| Attempt to call undefined method: 'IDLITSYMBOL::IS3D' [message #82400] |
Wed, 12 December 2012 09:06  |
![Paul Van Delst[1] is currently offline](theme/default/images/xoffline.png.pagespeed.ic.XRkd1fkXye.png "Paul Van Delst[1] is currently offline")
Paul Van Delst[1]
Messages: 1157
Registered: April 2002
|
Senior Member |
|
|
Hello,
I'm creating a graphics window and storing the id in a hash like so:
gref[tag] = WINDOW( WINDOW_TITLE=tag, $
DIMENSIONS = [600,800] )
I then do a bunch of plotting and legend-ing in that window using the
/CURRENT keyword.
The "gref" variable is passed out of my procedure so that, if I decide
to modify the plots (e.g. zoom, add arrows, text, whatever)
interactively, I can still save it. I've done this in the past and it
has always worked fine.
Now when I try to do it I get an error:
IDL> gref['afn'].save,'afn.png',height=800
% SAVE: Attempt to call undefined method: 'IDLITSYMBOL::IS3D'.
% Execution halted at: $MAIN$
Has anyone seen this error before? Or explain it to me? I can find
anything in the IDL help about either the object (IDLITSYMBOL) or the
method (IS3D)
Thanks for any info.
cheers,
paulv
p.s. IDL> print, !version
{ x86 linux unix linux 8.2 Apr 10 2012 32 64}
|
|
|
|
| Re: Attempt to call undefined method: 'IDLITSYMBOL::IS3D' [message #82464 is a reply to message #82400] |
Fri, 14 December 2012 08:54  |
Mark Piper
Messages: 198
Registered: December 2009
|
Senior Member |
|
|
On Thursday, December 13, 2012 7:23:33 AM UTC-7, Paul van Delst wrote:
>
> Re: the tickmark issue:
>
>
>
> In FG when I use [XY]ticklen = 1.0 and [XY]gridstyle=1 *all* the
>
> tickmarks are gridded, both major and minor. Try:
>
>
>
> IDL> p=plot(/test,layout=[1,3,1],$ &
>
> IDL> yticklen=1.0,ygridstyle=1,$ &
>
> IDL> xticklen=1.0,xgridstyle=1)
>
>
>
> My result:
>
> http://ftp.emc.ncep.noaa.gov/jcsda/CRTM/.plots/all_ticks_gri dded.png
>
>
>
> In DG only the major tickmarks were affected -- which is how the
>
> documentation describes the behaviour for both FG and DG.
>
>
>
> As you can imagine, having all the tickmarks gridded makes for a rather
>
> busy plot.
>
>
>
> Upon further reading of the docs I saw the [XY]subticklen keyword. I
>
> gave that a try, setting it to 0.05. But the minor tickmarks are *still*
>
> subject to the [XY]gridstyle keyword (I tend to use [XY]gridstyle=1,
>
> dotted, for this).
>
>
>
> Try:
>
> IDL> p=plot(/test,layout=[1,3,1],$ &
>
> IDL> yticklen=1.0,ygridstyle=1,ysubticklen=0.05,$ &
>
> IDL> xticklen=1.0,xgridstyle=1,xsubticklen=0.05)
>
>
>
> My result:
>
> ftp://ftp.emc.ncep.noaa.gov/jcsda/CRTM/.plots/using_subtickl en.png
>
>
>
> So, as you can see, what I have are uneven length minor tickmarks. And
>
> they are still dotted due the gridstyle keyword.
>
>
>
> What I want is the equivalent of the following DG:
>
> IDL> !p,multi=[0,1,3]
>
> IDL> plot, lindgen(100),$ &
>
> IDL> yticklen=1.0,ygridstyle=1,$ &
>
> IDL> xticklen=1.0,xgridstyle=1
>
>
>
> (Actually, even there I would like the x- and y minor tickmarks to be
>
> the same *absolute*, not relative, length. But, that's a quibble.)
>
>
>
> The FG documentation clearly states that [XY]ticklen = 1.0 is for the
>
> *major* tickmarks only - so that behaviour is a definite bug.
>
>
>
> The [XY]gridstyle docs do not explicitly state that only the major ticks
>
> are affected, but that was the behaviour with DG so I expect that to be
>
> the behaviour for FG.
>
>
>
> You might want to consider adding some examples with these keywords set
>
> to the "Plot Examples" section of the docs. Maybe I am doing something
>
> wrong in FG-space, but after reading the docs I have nothing but my
>
> DG-experience to go by.
>
>
>
> cheers,
>
>
>
> paulv
>
Hi Paul,
I'll log this tick behavior as a bug. For a temporary (but not really convenient) fix, try using CURRENT to overlay a second plot with the same data ranges:
q1 = plot(/test,$
yticklen=1.0, ygridstyle=1, ysubticklen=0.0, $
xticklen=1.0, xgridstyle=1, xsubticklen=0.0)
q2 = plot(/test, /nodata, /current)
Also, whenever you encounter a bug, please send it to Tech Support (support at exelisvis dot com); it's the best way to get this info to us.
mp
|
|
|
|
| Re: Attempt to call undefined method: 'IDLITSYMBOL::IS3D' [message #82481 is a reply to message #82400] |
Thu, 13 December 2012 06:23 |
Paul Van Delst[1]
Messages: 1157
Registered: April 2002
|
Senior Member |
|
|
Hi Chris,
Re: IDLITSYMBOL::IS3D error - I will try and come up with a small
reproducible case.
Re: the tickmark issue:
In FG when I use [XY]ticklen = 1.0 and [XY]gridstyle=1 *all* the
tickmarks are gridded, both major and minor. Try:
IDL> p=plot(/test,layout=[1,3,1],$ &
IDL> yticklen=1.0,ygridstyle=1,$ &
IDL> xticklen=1.0,xgridstyle=1)
My result:
http://ftp.emc.ncep.noaa.gov/jcsda/CRTM/.plots/all_ticks_gri dded.png
In DG only the major tickmarks were affected -- which is how the
documentation describes the behaviour for both FG and DG.
As you can imagine, having all the tickmarks gridded makes for a rather
busy plot.
Upon further reading of the docs I saw the [XY]subticklen keyword. I
gave that a try, setting it to 0.05. But the minor tickmarks are *still*
subject to the [XY]gridstyle keyword (I tend to use [XY]gridstyle=1,
dotted, for this).
Try:
IDL> p=plot(/test,layout=[1,3,1],$ &
IDL> yticklen=1.0,ygridstyle=1,ysubticklen=0.05,$ &
IDL> xticklen=1.0,xgridstyle=1,xsubticklen=0.05)
My result:
ftp://ftp.emc.ncep.noaa.gov/jcsda/CRTM/.plots/using_subtickl en.png
So, as you can see, what I have are uneven length minor tickmarks. And
they are still dotted due the gridstyle keyword.
What I want is the equivalent of the following DG:
IDL> !p,multi=[0,1,3]
IDL> plot, lindgen(100),$ &
IDL> yticklen=1.0,ygridstyle=1,$ &
IDL> xticklen=1.0,xgridstyle=1
(Actually, even there I would like the x- and y minor tickmarks to be
the same *absolute*, not relative, length. But, that's a quibble.)
The FG documentation clearly states that [XY]ticklen = 1.0 is for the
*major* tickmarks only - so that behaviour is a definite bug.
The [XY]gridstyle docs do not explicitly state that only the major ticks
are affected, but that was the behaviour with DG so I expect that to be
the behaviour for FG.
You might want to consider adding some examples with these keywords set
to the "Plot Examples" section of the docs. Maybe I am doing something
wrong in FG-space, but after reading the docs I have nothing but my
DG-experience to go by.
cheers,
paulv
On 12/12/12 22:14, Chris Torrence wrote:
> Hi Paul,
>
> This sounds like a bug to me. I'll try to reproduce it, but if you
> could come up with a simple reproduce case, that would be very
> helpful.
>
> By the way, you said that the "ticklen/gridstyle is hopelessly busted
> in v8.2". What's wrong with it? If there is a bug, I'd like to fix
> that before IDL 8.2.2.
>
> Thanks!
>
> -Chris ExelisVIS
|
|
|
|
| Re: Attempt to call undefined method: 'IDLITSYMBOL::IS3D' [message #82487 is a reply to message #82400] |
Wed, 12 December 2012 19:14 |
chris_torrence@NOSPAM
Messages: 528
Registered: March 2007
|
Senior Member |
|
|
Hi Paul,
This sounds like a bug to me. I'll try to reproduce it, but if you could come up with a simple reproduce case, that would be very helpful.
By the way, you said that the "ticklen/gridstyle is hopelessly busted in v8.2". What's wrong with it? If there is a bug, I'd like to fix that before IDL 8.2.2.
Thanks!
-Chris
ExelisVIS
|
|
|
|
| Re: Attempt to call undefined method: 'IDLITSYMBOL::IS3D' [message #82489 is a reply to message #82400] |
Wed, 12 December 2012 14:25 |
Paul Van Delst[1]
Messages: 1157
Registered: April 2002
|
Senior Member |
|
|
Alrighty, I've figured out the cause.
After the main plots (looping over some data) I plot simple grid lines
on teh first loop iteration (because the FG ticklen/gridstyle is
hopelessly busted in v8.2):
-----%<-----
...looping over "i"...
afn_plotref[2,i] = PLOT(f_afn.x, f_afn.y, $
TITLE='Band '+STRTRIM(band,2)+' CrIS Apodisation Function @ +Xmax', $
XTITLE='Optical delay (cm)', $
XRANGE=[xmax-delta,xmax+delta], $
YRANGE=[-0.01,0.1], $
COLOR=color, $
SYMBOL=symbol, $
LAYOUT=[1,3,3], $
OVERPLOT=overplot[2], $
/CURRENT)
IF ( i EQ 0 ) THEN BEGIN
pm = PLOT([xmax,xmax],afn_plotref[2,i].Yrange, $
LINESTYLE=1, $
OVERPLOT=afn_plotref[2,i], $
/CURRENT)
pm.Order, /SEND_TO_BACK
pz = PLOT(afn_plotref[2,i].Xrange,[0,0], $
LINESTYLE=1, $
OVERPLOT=afn_plotref[2,i], $
/CURRENT)
pz.Order, /SEND_TO_BACK
ENDIF
-----%<-----
If I comment out *both* the
pm.Order, /SEND_TO_BACK
and
pz.Order, /SEND_TO_BACK
statements, everything works.
Otherwise, I get the
Attempt to call undefined method: 'IDLITSYMBOL::IS3D'
error when I invoke the SAVE method on the graphic.
This little quest (and the idiosyncracies of FG in general) reminds me
of a question I had in my final Physics exam back in high school. I
don't remember the actual question, just our post-exam reactions to it
and our subsequent morphing of the problem to ridiculous proportions. It
sorta went like:
Facing due north, you start a ball rolling down a plane
inclined at 10deg and of length 10m. A westerly wind is
blowing at 2.5ms^-1. Using g=10.0ms^-2, what is the orbital
velocity of Jupiter?
Show all workings.
How are these things at all related?
:o)
cheers,
paulv
On 12/12/12 12:06, Paul van Delst wrote:
> Hello,
>
> I'm creating a graphics window and storing the id in a hash like so:
>
> gref[tag] = WINDOW( WINDOW_TITLE=tag, $
> DIMENSIONS = [600,800] )
>
> I then do a bunch of plotting and legend-ing in that window using the
> /CURRENT keyword.
>
> The "gref" variable is passed out of my procedure so that, if I decide
> to modify the plots (e.g. zoom, add arrows, text, whatever)
> interactively, I can still save it. I've done this in the past and it
> has always worked fine.
>
> Now when I try to do it I get an error:
>
> IDL> gref['afn'].save,'afn.png',height=800
> % SAVE: Attempt to call undefined method: 'IDLITSYMBOL::IS3D'.
> % Execution halted at: $MAIN$
>
> Has anyone seen this error before? Or explain it to me? I can find
> anything in the IDL help about either the object (IDLITSYMBOL) or the
> method (IS3D)
>
> Thanks for any info.
>
> cheers,
>
> paulv
>
>
> p.s. IDL> print, !version
> { x86 linux unix linux 8.2 Apr 10 2012 32 64}
>
|
|
|
|
|
|
|
|
| Re: Attempt to call undefined method: 'IDLITSYMBOL::IS3D' [message #82495 is a reply to message #82400] |
Wed, 12 December 2012 12:59 |
Paul Van Delst[1]
Messages: 1157
Registered: April 2002
|
Senior Member |
|
|
A little bit more digging found this:
-----%<-----
IDLitVisualization::Is3D
The IDLitVisualization::Is3D function method determines
whether or not this visualization is marked as being
three-dimensional.
Syntax
Result = Obj->[IDLitVisualization::]Is3D()
Return Value
Returns 1 if this visualization is marked as being
three-dimensional, or 0 if it is not three-dimensional.
Arguments
None
Keywords
None
-----%<-----
Based on this I started digging into the various class hierarchies
trying to figure out what is inherited by whom.
Following the list of <Direct> inherited objects...
IDL> help, obj_new('idlitsymbol'), /objects
** Object class IDLITSYMBOL, 1 direct superclass, 4 known methods
Superclasses:
IDLGRMODEL <Direct>
IDLGRCONTAINER
IDL_CONTAINER
IDLGRCOMPONENT
IDLITCOMPONENT
IDL> help, obj_new('idlgrmodel'), /objects
** Object class IDLGRMODEL, 1 direct superclass, 3 known methods
Superclasses:
IDLGRCONTAINER <Direct>
IDL_CONTAINER
IDLGRCOMPONENT
IDLITCOMPONENT
IDL> help, obj_new('idlgrcontainer'), /objects
** Object class IDLGRCONTAINER, 2 direct superclasses, 2 known methods
Superclasses:
IDL_CONTAINER <Direct>
IDLGRCOMPONENT <Direct>
IDLITCOMPONENT
IDL_CONTAINER has no superclasses (I think) so:
IDL> help, obj_new('idlgrcomponent'), /objects
** Object class IDLGRCOMPONENT, 1 direct superclass, 2 known methods
Superclasses:
IDLITCOMPONENT <Direct>
And IDLITCOMPONENT similarly has no superclasses.
So how can IDLITSYMBOL ever get access to the IS3D methjod of
IDLITVISUALIZATION? Via a contained object somehow?
Hmm. Let's go the other way:
IDL> help, obj_new('idlitvisualization'), /objects
** Object class IDLITVISUALIZATION, 2 direct superclasses, 6 known methods
Superclasses:
_IDLITVISUALIZATION <Direct>
_IDLITCONTAINER
_IDLITPROPERTYAGGREGATE
IDLGRMODEL
IDLGRCONTAINER
IDL_CONTAINER
IDLGRCOMPONENT
IDLITCOMPONENT
IDLITSELECTPARENT
IDLITIMESSAGING
IDLITPARAMETER <Direct>
Oi vey!
IDL> help, obj_new('_idlitvisualization'), /objects
** Object class _IDLITVISUALIZATION, 5 direct superclasses, 7 known methods
Superclasses:
_IDLITCONTAINER <Direct>
_IDLITPROPERTYAGGREGATE <Direct>
IDLGRMODEL <Direct>
IDLGRCONTAINER
IDL_CONTAINER
IDLGRCOMPONENT
IDLITCOMPONENT
IDLITSELECTPARENT <Direct>
IDLITIMESSAGING <Direct>
Crikey.
On 12/12/12 12:06, Paul van Delst wrote:
> Hello,
>
> I'm creating a graphics window and storing the id in a hash like so:
>
> gref[tag] = WINDOW( WINDOW_TITLE=tag, $
> DIMENSIONS = [600,800] )
>
> I then do a bunch of plotting and legend-ing in that window using the
> /CURRENT keyword.
>
> The "gref" variable is passed out of my procedure so that, if I decide
> to modify the plots (e.g. zoom, add arrows, text, whatever)
> interactively, I can still save it. I've done this in the past and it
> has always worked fine.
>
> Now when I try to do it I get an error:
>
> IDL> gref['afn'].save,'afn.png',height=800
> % SAVE: Attempt to call undefined method: 'IDLITSYMBOL::IS3D'.
> % Execution halted at: $MAIN$
>
> Has anyone seen this error before? Or explain it to me? I can find
> anything in the IDL help about either the object (IDLITSYMBOL) or the
> method (IS3D)
>
> Thanks for any info.
>
> cheers,
>
> paulv
>
>
> p.s. IDL> print, !version
> { x86 linux unix linux 8.2 Apr 10 2012 32 64}
>
|
|
|
|
comp.lang.idl-pvwave archive
Members
Search
Help
Login
Home
