Snap object in mouse input

[Helper]

On 08/01/2013 at 09:19, xxxxxxxx wrote:

What do you mean with "the snap does not work"? You are not even trying to snap the
object, just setting it to the cursor position.

joint.SetAbsPos(cursorpos)

If you want to to snap to a grid with a spacing of 50 units (just for the sake of example), you
need to adjust the position vector.

def snap_value(value, step) :
      rest = value % step
      on_grid = value - rest

if rest / float(step) >= 0.5:
          on_grid += step

return on_grid

def snap_grid(vector, gridsize) :
      vector = c4d.Vector(vector)
      vector.x = snap_value(vector.x, gridsize)
      vector.y = snap_value(vector.y, gridsize)
      vector.z = snap_value(vector.z, gridsize)
      return vector

# ...

gridsize = 50
  position = snap_grid(cursorpos)
  joint.SetAbsPos(position)

See the full code here.

-Niklas

[Helper]

On 08/01/2013 at 09:41, xxxxxxxx wrote:

Thanks NiklasR, excuse my bad English ... I would like to know how to snap the active object on another object...

[Helper]

On 08/01/2013 at 09:45, xxxxxxxx wrote:

And what do you mean with "snapping it to another object"? To what part of the other object?
The other objects' position, points, surface, or whatever?

-Nik

[Helper]

On 08/01/2013 at 09:56, xxxxxxxx wrote:

Surface 

[Helper]

On 08/01/2013 at 10:15, xxxxxxxx wrote:

You need to use the c4d.utils.GeRayCollider class to find an intersection on the surface of the
destination-object, intersected by the line defined by the mouse-position in global space and
the camera position.

op = doc.GetActiveObject()
      op_next = op.GetNext()
      if not op_next:
          return True

# Convert op_next to a polygon object.
      # Simplified here, just for demonstration purpose.
      op_next = op_next.GetCache()
      if not op_next or not op_next.CheckType(c4d.Opolygon) :
          print "Cache is not existent or is not a polygon object."
          return True
     
      win.MouseDragStart(button=device, mx=int(mx), my=int(my), flags=c4d.MOUSEDRAGFLAGS_DONTHIDEMOUSE)
      result, dx, dy, channel = win.MouseDrag()
      while result==c4d.MOUSEDRAGRESULT_CONTINUE:
          mx += dx
          my += dy            
 
          cursorpos = bd.SW(c4d.Vector(mx,my,500))     #screen to world conversion
          if(axis == 1) :
              cursorpos.x = 0
          elif(axis == 2) :
              cursorpos.y = 0                          #Constrain drawing along an axis based on the comboButton's value
          elif(axis == 3) :
              cursorpos.z = 0

cam_pos = bd.GetSceneCamera(doc).GetAbsPos()
          collider = c4d.utils.GeRayCollider()
          collider.Init(op_next)

length = 500000
          direction = -(cam_pos - cursorpos).GetNormalized()

did_intersect = collider.Intersect(cam_pos, direction, length)
          if did_intersect:
              position = collider.GetNearestIntersection()['hitpos']
              print "Intersection at", position
          else:
              print "No intersection."
              position = cursorpos

doc.AddUndo(c4d.UNDOTYPE_CHANGE, op)
          op.SetAbsPos(position)
          op.Message(c4d.MSG_UPDATE)

c4d.DrawViews(c4d.DA_ONLY_ACTIVE_VIEW|c4d.DA_NO_THREAD|c4d.DA_NO_ANIMATION)
          result, dx, dy, channel = win.MouseDrag()

-Nik

[Helper]

On 08/01/2013 at 10:25, xxxxxxxx wrote:

what do you mean with surface snapping

1. snap onto the polygon polygon centers, which is basically just a weighted
point snapping. get the vectors between your snap position and  the points of your
snapping objects, select the three shortest vectors out of that list, get the polygon,
use the points to calc the center.

2. actually snap onto the closest point on a polygon (in a retopo fashion). you would 
have to dig yourself through this. i have actually done this once, but i cannot share
the code but i still have the links i relied on then. you hvae basically do everything like
for no. 1 and then solve the the closest point in triangle problem.

http://www.geometrictools.com/Documentation/DistancePoint3Triangle3.pdf
http://www.gamedev.net/topic/552906-closest-point-on-triangle/
http://www.blackpawn.com/texts/pointinpoly/default.html

edit : ok that is also a way. i am usally not so good at this wrap your head arround math stuff, 
but won't a camera 'normal' based snapping lead to funny results ?  the point which is sitting
perpendicular on the viewplane does not have to be closest point, which at least I would
associate with polygon snapping. i would call it view dependent obstacle snapping or something 
like that or am i missing something here (again) ?

[Helper]

On 08/01/2013 at 10:42, xxxxxxxx wrote:

Hi Ferdinand,

I'm not sure what you mean with "camera 'normal' based snappin lead to funny results". Possibly
I just should get some sleep, but would you mind to elaborate a little bit more on what you are
talking about?

Thanks for the resources, I've always been interested into that topic but these links are new
to me!

-Nik

[Helper]

On 08/01/2013 at 10:48, xxxxxxxx wrote:

Thank you so much for your help, now everything is much more understandable ^_^

Thanks NiklasR  and Thanks littledevil 

[Helper]

On 08/01/2013 at 10:53, xxxxxxxx wrote:

as I already said this might be based on a misconeption of your code,
but this is what is was talking about. (awesome paint pic inc.) :

the red line is what your code will return for my understanding, the green line is
the closest point on surface. at least i would consider the red result as unwanted
or 'funny'.

edit : i hope this does make any sense for someone else than me

[Helper]

On 09/01/2013 at 11:33, xxxxxxxx wrote:

Originally posted by xxxxxxxx

You need to use the c4d.utils.GeRayCollider class to find an intersection on the surface of the
destination-object, intersected by the line defined by the mouse-position in global space and
the camera position.

op = doc.GetActiveObject()
      op_next = op.GetNext()
      if not op_next:
          return True

# Convert op_next to a polygon object.
      # Simplified here, just for demonstration purpose.
      op_next = op_next.GetCache()
      if not op_next or not op_next.CheckType(c4d.Opolygon) :
          print "Cache is not existent or is not a polygon object."
          return True
     
      win.MouseDragStart(button=device, mx=int(mx), my=int(my), flags=c4d.MOUSEDRAGFLAGS_DONTHIDEMOUSE)
      result, dx, dy, channel = win.MouseDrag()
      while result==c4d.MOUSEDRAGRESULT_CONTINUE:
          mx += dx
          my += dy            
 
          cursorpos = bd.SW(c4d.Vector(mx,my,500))     #screen to world conversion
          if(axis == 1) :
              cursorpos.x = 0
          elif(axis == 2) :
              cursorpos.y = 0                          #Constrain drawing along an axis based on the comboButton's value
          elif(axis == 3) :
              cursorpos.z = 0

cam_pos = bd.GetSceneCamera(doc).GetAbsPos()
          collider = c4d.utils.GeRayCollider()
          collider.Init(op_next)

length = 500000
          direction = -(cam_pos - cursorpos).GetNormalized()

did_intersect = collider.Intersect(cam_pos, direction, length)
          if did_intersect:
              position = collider.GetNearestIntersection()['hitpos']
              print "Intersection at", position
          else:
              print "No intersection."
              position = cursorpos

doc.AddUndo(c4d.UNDOTYPE_CHANGE, op)
          op.SetAbsPos(position)
          op.Message(c4d.MSG_UPDATE)

c4d.DrawViews(c4d.DA_ONLY_ACTIVE_VIEW|c4d.DA_NO_THREAD|c4d.DA_NO_ANIMATION)
          result, dx, dy, channel = win.MouseDrag()

-Nik

I again disorder 
I wanted that the active object changed the axis alignment, as the normal to the surfaced intersection?
example:

the current code

my request:

The axis changed by surface normal...

[Helper]

On 09/01/2013 at 11:48, xxxxxxxx wrote:

What about trying it yourself? 
Slight tip: The dictionary returned by GeRayCollider.GetNearestIntersection() contains information
like what polygon was interesected. You can then compute the normal of that polygon and rotate
the destination object accordingly.

@ littledevil :
Well, we don't want the closes point on the surface, do we? I'm sorry, I don't see why the green
line is the one we would like to use instead of the red line.. :// I also don't get how you would
come to that green line, I don't see any relation to the camera position or view-direction or similar.

Hope that doesn't sound offending or so, because that was not my intention. I always
appreciate constructive criticism or questions.
Best,
-Niklas

[Helper]

On 09/01/2013 at 12:13, xxxxxxxx wrote:

NiklasR we are successful!  Thank you for your precious help 

[Helper]

On 12/01/2013 at 03:43, xxxxxxxx wrote:

i have modified the  **op_next  **with Linkbox but... no work...
The object is not recognized...

I've never had much luck with the linkbox 

MY_LINKBOX = 100014

def CreateLayout(self) :
        self.GroupBegin(id=1000, flags=c4d.BFH_SCALEFIT, cols=1, rows=1)
        self.GroupBorderSpace(10, 10, 10, 10)
        self.element = self.AddStaticText(id=1001, flags=c4d.BFH_MASK, initw=120, name="Inserire Oggetto:", borderstyle=c4d.BORDER_NONE)
        self.linkBox = self.AddCustomGui(MY_LINKBOX, c4d.CUSTOMGUI_LINKBOX, "", c4d.BFH_SCALEFIT|c4d.BFV_SCALEFIT, 0, 0)
        self.GroupEnd()
        return True

def MouseInput(self, doc, data, bd, win, msg) :
        mx = msg[c4d.BFM_INPUT_X]
        my = msg[c4d.BFM_INPUT_Y]
 
        device = 0
        if msg[c4d.BFM_INPUT_CHANNEL]==c4d.BFM_INPUT_MOUSELEFT:
            device = c4d.KEY_MLEFT
        elif msg[c4d.BFM_INPUT_CHANNEL]==c4d.BFM_INPUT_MOUSERIGHT:
             device = c4d.KEY_MRIGHT
        else:
            return True

op = doc.GetActiveObject()
        op_next = self.linkBox.GetLink()
        if not op_next:
return True

win.MouseDragStart(button=device, mx=int(mx), my=int(my), flags=c4d.MOUSEDRAGFLAGS_DONTHIDEMOUSE)
        result, dx, dy, channel = win.MouseDrag()
        while result==c4d.MOUSEDRAGRESULT_CONTINUE:
            mx += dx
            my += dy

cursorpos = bd.SW(c4d.Vector(mx,my,500))     #screen to world conversion

cam_pos = bd.GetSceneCamera(doc).GetAbsPos()
            collider = c4d.utils.GeRayCollider()
            collider.Init(op_next)

length = 500000
            direction = -(cam_pos - cursorpos).GetNormalized()

did_intersect = collider.Intersect(cam_pos, direction, length)
            if did_intersect:
                position = collider.GetNearestIntersection()["hitpos"]
                print "Intersection at", position
            else:
                print "No intersection."
                position = cursorpos

doc.AddUndo(c4d.UNDOTYPE_CHANGE, op)
            op.SetAbsPos(position)
            op.Message(c4d.MSG_UPDATE)

c4d.DrawViews(c4d.DA_ONLY_ACTIVE_VIEW|c4d.DA_NO_THREAD|c4d.DA_NO_ANIMATION)
            result, dx, dy, channel = win.MouseDrag()

[Helper]

On 12/01/2013 at 07:22, xxxxxxxx wrote:

try :

op_next = self.linkBox.GetLink(c4d.documents.GetActiveDocument(), c4d.Obase)

[Helper]

On 12/01/2013 at 09:11, xxxxxxxx wrote:

Originally posted by xxxxxxxx

try :

op_next = self.linkBox.GetLink(c4d.documents.GetActiveDocument(), c4d.Obase)

"object has no attribute 'linkBox'"....  
I am beginning to think that linkbox they hate me 

Thanks littledevil, you are a generous person, always ready to help all friends in this forum 

[Helper]

On 12/01/2013 at 10:37, xxxxxxxx wrote:

could be some kind of call order problem. try something like this

class myclass(c4d.Something) :
	def __init__(self) :
		self.linkBox = None
	
	def CreateLayout(self, *args) :
		self.linkBox = Something
	
	def AnotherMethod(self) :
		if (not self.linkBox) :
			return False
		else:
			x = self.linkBox.Do(blah)

MouseInput might be called the first time before CreateLayout has been called and as you 
haven't defined linkBox as a variable in the class constructor it does not yet exist at that point.

[Helper]

On 12/01/2013 at 12:08, xxxxxxxx wrote:

littledevil, i have pasted the entire code. 
You can control it and correct it?
I would like to understand where I'm wrong ...
Thanks for your help 

import os
import sys
import c4d
from c4d import plugins, utils, bitmaps, gui, documents
 
PLUGIN_ID = 10000008 # THIS id IS FOR TESTING PUPOSES ONLY!!!
 
#for GeLoadString values must match with the header file
IDS_ESTENSIONE = 100000
MY_LINKBOX = 100015
  
  
class SettingsDialog(gui.SubDialog) :
 
 
    def CreateLayout(self) :
        self.GroupBegin(id=1000, flags=c4d.BFH_SCALEFIT, cols=1, rows=1)
        self.GroupBorderSpace(10, 10, 10, 10)
        self.element = self.AddStaticText(id=1001, flags=c4d.BFH_MASK, initw=120, name="Inserire Oggetto:", borderstyle=c4d.BORDER_NONE)
 **        self.linkBox = self.AddCustomGui(MY_LINKBOX, c4d.CUSTOMGUI_LINKBOX, "", c4d.BFH_SCALEFIT|c4d.BFV_SCALEFIT, 0, 0)**
  
		
		
        self.GroupEnd()
        return True
    def InitValues(self) :
        
        return True
 
    def Command(self, id, msg) :
 
        return True
 
class Caleidos(plugins.ToolData) :
  
    def __init__(self) :
        self.data = 0
  
    def KeyboardInput(self, doc, data, bd, win, msg) :
        key = msg.GetLong(c4d.BFM_INPUT_CHANNEL)
        cstr = msg.GetString(c4d.BFM_INPUT_ASC)
        if key==c4d.KEY_ESC:
            #do what you want            
            #return True to signal that the key is processed
            return True
        return False
 
 
    def MouseInput(self, doc, data, bd, win, msg) :
	
		if msg.GetLong(c4d.BFM_INPUT_CHANNEL)==c4d.BFM_INPUT_MOUSELEFT:
			print "Begin Mouse Left Pressed"
			while True:
			
				bc = c4d.BaseContainer()
				if gui.GetInputState(c4d.BFM_INPUT_MOUSE, c4d.BFM_INPUT_MOUSELEFT, bc) :
					if bc.GetLong(c4d.BFM_INPUT_CHANNEL)==c4d.BFM_INPUT_MOUSELEFT:
						print "Mouse Left Pressed"
			
						mx = msg[c4d.BFM_INPUT_X]
						my = msg[c4d.BFM_INPUT_Y]
				 
						device = 0
						if msg[c4d.BFM_INPUT_CHANNEL]==c4d.BFM_INPUT_MOUSELEFT:
							device = c4d.KEY_MLEFT
						elif msg[c4d.BFM_INPUT_CHANNEL]==c4d.BFM_INPUT_MOUSERIGHT:
							 device = c4d.KEY_MRIGHT
						else:
							return True
  
							
						doc.GetActiveObject()
						doc.StartUndo() 
						myobject = **self.linkBox.GetLink()**
  
  
						op = myobject
						op_next = doc.GetActiveObject()
						
						win.MouseDragStart(button=device, mx=int(mx), my=int(my), flags=c4d.MOUSEDRAGFLAGS_DONTHIDEMOUSE)
						result, dx, dy, channel = win.MouseDrag()
						while result==c4d.MOUSEDRAGRESULT_CONTINUE:
							mx += dx
							my += dy            
  
							cursorpos = bd.SW(c4d.Vector(mx,my,500))     #screen to world conversion
							cam_pos = bd.GetSceneCamera(doc).GetAbsPos()
							collider = c4d.utils.GeRayCollider()
							collider.Init(op_next)
							length = 500000
							direction = -(cam_pos - cursorpos).GetNormalized()
							did_intersect = collider.Intersect(cam_pos, direction, length)
							if did_intersect:
								position = collider.GetNearestIntersection()["hitpos"]
								print "Intersection at", position
							else:
								print "No intersection."
								position = cursorpos
							doc.AddUndo(c4d.UNDOTYPE_CHANGE, op)
							op.SetAbsPos(position)
							op.Message(c4d.MSG_UPDATE)  
							c4d.DrawViews(c4d.DA_ONLY_ACTIVE_VIEW|c4d.DA_NO_THREAD|c4d.DA_NO_ANIMATION)
							result, dx, dy, channel = win.MouseDrag()
   						if not bc.GetBool(c4d.BFM_INPUT_VALUE) : break
			print "End Mouse Left Pressed"
    def AllocSubDialog(self, bc) :
        return SettingsDialog(self.data) #always return new instance(self.data)
 
 
  
if __name__ == "__main__":
    bmp = bitmaps.BaseBitmap()
    dir, file = os.path.split(__file__)
    fn = os.path.join(dir, "res", "Icon.tif")
    bmp.InitWith(fn)
    plugins.RegisterToolPlugin(id=PLUGIN_ID, str="Caleidos",info=0, icon=bmp, help="Statusbar Text",dat=Caleidos())
  

[Helper]

On 12/01/2013 at 13:03, xxxxxxxx wrote:

this obviously won't work. i think you do not really understand what self means in python.
it is similar to this in c# or java. it is a referecne to the actual instance of the class from which
the code is being executed from. you are doing more or less this :

class clsA() :
	def __init__(self, x) :
		self.X = x
  
class clsB() :
	def someMethod() :
		# raises a runtime error as class b does not have a memeber called X
		print self.X

you have either make your dialog class instance a globally available variable (but this
could be considered as really bad style) :

class clsA() :
    def __init__(self, x) :
        self.X = x
  
class clsB() :
    def do (self, classAinstance) :
        print classAinstance.X
       
def main() :
    A = clsA()
    B = clsB()
    B.do(A)

or you have to make an instance of your GeDialog class a memeber of your
ToolData class.

class ToolDataClass() :
    def __init__(self) :
        self.dialog = None
        self.data = None
    
    def doSomething(self) :
        if (self.dialog) :
            print self.Dialog.linkBox.GetLink(document, c4d.Obase)
        else:
            print 'dialog has not been raised yet.'
        
    def AllocSubDialog(self, bc) :
        self.Dialog = SettingsDialog(self.data)
        return self.Dialog

edit : fixed a typo,  should be if (self.dialog) : of course and not if(self.data) :

[Helper]

On 12/01/2013 at 14:28, xxxxxxxx wrote:

Originally posted by xxxxxxxx

or you have to make an instance of your GeDialog class a memeber of your
ToolData class.

class ToolDataClass() :
    def __init__(self) :
        self.dialog = None
        self.data = None
    
    def doSomething(self) :
        if (self.dialog) :
            print self.Dialog.linkBox.GetLink(document, c4d.Obase)
        else:
            print 'dialog has not been raised yet.'
        
    def AllocSubDialog(self, bc) :
        self.Dialog = SettingsDialog(self.data)
        return self.Dialog

edit : fixed a typo,  should be if (self.dialog) : of course and not if(self.data) :

Work!!! I succeeded! littledevil Thanks!!!!!!! 

[Helper]

On 19/07/2013 at 09:28, xxxxxxxx wrote:

Originally posted by xxxxxxxx

You need to use the c4d.utils.GeRayCollider class to find an intersection on the surface of the
destination-object, intersected by the line defined by the mouse-position in global space and
the camera position.

op = doc.GetActiveObject()
      op_next = op.GetNext()
      if not op_next:
          return True

# Convert op_next to a polygon object.
      # Simplified here, just for demonstration purpose.
      op_next = op_next.GetCache()
      if not op_next or not op_next.CheckType(c4d.Opolygon) :
          print "Cache is not existent or is not a polygon object."
          return True
     
      win.MouseDragStart(button=device, mx=int(mx), my=int(my), flags=c4d.MOUSEDRAGFLAGS_DONTHIDEMOUSE)
      result, dx, dy, channel = win.MouseDrag()
      while result==c4d.MOUSEDRAGRESULT_CONTINUE:
          mx += dx
          my += dy            
 
          cursorpos = bd.SW(c4d.Vector(mx,my,500))     #screen to world conversion
          if(axis == 1) :
              cursorpos.x = 0
          elif(axis == 2) :
              cursorpos.y = 0                          #Constrain drawing along an axis based on the comboButton's value
          elif(axis == 3) :
              cursorpos.z = 0

cam_pos = bd.GetSceneCamera(doc).GetAbsPos()
          collider = c4d.utils.GeRayCollider()
          collider.Init(op_next)

length = 500000
          direction = -(cam_pos - cursorpos).GetNormalized()

did_intersect = collider.Intersect(cam_pos, direction, length)
          if did_intersect:
              position = collider.GetNearestIntersection()['hitpos']
              print "Intersection at", position
          else:
              print "No intersection."
              position = cursorpos

doc.AddUndo(c4d.UNDOTYPE_CHANGE, op)
          op.SetAbsPos(position)
          op.Message(c4d.MSG_UPDATE)

c4d.DrawViews(c4d.DA_ONLY_ACTIVE_VIEW|c4d.DA_NO_THREAD|c4d.DA_NO_ANIMATION)
          result, dx, dy, channel = win.MouseDrag()

-Nik

an example for snap on point object? thanks