TRANS for rotation angle?

Good morning,

I am looking for help with a TRANS for a rotation angle. I looked at points
conversion in help, and this is way over my head.

I'm copying blocks using insert from one ucs to another, and placing them on
a different layer in the new ucs. Everything is working except for the
rotation angle. A hardcoded formula doesn't work, as I can't predict new
ucs.

Any help here would be greatly appreciated.
Thanks,
Mark

 

Search the newsgroup for a 2 or 3 weeks ago, it was discussed.

9-14-04 "Trans or other"

 

James, thank you for the reply.

I am taking blocks from WCS to a known, named UCS. I am able to insert the
blocks in the known UCS in a correct position and on the new layer, just
don't know how to handle the rotation. I understand the dxf is radians, but
point matrix is beyond me.

I don't need 3d orientation, all insert will be at 0 z.

I looked for the posts you mentioned, and had no luck - using OE6, not web
access. I will look at vla-transformby, but again, a point matrix melts my
brain.

BTW, I need to re-insert the block, not just copy it, as I use a different
block.

Any help getting the new rotation would be great.

Thanks again,
Mark

 

Thanks Rudy, but I didn't see anything providing an angle, or I don't know
what I'm looking at.

Mark

 

Sorry, I can't see it. Could you repost attachment here?

Thank you,
Mark

 

How about if we take this back to basics?

The rotaion angle in the WCS is the same as the rotation angle in the UCS.

Unless for some reason you want to have your inserts different than the ones on the WCS.


Bill

 

I guess in short...

<in degrees or radians, I don't care>
(functionname rotation_from_WCS) = returns rotation in current/known UCS

or

(functionname INS_from_WCS rotation_from_WCS) = returns INSertion point &
rotation in current/known UCS as list

Sorry if I am unclear, I'm trying
Mark

 

Maybe this will help:

You may be able to modify it.

; == Function WvpVec2Ang
; Converts a vector to an angle.
; Arguments [Type]:
; Vec = Vector value

• 
  ; Return [Type]:
  ; > Angle value [REAL]
  ; Notes:
  ; Credits to Luis Esquivel
  ;
  (defun Vec2Ang (Vec / XvcVal YvcVal) ;get angle of each segment of pline at vertex.
  (setq XvcVal (car Vec)
  YvcVal (cadr Vec)
  )
  (cond ((equal XvcVal 0 1E-4)
  (if (minusp YvcVal)(* pi 1.5)(* pi 0.5))
  )
  ((equal XvcVal -1 1E-4) pi)
  ((equal YvcVal 0 1E-4)(if (minusp XvcVal) pi 0)
  )
  (T (atan YvcVal XvcVal))
  )
  ) ;defun
  
  ;;
  ;;2d LWPOLYLINES
  ;;Usage (Insert_@_point_lst)
  
  (defun c:Insert_@_point_list (/ block_name CurPar InsAng rslt)
  (vl-load-com)
  (setvar "cmdecho" 0)
  (setq *acad* (vlax-get-acad-object)
  *acad_doc* (vla-get-activedocument *acad*)
  *mspace* (vla-get-modelspace *acad_doc*)
  )
  (initget 1)
  (setq plineObj (vlax-ename->vla-object (car (entsel "Pick a 2D Polyline: <> ")))
  )
  (setq block_name (getstring "\nEnter block name: <> ")
  )
  (setq vlist ;get list of xyz reals from polyline.
  (vlax-safearray->list
  (vlax-variant-value
  (vla-get-Coordinates
  plineObj
  )
  )
  )
  )
  ;---; ;convert to list of 2d points.
  (while
  (setq rslt (cons (list (car vlist)(cadr vlist)) rslt)
  vlist (cddr vlist)
  )
  )
  ;---;
  (if (tblsearch "BLOCK" block_name) ;insert block
  (foreach Pnt (reverse rslt)
  (setq CurPar (vlax-curve-getParamAtPoint plineObj Pnt)
  InsAng (Vec2Ang (vlax-curve-getFirstDeriv plineObj CurPar))
  )
  (vla-InsertBlock *mspace* (vlax-3d-point Pnt) block_name 1 1 1 InsAng)
  )
  
  (prompt "\nBlock name not found.")
  )
  (princ)
  )
  
  HTH
  
  Gotta go now.
  
  Bill

 

Great question. The 'other' ucs would be maybe opposite hand / maybe rotated
/ maybe flipped floor plan. This would make the rotation angle opposite in
whatever way to the WCS.

The objective is to have duplicated annotation in the same plan but
different UCS orientation and different layers.

The scenario of the new UCS is not predefined by the system, but created by
the user. The name of the new UCS is known, and consistent, and current
during this process.

Thanks,
Mark

 

Thanks Bill, I'll check it out.

Mark

 

Thanks Bill, I'll check it out.

Mark

 

I may not be able to post for a few hours, please don't take my lack of
response as disrespect for your help, I am grateful.

Thank you very much for your help,
Mark

 

Hi Mark.

Using vlisp methods, it sounds like all you need for each insert is a
vla-copy and vla-transformby. Two operations per insert, that's the bottom
line. Should you choose to accept that mission, the only trick is getting
the transformation matrix to feed to the transformby method.

The key question to answer is what defines the two coordinate systems. Are
they two saved ucs's, one wcs and one ucs, ocs to ucs? For each insert,
what cs are we going from and to? I wasn't sure from your post. Also, how
powerful do you want the cs change to be? Are you changing just rotation,
position as well, or even non-uniform scaling and 3d orientation. All of
this is possible, and again I'm not sure.

With that information in hand you can do just what you want. It is possible
that you can do it without matrices or transformby at all, but depending on
exactly what you are doing, it might actually be more work IMO.

I'll be watching for your response. In the meantime, if you want to study
some example code you can look at some of my recent posts. Find my
"BlockSwap.lsp" and "rcb2.lsp" attachments (5Oct & 10Oct) in the recent
thread "how to change the insertpoint of a block,and keep the insert
objects's place". Or I know John Uhden and Joe Burke (probably others as
well) have some good code out there as well. Perhaps none of it will
exactly fit your need, but under the hood it's all fundamentally the same
stuff.

 

Hi Rudy.

I wish I hadn't missed that thread you referred to.

Rudy wrote> "Thanks Luis... but it would be a killer utility to incorporate
all the other factors."

I think you may be interested in "MWE:BlockTrans" in "rcb2.lsp" that I
referred Mark to. And perhaps some more of that very long thread.

 

Mark, I think Bill's got you going in the right direction. Thank you Bill.
(We see where my mind's been lately.)

Another tip: (angle '(0 0 0) (trans '(1 0 0) 1 0 t)) returns the rotation of
the current ucs in radians. Notice that '(1 0 0) is a vector that defines
the x-axis, or 0d rotation. So we translate the vector and find the
resulting angle.

When you say things like opposite hand and flipped, I'm still not sure that
ALL we're talking about is position and rotation, but I'll hold off on that
for now.

James

 

That may be what I was looking for... off to test.
Thanks
Mark

 

You're welcome.

James