Topic: LISP Challenge: Ducks in a row

[pkohut]

Ah, mine has a side effect!  The test functions variables are being used in the IndexPoints
routines.  So the change to your test suite caught the bug.  :ugly:  Still the 3D test
and duplicate points run will fail.

Paul

[gile]

My correction too, an a little optimisation (2 lists instead of a dotted pair list)

GC = Gilles Chanteau

Code: [Select]

(defun GC:IndexPoints (control_points test_points tolerance / pl pt l1 l2)
  (vl-load-com)
  (setq pl
(entmakex
   (append
     (list
       '(0 . "lWPOLYLINE")
       '(100 . "AcDbEntity")
       '(100 . "AcDbPolyline")
       '(70 . 0)
       (cons 90 (length control_points))
     )
     (mapcar '(lambda (x) (cons 10 x)) control_points)
   )
)
  )
  (foreach p (reverse test_points)
    (setq pt (vlax-curve-getClosestPointTo pl p)
  l1 (cons (vlax-curve-getParamAtPoint pl pt) l1)
  l2 (cons (distance pt (list (car p) (cadr p))) l2)
    )
  )
  (entdel pl)
  (vl-remove-if
    '(lambda (x) (< tolerance (nth x l2)))
    (vl-sort-i l1 '<)
  )
)

[dgorsman]

While I don't have anything to contribute (on pseudo-holiday at the moment), this could be potentially quite useful for some of my users.  Thank you all for the ideas, I will definitely be keeping an eye on this.

Any plans for testing this with point cloud data?

[MP]

Just to cap this thread ... Giles' and Evgeniy's last entries ...

• showed performance increases, tho in the same magnitude as prior efforts
• pass the 2D and 3D tests
• fail the duplicates test

For my own use functions based upon entity creation/deletion are, while impressive performance wise, not usable for production work.

All in all a pretty fun exercise, I thank you all for your spirited participation.

Lots of functioning gray matter herein!

[gile]

• fail the duplicates test

I don't agree.

_$ (MP:IndexPoints control_points (append test_points test_points) 1.0)
(26 2 38 14 46 22 3 27 4 28 9 33 39 15 31 7 37 13 17 41 11 35 1 25 34 10 45 21 47 23)
_$ (GC:IndexPoints control_points (append test_points test_points) 1.0)
(26 2 38 14 46 22 27 3 28 4 33 9 39 15 31 7 37 13 41 17 35 11 25 1 34 10 45 21 47 23)
_$ (EE:IndexPoints-2 control_points (append test_points test_points) 1.0)
(26 2 38 14 46 22 27 3 28 4 33 9 39 15 31 7 37 13 41 17 35 11 25 1 34 10 45 21 47 23)
_$ (EE:IndexPoints-1 control_points (append test_points test_points) 1.0)
(26 2 38 14 46 22 27 3 28 4 33 9 39 15 31 7 37 13 41 17 35 11 25 1 34 10 45 21 47 23)

The inversions as 3 27 vs 27 3, 4 28 vs 28 4, 9 33 vs 33 9 and so on, always refer to the same overlapped points (a 24 difference).

[MP]

Don't know what to tell you Gile, when I ran my tests (which cared not if a pair was inverted, only that the pairs were in the proper sequence) it this is what I got:

NO DUPLICATES SORT
(eval (MP:INDEXPOINTS CONTROL_POINTS TEST_POINTS TOLERANCE)):
(02 14 22 03 04 09 15 07 13 17 11 01 10 21 23)

DUPLICATES SORT
(eval (E1:INDEXPOINTS CONTROL_POINTS TEST_POINTS TOLERANCE)): FAIL
(26/02  38/14  46/22  28/04  33/09  39/15  31/07 37/13  ...)

(eval (E2:INDEXPOINTS CONTROL_POINTS TEST_POINTS TOLERANCE)): FAIL
(26/02  38/14  46/22 28/04  33/09  39/15 31/07  37/13 ...)

(eval (GC:INDEXPOINTS CONTROL_POINTS TEST_POINTS TOLERANCE)): FAIL
(26/02  38/14  46/22  28/04  33/09  39/15  31/07  37/13 ...)

(eval (MP:INDEXPOINTS CONTROL_POINTS TEST_POINTS TOLERANCE)):
(26/02  38/14  46/22  03/27  04/28  09/33  39/15  31/07  ...)
 
{shrug}

I'll run fresh tests later on; gotta gun to me head right now.

[ElpanovEvgeniy]

DUPLICATES SORT
(eval (E1:INDEXPOINTS CONTROL_POINTS TEST_POINTS TOLERANCE)): FAIL
(26/02  38/14  46/22  28/04  33/09  39/15  31/07 37/13  ...)

(eval (E2:INDEXPOINTS CONTROL_POINTS TEST_POINTS TOLERANCE)): FAIL
(26/02  38/14  46/22 28/04  33/09  39/15 31/07  37/13 ...)

I have received other results!

Code: [Select]

(EE:INDEXPOINTS-1 a b c)
(26 2 38 14 46 22 27 3 28 4 33 9 39 15 31 7 37 13 41 17 35 11 25 1 34 10 45 21 47 23)
(EE:INDEXPOINTS-2 a b c)
(26 2 38 14 46 22 27 3 28 4 33 9 39 15 31 7 37 13 41 17 35 11 25 1 34 10 45 21 47 23)
(GC:IndexPoints a b c)
(26 2 38 14 46 22 27 3 28 4 33 9 39 15 31 7 37 13 41 17 35 11 25 1 34 10 45 21 47 23)
(MP:IndexPoints a b c)
(26 2 38 14 46 22 3 27 4 28 9 33 39 15 31 7 37 13 17 41 11 35 1 25 34 10 45 21 47 23)


[MP]

Weird, I know I used the most current versions of all your submissions. If you're both getting some thing different than me I must be in err, so my apologies. I'll rerun anew ASAP. Forgive me / thanks for your patience.