f = Param(TYPE_FLOAT); return *(float*)&f;}
inline float xgfParam(LPRDATA rdPtr, int param1) {int f = ExParam(TYPE_FLOAT); return *(float*)&f;}

//Actions/Conditions
//Object Parameter
#define oParam() (LPRO)Param(-1);
//Undefined Parameter
#define uParam() Param(-1);
//Long/Int Parameter
#define lParam() Param(TYPE_INT);
//Float Parameter
#define fParam() gfParam(rdPtr,param1);
//String Parameter
#define sParam() (LPCSTR)Param(TYPE_STRING);

//Expressions
//Long/Int Parameter
#define xlParam() ExParam(TYPE_INT);
//String Parameter
#define xsParam() (LPCSTR)ExParam(TYPE_STRING);
//Float Parameter
#define xfParam() xgfParam(rdPtr,param1);

---

Thanks Werbad, so it seems the actual float type is wrong then. The macro seems like what i need but the code sample confuses me slightly though because i am guessing when you say macro it means you just use parts from the above when needed? Although i am not fully sure how to use that method as i am new to it so hopefully you don't mind me asking.

I am guessing i need the help bit in place then use the define params where you need a param. If you keep the code format i used above (default rsdk layout) using the 2 parameters 'LastSetValue' and 'CounterValue' how would you write the same thing using the macro method?

Currently i have this code, this goes before the condition code block -

Code:

---

// Default values
float CounterValue = 0.0;
float LastSetValue = 0.0;

//Float Param Help Functions
inline float gfParam(LPRDATA rdPtr, int param1) {int f = Param(TYPE_FLOAT); return *(float*)&f;}
inline float xgfParam(LPRDATA rdPtr, int param1) {int f = ExParam(TYPE_FLOAT); return *(float*)&f;}

---

The action -

Code:

---

ACTION(
        /* ID */                        0,
        /* Name */                        "Set The Counter",
        /* Flags */                        0,
        /* Params */                #define fParam() gfParam(rdPtr,param1);
)
{
        LastSetValue = fParam(TYPE_FLOAT);
        CounterValue = LastSetValue;
}

---

The expression -

Code:

---

EXPRESSION(
        /* ID */                        0,
        /* Name */                        "Get Counter Value(",
        /* Flags */                        EXPFLAG_DOUBLE,
        /* Params */                (0)
) {
    ReturnFloat(CounterValue);
}

---

But if i enter something like 120.575 as the value it returns 120 back so i am guessing i am still not doing it correctly.

I prefer to avoid all rSDK facilities when dealing with floating point in actions, all of it seems to break in some scenario or another. Here's a couple routines I use in XLua - you can pretty much ignore anything that starts with rdPtr-> since that's internal to XLua, manipulating the stacks.

Code:

---

ACTION(
  /* ID */                                39,
  /* Name */                        "Push Floating Parameter: %0",
  /* Flags */                        0,
  /* Params */                        (1, PARAM_EXPRESSION, "Number")
) {
        //float p1 = (float) Param(TYPE_FLOAT);
        long p1temp = CNC_GetFloatParameter(rdPtr);
        float p1 = *(float*) &p1temp;

        if (rdPtr->luaMan->state) {
                rdPtr->luaMan->state->stack.PushParam(p1);
        }
}

---

Note above the commented out defunct code (actually it might have worked in this instance, but broke in other actions). Instead I'm using Clickteam's interface directly.

Code:

---

ACTION(
  /* ID */                                46,
  /* Name */                        "Set Floating Variable %0 to %1",
  /* Flags */                        0,
  /* Params */                        (2, PARAM_EXPSTRING, "Variable Name", PARAM_EXPRESSION, "Numeric Value")
) {
        char* p1 = (char*) CNC_GetStringParameter(rdPtr);
        long p2temp = CNC_GetFloatParameter(rdPtr);

        float p2 = *(float*) &p2temp;

        if (rdPtr->luaMan->state) {
                rdPtr->luaMan->state->SetFloatingVariable(p1, p2);
        }
}

---

Another example, using multiple parameters, note I use CNC_* functions for all the parameters if I use them for any.

Code:

---

EXPRESSION(
  /* ID */                                21,
  /* Name */                        "LF_FloatingReturn(",
  /* FLAGS */                        0,
  /* Params */                        (1,EXPPARAM_NUMBER,"1-based index")
) {
        int p1 = ExParam(TYPE_INT);

        if (rdPtr->luaMan->state == NULL) {
                ReturnFloat(0);
        }
        ReturnFloat(rdPtr->luaMan->state->stack.GetNumberReturn(p1));
}

---

Expressions don't do anything special. No idea what you're doing with the EXPFLAG_DOUBLE flag, I've never used it and I get correct float behavior. The Return* rSDK functions already set the correct flags for you.

You had to set it in the old SDK I think, but as I said it works fine without it so whatever...
My macros were made to simplify the parameter retrieval to one-word functions rather than the many different ways used for different parameter types.

Here's an example from Box2D of how they're used:

Code:

---

ACTION(
        /* ID */                        4,
        /* Name */                        "World - Set Gravity to (%0,%1)",
        /* Flags */                        0,
        /* Params */                (2,PARAM_NUMBER,"Gravity X",PARAM_NUMBER,"Gravity Y")
) {
        float x = fParam();
        float y = fParam();

        rdPtr->world->SetGravity(b2Vec2(x,y));
}

---

Code:

---

EXPRESSION(
        /* ID */                        144,
        /* Name */                        "BodyVelocityAtWorldPointX(",
        /* Flags */                        EXPFLAG_DOUBLE,
        /* Params */                (3,EXPPARAM_NUMBER,"Body ID",EXPPARAM_NUMBER,"X Coordinate",EXPPARAM_NUMBER,"Y Coordinate")
) {
        int n = xlParam();
        float x = xfParam();
        float y = xfParam();

        b2Body* b = getBody(n,rdPtr);
       
        if(!b) ReturnFloat(0);
       
        ReturnFloat(b->GetLinearVelocityFromWorldPoint(b2Vec2(x/rdPtr->scale,y/rdPtr->scale)).x);
}

---

If you want to use them, just add the defines to the top of your main.cpp file.

Thanks everyone, hopefully i am not being annoying here but i still can't get this working for some reason. The confusion i am currently having is your code snippets are written from other code sources (xlua and box2d) but i was hoping someone would show it in the style i had. In these they use different param names, id's, have extra param inputs and inputs on the expression but i was trying to just get it directly from the action without user input.

Trying to use the macro method and using just a single value rather than the other temp also now the top code looks like -

Code:

---

#include "common.h"

// Set default values
float CounterValue = 0.0;

//Float Param Help Functions
inline float gfParam(LPRDATA rdPtr, int param1) {int f = Param(TYPE_FLOAT); return *(float*)&f;}
inline float xgfParam(LPRDATA rdPtr, int param1) {int f = ExParam(TYPE_FLOAT); return *(float*)&f;}

//Actions/Conditions
//Object Parameter
#define oParam() (LPRO)Param(-1);
//Undefined Parameter
#define uParam() Param(-1);
//Long/Int Parameter
#define lParam() Param(TYPE_INT);
//Float Parameter
#define fParam() gfParam(rdPtr,param1);
//String Parameter
#define sParam() (LPCSTR)Param(TYPE_STRING);

//Expressions
//Long/Int Parameter
#define xlParam() ExParam(TYPE_INT);
//String Parameter
#define xsParam() (LPCSTR)ExParam(TYPE_STRING);
//Float Parameter
#define xfParam() xgfParam(rdPtr,param1);

---

Then the action is -

Code:

---

ACTION(
        /* ID */                        0,
        /* Name */                        "Set The Counter",
        /* Flags */                        0,
  /* Params */                        (1,PARAM_NUMBER, "Number")
) {
    float CounterValue = fParam();
}

---

And the expression (which is not using a param input) is -

Code:

---

EXPRESSION(
  /* ID */                                0,
  /* Name */                        "Get Counter(",
        /* Flags */                        EXPFLAG_DOUBLE,
  /* Params */                        (0)
) {
    ReturnFloat(CounterValue);
}

---

So i am confused as to why it still passes a int, i just read what both of you said and tried various things yet it still doesn't work. I am also confused as to why the SDK is broken/needs workarounds with the default download (also has other errors that are fixed, float still does not work for me though). Hopefully with some help i can get floats working though, usually in other coding styles i have them working fine.

It builds fine and has no errors but when testing it still rounds any input decimals to a int rather than keeping the float. If anyone could correct that simple code i posted and maybe give pointers as to where i am having the problems that would be a huge help.

Are you sure you didn't declare CounterValue as an int instead of a float? I don't seem to see any other reason it wouldn't work.

Also, is CounterValue a global value? If it is you will have problems when you use multiple instances of the extension. You should store the data in the runData structure and access it via rdPtr->CounterValue to ensure it's unique for every object.

Edit:
Did even you declare the value globally at all? Looking at it again it seems like the value is local to the ACTION function...

Quote:

---

Originally Posted by Werbad

Did even you declare the value globally at all? Looking at it again it seems like the value is local to the ACTION function...

---

That confuses me as well, but there is "float CounterValue = 0.0;" in the first code, which seems to be at the top of Main.cpp.

I just remembered something, I have made a modification to rSDK which is required for the float macro to work. Since you have the original you need to change the helper function for fParam to:

Code:

---

inline float gfParam(LPRDATA rdPtr, int param1) {int f = CNC_GetFloatParameter(rdPtr); return *(float*)&f;}

---

Arggg, why is this not working..... :(

I just replaced the text with what Werbad says but it still won't work. The float CounterValue = 0.0; is how i thought you set a temp value, i only did that though because i would get errors that the used parameter was not set with things i tested before and that fixed it.

I also don't have much idea to doing certain things because the tutorials for the SDK are well... not there other than the getting started tutorial on the wiki which does not explain stuff like this. So basically i am fairly new to it although i know basic c++ from other places yet i have found i can't just get started with this like i can other things as it starts you off with too many problems. I have seen a few source codes, none use float like this etc. Basically i think clickteam needs more resources for this and also to make sure any SDKs they have available build well on first try and use working float types etc.

Is my code suppose to work then? because it definitely doesn't for me. I don't know why i am needing a workaround for floats though but this seems like a brick wall now, i am frustrated with this, have wasted too much time on trying to do what should be basic and i am at the point where i feel totally stupid and i am just going ask now...

If possible can someone please post the full template as zip etc of a basic 1 action and 1 expression object where you set a float value and then return this value in the expression. So in mmf you could say enter 3.555 in the Set Counter action and then in a expression get this value back as 3.555.

Btw - Sorry for over reacting and also what is the most current RSDK just so i can confirm i am using a new version rather than one that is out of date and might be the reason i am getting this.

It's been about 2 days now and i still need to know this info to get floats working. I am still not sure why floats are broken by default but i want to work with them in the SDK so...

Bump

I don´t know if this is what you are looking for;
ACTION
long p2;
float pf2;

p2 = CNC_GetFloatValue(rdPtr, 1);
//Move the float into the long placeholder.
memcpy((void *)&pf2,(void *)&p2,sizeof(float));
// now you float is pf2
.
.
.

EXPRESSION
long rdPtr->value_f = 0;
float fp2 = (float)fp1;

//Setting the HOF_FLOAT flag lets MMF know that you are returning a float.
rdPtr->rHo.hoFlags |= HOF_FLOAT;

//Move the float into the long placeholder.
memcpy((void *)&rdPtr->value_f,(void *)&fp2,sizeof(float));

//Now we can return the Float safely.
return ((long)rdPtr->value_f);
}

EDIT: FORGOT TO MENTION EXPRESSION WAY

Regards

Thanks for the help FVivolo, when i use this code though it gives me build errors.

I still haven't been able to get the code Werbad etc posted to work yet either but i think that is because i need to declare CounterValue as a float and store in runData structure etc but i have no idea how to do this at the moment.

I try to use what FVivolo posted but i get build errors with the Rsdk still -

Code:

---

#include "common.h"

// ============================================================================
//
// CONDITIONS
//
// ============================================================================

// ============================================================================
//
// ACTIONS
//
// ============================================================================

ACTION(
        /* ID */                        0,
        /* Name */                        "Set The Counter",
        /* Flags */                        0,
  /* Params */                        (1,PARAM_NUMBER, "Number")
) {
        long p2;
    float pf2;
        p2 = CNC_GetFloatValue(rdPtr, 1);

// Move the float into the long placeholder.
    memcpy((void *)&pf2,(void *)&p2,sizeof(float));
// Now you float is pf2
}

// ============================================================================
//
// EXPRESSIONS
//
// ============================================================================

EXPRESSION(
  /* ID */                                0,
  /* Name */                        "Get Counter(",
        /* Flags */                        0,
  /* Params */                        (0)
) {
long rdPtr->value_f = 0;
float fp2 = (float)fp1;

//Setting the HOF_FLOAT flag lets MMF know that you are returning a float.
rdPtr->rHo.hoFlags |= HOF_FLOAT;

//Move the float into the long placeholder.
memcpy((void *)&rdPtr->value_f,(void *)&fp2,sizeof(float));

//Now we can return the Float safely.
return ((long)rdPtr->value_f);
}

---

Would i need some other include maybe? because the debug tells me things like it needs a ; after -> etc so it seems it doesn't recognize the command.

The problem is
long rdPtr->value_f = 0;

You can't declare variables in a struct after it has been created.
Instead of doing this, go to Data.h and add "long value_f;" in the LPRDATA struct.
However, I don't see a reason why you would store the float in rdPtr at all.
Just remove the "rdPtr->" from all "value_f" and it should work..

Thanks Looki, that fixes all errors but one which is -

1>.\Main.cpp(43) : error C2065: 'fp1' : undeclared identifier

So it seems i need to make it declared as float and maybe a default value but i am not fully sure how to do that.

I am confused as to why in the expression it is not using pf2 as it actually looks like it was setup to require a parameter input. For the expression though i am just just wanting to collect the value set in the action so basically nothing but returning a float when triggered, most of the code people have posted have parameters used though.

Bump..

Floats still can't be passed from a action to a expression (without parameters on the expression), i am surprised i am having to bump this thread so much this to be honest. Sorry for being annoying but i want to work with them in the SDK so i need to find a solution.

Not to be rude, but you've repeatedly been given code that works, you just seem to be unable to use it.

Not to be rude to anyone but i don't think so.

I say this because when i first asked i wanted something else than what i was given, for your reply it was very short with not much info so kind of the same thing as when someone asks how to do something in MMF and someone just posts a object name etc. Not saying that it was not helpful of course just it makes things harder to get working with not much info.

When i was given code snippets these were all clearly written in another way to what i was asking for so a snippet from the XLua code etc. This made things confusing as it included extra parts of code and also the names of parameters were not similar again making things more confusing.

In fact if i just wanted to set and output a float there is no problem it is i am guessing when sending it from the action to the expression. People have clearly mistaken what i asked for also which is to use the action with a float parameter and then pass the float to be returned in the expression. All of the code posted has had a parameter input on the expression which is not what i wanted either, i just want to receive the value with that not input one. It is maybe the sending from the action to expression which is causing problems but i am not sure how to do this correctly as everything is returning a int and why i need help with this.

Also the only reason i keep bumping this thread now is because people keep ignoring it, i am guessing due to other people posting in other threads for betas of objects. I think this forum area should be renamed to extension beta testing because it's mainly for that and any SDK dev questions that get asked go down the page fast and get forgotten. As i said i am sorry for being annoying but i need a solution to this, if you don't like me asking so much please could someone just post a complete and working Main.cpp snippet of what i am asking for so i can see how it is done correctly. In most other SDKs this would not be a issue for me so again sorry but as Werbad said earlier float seems to return integer which is making things annoying and confusing.

Hi Atom,

First of all, sorry for the delay, Looki indeed is correct "RdPtr->" was taken from my Data structure of Basic object, if you try the example basictest_X, in the edit box named "I" you will get an answer corresponding to the number you put inside it, in this case a float, i really don't understand your question, but i am passing float value from action to expression indeed as variable inside a basic program block. (using these small pieces of codes).

consider if you only define a float inside the data structure rdPtr->myfloatA and you can read from action and set the value using the small code and copy using
//Move the float into the long placeholder.
memcpy((void *)&rdPtr->value_f,(void *)&fp2,sizeof(float));

define an Action like this

short WINAPI DLLExport SetFVar(LPRDATA rdPtr, long param1, long param2)
{

long p2;

float pf2;

//Actions work just like Conditions
p2 = CNC_GetFloatValue(rdPtr, 1);
//Move the float into the long placeholder.
memcpy((void *)&pf2,(void *)&p2,sizeof(float));

//copy the value, you may copy directly without using pf2

rdPtr->myfloatA = pf2;

return 0;
}

now you have it for future use inside the data structure of the object so when you get to expression, define a new expression without parameters, but this is the key;

//Setting the HOF_FLOAT flag lets MMF know that you are returning a float.
rdPtr->rHo.hoFlags |= HOF_FLOAT;


//Now we can return the Float safely.
return ((long)rdPtr->myfloatA);


IT SHOULD WORKS!
Related to SDK, i am using the original SDK.

Hope this help you.

Thanks for the help, i have been using the RSDK though so i am not fully sure how to use the normal one yet and get errors back which is why i was hoping someone would just post a simple code block i could copy and paste to test. It gives 6 errors back, mainly undeclared identifiers and these type -

error C2039: 'myfloatA' : is not a member of 'tagRDATA'

The full Main.cpp code i currently have is -

Code:

---

// Common Include
#include        "common.h"

// Quick memo: content of the eventInformations arrays
// ---------------------------------------------------
// Menu ID
// String ID
// Code
// Flags
// Number_of_parameters
// Parameter_type [Number_of_parameters]
// Parameter_TitleString [Number_of_parameters]

// Definitions of parameters for each condition
short conditionsInfos[]=
                {
                IDMN_CONDITION, M_CONDITION, CND_CONDITION, EVFLAGS_ALWAYS, 3, PARAM_EXPRESSION, PARAM_EXPRESSION, PARAM_EXPRESSION, M_CND_P1, M_CND_P2, M_CND_P3,
                };

// Definitions of parameters for each action
short actionsInfos[]=
                {
                IDMN_ACTION, M_ACTION,        ACT_ACTION,        0, 0,
                };

// Definitions of parameters for each expression
short expressionsInfos[]=
                {
                IDMN_EXPRESSION, M_EXPRESSION, EXP_EXPRESSION, 0, 3, EXPPARAM_LONG, EXPPARAM_LONG, EXPPARAM_LONG, 0, 0, 0,
               
                //Note in the following.  If you are returning a string, you set the EXPFLAG_STRING.       
                IDMN_EXPRESSION2, M_EXPRESSION2, EXP_EXPRESSION2, EXPFLAG_STRING, 1, EXPPARAM_STRING, 0,
               
                //Note in the following.  If you are returning a float, you set the EXPFLAG_DOUBLE
                IDMN_EXPRESSION3, M_EXPRESSION3, EXP_EXPRESSION3, EXPFLAG_DOUBLE, 1, EXPPARAM_LONG, 0,
                };

// ============================================================================
//
// CONDITION ROUTINES
//
// ============================================================================

// ============================================================================
//
// ACTIONS ROUTINES
//
// ============================================================================

short WINAPI DLLExport SetFVar(LPRDATA rdPtr, long param1, long param2)
{

long p2;

float pf2;

//Actions work just like Conditions
p2 = CNC_GetFloatValue(rdPtr, 1);
//Move the float into the long placeholder.
memcpy((void *)&pf2,(void *)&p2,sizeof(float));

//copy the value, you may copy directly without using pf2

rdPtr->myfloatA = pf2;

return 0;
}

// ============================================================================
//
// EXPRESSIONS ROUTINES
//
// ============================================================================

long WINAPI DLLExport Expression(LPRDATA rdPtr,long param1)
{
//Setting the HOF_FLOAT flag lets MMF know that you are returning a float.
rdPtr->rHo.hoFlags |= HOF_FLOAT;

//Now we can return the Float safely.
return ((long)rdPtr->myfloatA);
}

// ----------------------------------------------------------
// Condition / Action / Expression jump table
// ----------------------------------------------------------
// Contains the address inside the extension of the different
// routines that handle the action, conditions and expressions.
// Located at the end of the source for convinience
// Must finish with a 0
//
long (WINAPI * ConditionJumps[])(LPRDATA rdPtr, long param1, long param2) =
                        {
                        Condition,
                        0
                        };
       
short (WINAPI * ActionJumps[])(LPRDATA rdPtr, long param1, long param2) =
                        {
                        Action,
                        0
                        };

long (WINAPI * ExpressionJumps[])(LPRDATA rdPtr, long param) =
                        {   
                        Expression,
                        Expression2,
                        Expression3,
                        0
                        };

---

Also i notice that this inputs 2 parameters on the action but i just need 1 for the action and none on the expression. I am not sure how to set the inputs to 0 for the expression with this SDK either so in my code it has 1 still. Using floats is actually no issue if i use it for the expression only but the problem is when i send from the action to the expression. It may have been better to have used the standard SDK from the start though but the RSDK seems to have a nicer organized structure in my opinion. Sorry for all the questions, hopefully i can get this working though, many thanks

Quote:

---

Originally Posted by Atom

error C2039: 'myfloatA' : is not a member of 'tagRDATA'

---

This is because you don't have 'myfloatA' declared inside 'tagRDATA'.

Open main.h, and add "float myfloatA;" to the rdata struct.

Thanks Dynasoft, this fixed the myfloatA errors :)

Now it gives 4 errors for these -

error C2065: 'Condition' : undeclared identifier
error C2065: 'Action' : undeclared identifier
error C2065: 'Expression2' : undeclared identifier
error C2065: 'Expression3' : undeclared identifier

But i am not sure how to declare these as they are not a float etc as far as i know, if i delete the Condition / Action / Expression jump table it gives more errors.

I also tried converting it to the RSDK like -

Code:

---

#include "common.h"

// ============================================================================
//
// ACTIONS
//
// ============================================================================

ACTION(
        /* ID */                        0,
        /* Name */                        "Set The Counter",
        /* Flags */                        0,
  /* Params */                        (1,PARAM_NUMBER, "Number a")
) {
long p2;
float pf2;

//Actions work just like Conditions
p2 = CNC_GetFloatValue(rdPtr, 1);
//Move the float into the long placeholder.
memcpy((void *)&pf2,(void *)&p2,sizeof(float));

//copy the value, you may copy directly without using pf2
rdPtr->myfloatA = pf2;

}

// ============================================================================
//
// EXPRESSIONS
//
// ============================================================================

EXPRESSION(
        /* ID */                        0,
        /* Name */                        "Get(",
        /* Flags */                        0,
        /* Parameters */        (0)
) {
//Setting the HOF_FLOAT flag lets MMF know that you are returning a float.
rdPtr->rHo.hoFlags |= HOF_FLOAT;

//Now we can return the Float safely.
return ((long)rdPtr->myfloatA);
}

---

With the myfloatA in data.h which i think is the place in RSDK (maybe not) like -

Code:

---

// --------------------------------
// RUNNING OBJECT DATA STRUCTURE
// --------------------------------
// If you want to store anything between actions/conditions/expressions
// you should store it here

typedef struct tagRDATA
{
        #include "MagicRDATA.h"
       
        // Declare myfloatA inside tagRDATA
        float myfloatA;
       
} RUNDATA;
typedef        RUNDATA        * LPRDATA;

---

This builds just fine however when i test it in MMF setting the action value at something like 3.56 and then get this in a counter it just shows as 0 so it seems i have made a mistake somewhere with the RSDK version.

Use the rSDK version to save you the headaches.

That said, I think I see two errors in your code:
1:
p2 = CNC_GetFloatValue(rdPtr, 1);
I think that should be 0 instead of 1, because you want the first parameter.

2:
return ((long)rdPtr->myfloatA);
This is going to screw up royally. You need to do something to pack the float into a long, not just cast it, like:
return *(long*)&rdPtr->myfloatA;

PS:
You can simplify the action code to:
long p2 = CNC_GetFloatValue(rdPtr, 0);
float pf2 = *(float*)&p2;
rdPtr->myfloatA = pf2;
or even:
long p2 = CNC_GetFloatValue(rdPtr, 0);
rdPtr->myfloatA = *(float*)&p2;

Thanks Dynasoft this works great :D

The code is quite simple and working with the RSDK which is what i have decided to use also which is good. I have made a note of this so i don't forget it in the future.

This seems to be the main things sorted for now, thanks for the help everyone :D

If you are using the rSDK you need to add it to Data.h. ;)