Re: [Visualpython-users] strange math

[Robert X. <nne...@gm...>]

Vector objects are mutable, which means they get passed by reference instead
of by value. It also means that statements like "a=b" do not copy the value,
but they copy the reference (making a and b point to the same vector). So,

a = vector(1,2,3)
b = a
b.x = 3
print a

yields
<3, 2, 3>

as expected, because a and b refer to the same vector. The same thing
happens with list objects in Python.

Numbers, tuples and strings are immutable in Python, so they are passed by
value.

Robert

2009/9/12 Lenore Horner <lh...@si...>

> (Sorry, forgot to cc the list.)
>
> Hi Steve,
>
> You're saying the following?
> a = 1
> b = a
> b = 3
> Now print a would give 3?
>
> I could almost understand if the following were true.
> air.acceleration = gravity
> gravity = gravity + (1,0,1)
> print acceleration gives (1,-9.8,1)
>
> That would mean the assignment was somehow permanent, which is to my
> mind quite weird but potentially useful.
>
> But it sounds like you're saying that assignment in python makes two
> variables identical rather than giving the one on the left the current
> value of the one on the right?  That seems to me to be incredibly
> awkward.  How is one supposed to initialize several things identically
> (no possibility of typos or slider misadjustments) and then let them
> evolve differently?
>
> -----
> Testing:
> Nope, that's not true.
>  >>> a = 1
>  >>> print a
> 1
>  >>> b = a
>  >>> print a, b
> 1 1
>  >>> b = 3
>  >>> print a, b
> 1 3
>
> Well maybe vectors behave differently than scalars.
> Nope.
>  >>> a = (0,0,1)
>  >>> b = a
>  >>> print b
> (0, 0, 1)
>  >>> b = (1,0,0)
>  >>> print a, b
> (0, 0, 1) (1, 0, 0)
>
> Maybe a vector is different than a tuple.
> Nope.
> from visual import *
>  >>> a = vector(0,0,1)
>  >>> b = a
>  >>> b = vector(1,0,0)
>  >>> print a, b
> <0, 0, 1> <1, 0, 0>
>
> But the above isn't quite what I'm doing in my code.  Maybe component
> assignment works differently than all other types of assignment.
> Yup!
>  >>> a = vector(0,0,1)
>  >>> b = a
>  >>> c = a
>  >>> print a, b, c
> <0, 0, 1> <0, 0, 1> <0, 0, 1>
>  >>> c.y = 5
>  >>> print a, b, c
> <0, 5, 1> <0, 5, 1> <0, 5, 1>
>
> Ok, that's just plain spooky.
>
> Can someone give me an example where that would be helpful?
>
> Now I have to go figure out how to dodge this bullet.
>
> Lenore
>
>
> On Sep 12, 2009, at 06:36 , Steve Spicklemire wrote:
>
> > Hi Lenore,
> >
> > It's about python assignments. You are assigning the *same*
> > acceleration vector to both objects, then changing the y component
> > of the same vector.
> No.  I did not change the y component of the vector which was assigned
> to both of them.
> >
> > Instead of assigning both accelerations to the same vector (gravity)
> > try initializing them with different vectors:
> >
> > pixie.acceleration = vector(0,-9.8,0)
> > air.acceleration = vector(0,-9.8,0)
> >
> > then when you fiddle with the y components, it will be the y
> > components of different vector objects.
> >
> > That should work!
> > -steve
> >
> > On Sep 12, 2009, at 7:26 AM, Lenore Horner wrote:
> >
> >> Sorry.  I was in too big a hurry and forgot to paste the code.  Here
> >> it is.  I've put arrows off to the right indicating the lines where
> >> both air.acceleration and pixie.acceleration are being changed
> >> despite
> >> the fact that only air.acceleration appears in the code.  I've
> >> included all of the while loop.  I don't see anywhere that I'm
> >> reassigning pixie.acceleration from its original value of gravity but
> >> it changes as soon as it hits those to statements (and not before).
> >>
> >> Thanks,
> >> Lenore
> >>
> >>
> >> angle = math.pi/4
> >> speed = 13.0
> >>
> >> power = 2
> >> constant = 1
> >>
> >> gravity = vector(0,-9.8,0)
> >>
> >> launch_velocity =
> >> vector(speed*math.cos(angle),speed*math.sin(angle),0)
> >>
> >> pixie = sphere(pos=vector(-9,-3.55,0), radius=0.2, color=color.red)
> >> pixie.velocity = launch_velocity
> >> pixie.acceleration = gravity
> >>
> >> air = sphere(pos=vector(-9,-3.55,0.3), radius=0.15, color=color.blue)
> >> air.velocity = launch_velocity
> >> air.acceleration = gravity
> >>
> >> scene.mouse.getclick()    # Don't launch until we're watching
> >>
> >> dt = 0.01
> >> while 1:
> >>    rate(100)
> >>    air.pos = air.pos + air.velocity*dt
> >>    if air.y < -3.55:
> >>        air.velocity = vector(0,0,0)
> >>    else:
> >>        air.velocity = air.velocity + air.acceleration*dt
> >>        air.acceleration.y = gravity.y - sign(air.velocity.y) *
> >> constant * air.radius * air.velocity.y**power    <-----
> >>        print  "5", air.acceleration, pixie.acceleration
> >>        air.acceleration.x = -sign(air.velocity.x) * constant *
> >> air.radius * air.velocity.x**power              <-------
> >>        print "6", air.acceleration, pixie.acceleration
> >>
> >>    pixie.pos = pixie.pos + pixie.velocity*dt
> >>    if pixie.y < -3.55:
> >>        pixie.velocity = vector(0,0,0)
> >>    else:
> >>        pixie.velocity = pixie.velocity + pixie.acceleration*dt
> >>
> >>
> >> On Sep 11, 2009, at 18:20 , Bruce Sherwood wrote:
> >>
> >>> Can you explain the context of your question? I have no idea what
> >>> program or
> >>> documentation you're referring to.
> >>>
> >>> Bruce Sherwood
> >>>
> >>> Lenore Horner wrote:
> >>>> Perhaps I'm just blind, but I can't figure out why the assignment
> >>>> statements above print "5" and print "6" affect both acceleration
> >>>> vectors, but they do.  How do I change the acceleration vector of
> >>>> one
> >>>> object without touching the vector of other objects?
> >>>>
> >>>> Thanks,
> >>>> Lenore
> >>>
> >>>
> ------------------------------------------------------------------------------
> >>> Let Crystal Reports handle the reporting - Free Crystal Reports 2008
> >>> 30-Day
> >>> trial. Simplify your report design, integration and deployment - and
> >>> focus on
> >>> what you do best, core application coding. Discover what's new with
> >>> Crystal Reports now.  http://p.sf.net/sfu/bobj-july
> >>> _______________________________________________
> >>> Visualpython-users mailing list
> >>> Vis...@li...
> >>> https://lists.sourceforge.net/lists/listinfo/visualpython-users
> >>
> >>
> >>
> ------------------------------------------------------------------------------
> >> Let Crystal Reports handle the reporting - Free Crystal Reports
> >> 2008 30-Day
> >> trial. Simplify your report design, integration and deployment -
> >> and focus on
> >> what you do best, core application coding. Discover what's new with
> >> Crystal Reports now.  http://p.sf.net/sfu/bobj-july
> >> _______________________________________________
> >> Visualpython-users mailing list
> >> Vis...@li...
> >> https://lists.sourceforge.net/lists/listinfo/visualpython-users
> >
>
>
>
> ------------------------------------------------------------------------------
> Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day
> trial. Simplify your report design, integration and deployment - and focus
> on
> what you do best, core application coding. Discover what's new with
> Crystal Reports now.  http://p.sf.net/sfu/bobj-july
> _______________________________________________
> Visualpython-users mailing list
> Vis...@li...
> https://lists.sourceforge.net/lists/listinfo/visualpython-users
>