[Maxima-bugs] [ maxima-Bugs-1629043 ] assume: < doesn't imply not-equal

[SourceForge.net <no...@so...>]

Bugs item #1629043, was opened at 2007-01-05 15:13
Message generated for change (Comment added) made by dgildea
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=104933&aid=1629043&group_id=4933

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: Lisp Core - Assume
Group: None
>Status: Closed
>Resolution: Works For Me
Priority: 5
Private: No
Submitted By: Stavros Macrakis (macrakis)
Assigned to: Nobody/Anonymous (nobody)
Summary: assume: < doesn't imply not-equal

Initial Comment:
assume(r<1)$
is(equal(r,1)) => unknown (???)

Maxima 5.11.0 GCL 2.6.8 Windows 2k


----------------------------------------------------------------------

>Comment By: Dan Gildea (dgildea)
Date: 2007-12-30 09:04

Message:
Logged In: YES 
user_id=1797506
Originator: NO

Works in current cvs.

(%i1) assume(r<1)$
(%i2) is(equal(r,1));
(%o2)                                false
(%i3)  is(equal(r,2));
(%o3)                                false
(%i4)  is(equal(r,0));
(%o4)                               unknown



----------------------------------------------------------------------

Comment By: Barton Willis (willisbl)
Date: 2007-04-05 23:56

Message:
Logged In: YES 
user_id=895922
Originator: NO

Appending (simplifya e nil) might not be the right thing to do, but it
does
eliminate this bug:

(defun equal-facts-simp (e)
  (let ((f (margs ($facts))))
    (dolist (fi f (simplifya e nil))
      (if (op-equalp fi '$equal)
	  (setq e ($ratsubst (nth 2 fi) (nth 1 fi) e))))))


----------------------------------------------------------------------

Comment By: Barton Willis (willisbl)
Date: 2007-04-05 23:39

Message:
Logged In: YES 
user_id=895922
Originator: NO

What's the story:

(%i1) assume(r<1);
(%o1) [r<1]
(%i2) :lisp(trace meqp csign);
(MEQP CSIGN)
(%i2) is(equal(x,1)),prederror : false;
  1> (MEQP $X 1)
    2> (CSIGN ((MPLUS RATSIMP) $X -1)) <-- the argument to csign looks
specious ?
    <2 (CSIGN $PNZ)  <-- Yikes!
  <1 (MEQP (($EQUAL) $X 1))
(%o2) unknown

(%i8) ?csign(r-1);
  1> (CSIGN ((MPLUS SIMP) -1 $R))
  <1 (CSIGN $NEG)  
(%o8) neg  <-- OK


----------------------------------------------------------------------

Comment By: Stavros Macrakis (macrakis)
Date: 2007-01-05 15:19

Message:
Logged In: YES 
user_id=588346
Originator: YES

assume(r<1)$
is(equal(r,1)) => unknown (???)
is(equal(r,2)) => unknown (???)

On the other hand,

assume(s<t)$
is(equal(s,t)) => false (OK)
is(equal(s,t+1)) => unknown
       (oops, but probably not the same bug)

assume(equal(q,1))$
is(q<1) => false (OK)
is(q<=1) => true (OK)

Maxima 5.11.0 GCL 2.6.8 Windows 2k

----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=104933&aid=1629043&group_id=4933