| A possible bug in IDL 8.2.3 [message #84768] |
Fri, 07 June 2013 19:39  |
Xin Tao
Messages: 40
Registered: April 2011
|
Member |
|
|
I don't know whether this is a bug or not, but the behaviour of the following two lines is not what I expected
IDL> a=plot([0,1,3],yrange=[0,3])
IDL> b = plot([-1,1,2],/overplot, 'g')
The line of b will go outside the plotting box, because it tried to get to -1, but the yrange is limited to be [0,3] by a.
Here is my IDL version:
{ x86_64 darwin unix Mac OS X 8.2.3 May 2 2013 64 64}
Please let me know whether there is anything I can do about it.
Thanks,
Xin
|
|
|
|
| Re: A possible bug in IDL 8.2.3 [message #84769 is a reply to message #84768] |
Fri, 07 June 2013 20:32  |
David Fanning
Messages: 11724
Registered: August 2001
|
Senior Member |
|
|
Xin Tao writes:
> I don't know whether this is a bug or not, but the behaviour of the following two lines is not what I expected
>
> IDL> a=plot([0,1,3],yrange=[0,3])
> IDL> b = plot([-1,1,2],/overplot, 'g')
What did you expect?
Cheers,
David
--
David Fanning, Ph.D.
Fanning Software Consulting, Inc.
Coyote's Guide to IDL Programming: http://www.idlcoyote.com/
Sepore ma de ni thue. ("Perhaps thou speakest truth.")
|
|
|
|
| Re: A possible bug in IDL 8.2.3 [message #84770 is a reply to message #84768] |
Fri, 07 June 2013 21:03 |
chris_torrence@NOSPAM
Messages: 528
Registered: March 2007
|
Senior Member |
|
|
On Friday, June 7, 2013 8:39:57 PM UTC-6, Xin Tao wrote:
> I don't know whether this is a bug or not, but the behaviour of the following two lines is not what I expected
>
>
>
> IDL> a=plot([0,1,3],yrange=[0,3])
>
> IDL> b = plot([-1,1,2],/overplot, 'g')
>
>
>
> The line of b will go outside the plotting box, because it tried to get to -1, but the yrange is limited to be [0,3] by a.
>
>
>
> Here is my IDL version:
>
>
>
> { x86_64 darwin unix Mac OS X 8.2.3 May 2 2013 64 64}
>
>
>
> Please let me know whether there is anything I can do about it.
>
>
>
> Thanks,
>
>
>
> Xin
Hi Xin,
I would recommend either not setting the YRANGE, or, just setting the Yrange after you add the second plot. Either way should get you what you want.
Cheers,
Chris
ExelisVIS
|
|
|
|
| Re: A possible bug in IDL 8.2.3 [message #84771 is a reply to message #84769] |
Sat, 08 June 2013 00:07 |
Xin Tao
Messages: 40
Registered: April 2011
|
Member |
|
|
I expected it to behave in a normal way; i.e., the line of b should not go outside the box, since I have already set the yrange of the plot.
On Saturday, June 8, 2013 11:32:51 AM UTC+8, David Fanning wrote:
> Xin Tao writes:
>
>
>
>> I don't know whether this is a bug or not, but the behaviour of the following two lines is not what I expected
>
>>
>
>> IDL> a=plot([0,1,3],yrange=[0,3])
>
>> IDL> b = plot([-1,1,2],/overplot, 'g')
>
>
>
> What did you expect?
>
>
>
> Cheers,
>
>
>
> David
>
>
>
>
>
>
>
> --
>
> David Fanning, Ph.D.
>
> Fanning Software Consulting, Inc.
>
> Coyote's Guide to IDL Programming: http://www.idlcoyote.com/
>
> Sepore ma de ni thue. ("Perhaps thou speakest truth.")
|
|
|
|
| Re: A possible bug in IDL 8.2.3 [message #84772 is a reply to message #84770] |
Sat, 08 June 2013 00:15 |
Xin Tao
Messages: 40
Registered: April 2011
|
Member |
|
|
Hi Chris,
Thanks for your reply. The main purpose here is to set the YRANGE and make all plots obey the YRANGe set by a. I tried your second method, but I need to use the following three lines to get a "normal" plot
IDL> a=plot([0,1,3])
IDL> b = plot([-1,1,2],/overplot)
IDL> c=plot([-1,1,3],/overplot, yrange=[0,3])
If I put yrange keyword in b, it doesn't help at all. Part of the line of b is still outside the box. I have to use the third plot c to set the yrange to a value I want. This of course solved my issue, but it's very strange that I have to use three lines to achieve this.
Xin
On Saturday, June 8, 2013 12:03:26 PM UTC+8, Chris Torrence wrote:
> On Friday, June 7, 2013 8:39:57 PM UTC-6, Xin Tao wrote:
>
>> I don't know whether this is a bug or not, but the behaviour of the following two lines is not what I expected
>
>>
>
>>
>
>>
>
>> IDL> a=plot([0,1,3],yrange=[0,3])
>
>>
>
>> IDL> b = plot([-1,1,2],/overplot, 'g')
>
>>
>
>>
>
>>
>
>> The line of b will go outside the plotting box, because it tried to get to -1, but the yrange is limited to be [0,3] by a.
>
>>
>
>>
>
>>
>
>> Here is my IDL version:
>
>>
>
>>
>
>>
>
>> { x86_64 darwin unix Mac OS X 8.2.3 May 2 2013 64 64}
>
>>
>
>>
>
>>
>
>> Please let me know whether there is anything I can do about it.
>
>>
>
>>
>
>>
>
>> Thanks,
>
>>
>
>>
>
>>
>
>> Xin
>
>
>
> Hi Xin,
>
> I would recommend either not setting the YRANGE, or, just setting the Yrange after you add the second plot. Either way should get you what you want.
>
> Cheers,
>
> Chris
>
> ExelisVIS
|
|
|
|
| Re: A possible bug in IDL 8.2.3 [message #84773 is a reply to message #84772] |
Sat, 08 June 2013 00:51 |
lecacheux.alain
Messages: 325
Registered: January 2008
|
Senior Member |
|
|
Le samedi 8 juin 2013 09:15:24 UTC+2, Xin Tao a écrit :
> Hi Chris,
>
>
>
> Thanks for your reply. The main purpose here is to set the YRANGE and make all plots obey the YRANGe set by a. I tried your second method, but I need to use the following three lines to get a "normal" plot
>
>
>
> IDL> a=plot([0,1,3])
>
> IDL> b = plot([-1,1,2],/overplot)
>
> IDL> c=plot([-1,1,3],/overplot, yrange=[0,3])
>
>
>
> If I put yrange keyword in b, it doesn't help at all. Part of the line of b is still outside the box. I have to use the third plot c to set the yrange to a value I want. This of course solved my issue, but it's very strange that I have to use three lines to achieve this.
>
>
>
> Xin
>
>
>
> On Saturday, June 8, 2013 12:03:26 PM UTC+8, Chris Torrence wrote:
>
>> On Friday, June 7, 2013 8:39:57 PM UTC-6, Xin Tao wrote:
>
>>
>
>>> I don't know whether this is a bug or not, but the behaviour of the following two lines is not what I expected
>
>>
>
>>>
>
>>
>
>>>
>
>>
>
>>>
>
>>
>
>>> IDL> a=plot([0,1,3],yrange=[0,3])
>
>>
>
>>>
>
>>
>
>>> IDL> b = plot([-1,1,2],/overplot, 'g')
>
>>
>
>>>
>
>>
>
>>>
>
>>
>
>>>
>
>>
>
>>> The line of b will go outside the plotting box, because it tried to get to -1, but the yrange is limited to be [0,3] by a.
>
>>
>
>>>
>
>>
>
>>>
>
>>
>
>>>
>
>>
>
>>> Here is my IDL version:
>
>>
>
>>>
>
>>
>
>>>
>
>>
>
>>>
>
>>
>
>>> { x86_64 darwin unix Mac OS X 8.2.3 May 2 2013 64 64}
>
>>
>
>>>
>
>>
>
>>>
>
>>
>
>>>
>
>>
>
>>> Please let me know whether there is anything I can do about it.
>
>>
>
>>>
>
>>
>
>>>
>
>>
>
>>>
>
>>
>
>>> Thanks,
>
>>
>
>>>
>
>>
>
>>>
>
>>
>
>>>
>
>>
>
>>> Xin
>
>>
>
>>
>
>>
>
>> Hi Xin,
>
>>
>
>> I would recommend either not setting the YRANGE, or, just setting the Yrange after you add the second plot. Either way should get you what you want.
>
>>
>
>> Cheers,
>
>>
>
>> Chris
>
>>
>
>> ExelisVIS
My understanding is that OVERPLOT keyword in NG *does not* work like the OPLOT of DG. Indeed, by using OVERPLOT=gr (or 1), you will mix the added plot with the 'gr' (or current) one, but without necessarily retaining the initial plot axes.
To get what you want, you must do a true overlay as follows:
IDL> a = plot([0,1,3],yrange=[0,3])
IDL> b = plot([-1,1,2], POSITION=a.POSITION, YRANGE=a.YRANGE, /CURRENT, 'g')
alx.
|
|
|
|
| Re: A possible bug in IDL 8.2.3 [message #84775 is a reply to message #84773] |
Sat, 08 June 2013 05:21 |
Xin Tao
Messages: 40
Registered: April 2011
|
Member |
|
|
Maybe you're right. However in IDL 8.2.2, the two lines in my first post will just give I what I wanted. So the behaviour of overplot actually changed with the release of 8.2.3.
Xin
On Saturday, June 8, 2013 3:51:20 PM UTC+8, alx wrote:
> Le samedi 8 juin 2013 09:15:24 UTC+2, Xin Tao a écrit :
>
>> Hi Chris,
>
>>
>
>>
>
>>
>
>> Thanks for your reply. The main purpose here is to set the YRANGE and make all plots obey the YRANGe set by a. I tried your second method, but I need to use the following three lines to get a "normal" plot
>
>>
>
>>
>
>>
>
>> IDL> a=plot([0,1,3])
>
>>
>
>> IDL> b = plot([-1,1,2],/overplot)
>
>>
>
>> IDL> c=plot([-1,1,3],/overplot, yrange=[0,3])
>
>>
>
>>
>
>>
>
>> If I put yrange keyword in b, it doesn't help at all. Part of the line of b is still outside the box. I have to use the third plot c to set the yrange to a value I want. This of course solved my issue, but it's very strange that I have to use three lines to achieve this.
>
>>
>
>>
>
>>
>
>> Xin
>
>>
>
>>
>
>>
>
>> On Saturday, June 8, 2013 12:03:26 PM UTC+8, Chris Torrence wrote:
>
>>
>
>>> On Friday, June 7, 2013 8:39:57 PM UTC-6, Xin Tao wrote:
>
>>
>
>>>
>
>>
>
>>>> I don't know whether this is a bug or not, but the behaviour of the following two lines is not what I expected
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>> IDL> a=plot([0,1,3],yrange=[0,3])
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>> IDL> b = plot([-1,1,2],/overplot, 'g')
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>> The line of b will go outside the plotting box, because it tried to get to -1, but the yrange is limited to be [0,3] by a.
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>> Here is my IDL version:
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>> { x86_64 darwin unix Mac OS X 8.2.3 May 2 2013 64 64}
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>> Please let me know whether there is anything I can do about it.
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>> Thanks,
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>> Xin
>
>>
>
>>>
>
>>
>
>>>
>
>>
>
>>>
>
>>
>
>>> Hi Xin,
>
>>
>
>>>
>
>>
>
>>> I would recommend either not setting the YRANGE, or, just setting the Yrange after you add the second plot. Either way should get you what you want.
>
>>
>
>>>
>
>>
>
>>> Cheers,
>
>>
>
>>>
>
>>
>
>>> Chris
>
>>
>
>>>
>
>>
>
>>> ExelisVIS
>
>
>
>
>
> My understanding is that OVERPLOT keyword in NG *does not* work like the OPLOT of DG. Indeed, by using OVERPLOT=gr (or 1), you will mix the added plot with the 'gr' (or current) one, but without necessarily retaining the initial plot axes.
>
> To get what you want, you must do a true overlay as follows:
>
>
>
> IDL> a = plot([0,1,3],yrange=[0,3])
>
> IDL> b = plot([-1,1,2], POSITION=a.POSITION, YRANGE=a.YRANGE, /CURRENT, 'g')
>
>
>
> alx.
|
|
|
|
| Re: A possible bug in IDL 8.2.3 [message #84776 is a reply to message #84775] |
Sat, 08 June 2013 05:37 |
dg86
Messages: 118
Registered: September 2012
|
Senior Member |
|
|
On Saturday, June 8, 2013 8:21:51 AM UTC-4, Xin Tao wrote:
> Maybe you're right. However in IDL 8.2.2, the two lines in my first post will just give I what I wanted. So the behaviour of overplot actually changed with the release of 8.2.3.
>
>
>
> Xin
>
>
>
> On Saturday, June 8, 2013 3:51:20 PM UTC+8, alx wrote:
>
>> Le samedi 8 juin 2013 09:15:24 UTC+2, Xin Tao a écrit :
>
>>
>
>>> Hi Chris,
>
>>
>
>>>
>
>>
>
>>>
>
>>
>
>>>
>
>>
>
>>> Thanks for your reply. The main purpose here is to set the YRANGE and make all plots obey the YRANGe set by a. I tried your second method, but I need to use the following three lines to get a "normal" plot
>
>>
>
>>>
>
>>
>
>>>
>
>>
>
>>>
>
>>
>
>>> IDL> a=plot([0,1,3])
>
>>
>
>>>
>
>>
>
>>> IDL> b = plot([-1,1,2],/overplot)
>
>>
>
>>>
>
>>
>
>>> IDL> c=plot([-1,1,3],/overplot, yrange=[0,3])
>
>>
>
>>>
>
>>
>
>>>
>
>>
>
>>>
>
>>
>
>>> If I put yrange keyword in b, it doesn't help at all. Part of the line of b is still outside the box. I have to use the third plot c to set the yrange to a value I want. This of course solved my issue, but it's very strange that I have to use three lines to achieve this.
>
>>
>
>>>
>
>>
>
>>>
>
>>
>
>>>
>
>>
>
>>> Xin
>
>>
>
>>>
>
>>
>
>>>
>
>>
>
>>>
>
>>
>
>>> On Saturday, June 8, 2013 12:03:26 PM UTC+8, Chris Torrence wrote:
>
>>
>
>>>
>
>>
>
>>>> On Friday, June 7, 2013 8:39:57 PM UTC-6, Xin Tao wrote:
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>> > I don't know whether this is a bug or not, but the behaviour of the following two lines is not what I expected
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>> >
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>> >
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>> >
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>> > IDL> a=plot([0,1,3],yrange=[0,3])
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>> >
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>> > IDL> b = plot([-1,1,2],/overplot, 'g')
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>> >
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>> >
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>> >
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>> > The line of b will go outside the plotting box, because it tried to get to -1, but the yrange is limited to be [0,3] by a.
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>> >
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>> >
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>> >
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>> > Here is my IDL version:
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>> >
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>> >
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>> >
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>> > { x86_64 darwin unix Mac OS X 8.2.3 May 2 2013 64 64}
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>> >
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>> >
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>> >
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>> > Please let me know whether there is anything I can do about it.
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>> >
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>> >
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>> >
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>> > Thanks,
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>> >
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>> >
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>> >
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>> > Xin
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>> Hi Xin,
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>> I would recommend either not setting the YRANGE, or, just setting the Yrange after you add the second plot. Either way should get you what you want.
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>> Cheers,
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>> Chris
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>> ExelisVIS
>
>>
>
>>
>
>>
>
>>
>
>>
>
>> My understanding is that OVERPLOT keyword in NG *does not* work like the OPLOT of DG. Indeed, by using OVERPLOT=gr (or 1), you will mix the added plot with the 'gr' (or current) one, but without necessarily retaining the initial plot axes.
>
>>
>
>> To get what you want, you must do a true overlay as follows:
>
>>
>
>>
>
>>
>
>> IDL> a = plot([0,1,3],yrange=[0,3])
>
>>
>
>> IDL> b = plot([-1,1,2], POSITION=a.POSITION, YRANGE=a.YRANGE, /CURRENT, 'g')
>
>>
>
>>
>
>>
>
>> alx.
I agree with Xin that the new behavior is less desirable that the old. If I set the range in the original plot, then I probably want to keep that range as I build up the graphic. I wonder if this new behavior is related to the issue of plotting speed for complicated graphics that arose over 8.2.2.
David
|
|
|
|
| Re: A possible bug in IDL 8.2.3 [message #84778 is a reply to message #84776] |
Sat, 08 June 2013 06:41 |
lecacheux.alain
Messages: 325
Registered: January 2008
|
Senior Member |
|
|
Le samedi 8 juin 2013 14:37:28 UTC+2, David Grier a écrit :
> On Saturday, June 8, 2013 8:21:51 AM UTC-4, Xin Tao wrote:
>
>> Maybe you're right. However in IDL 8.2.2, the two lines in my first post will just give I what I wanted. So the behaviour of overplot actually changed with the release of 8.2.3.
>
>>
>
>>
>
>>
>
>> Xin
>
>>
>
>>
>
>>
>
>> On Saturday, June 8, 2013 3:51:20 PM UTC+8, alx wrote:
>
>>
>
>>> Le samedi 8 juin 2013 09:15:24 UTC+2, Xin Tao a écrit :
>
>>
>
>>>
>
>>
>
>>>> Hi Chris,
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>> Thanks for your reply. The main purpose here is to set the YRANGE and make all plots obey the YRANGe set by a. I tried your second method, but I need to use the following three lines to get a "normal" plot
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>> IDL> a=plot([0,1,3])
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>> IDL> b = plot([-1,1,2],/overplot)
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>> IDL> c=plot([-1,1,3],/overplot, yrange=[0,3])
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>> If I put yrange keyword in b, it doesn't help at all. Part of the line of b is still outside the box. I have to use the third plot c to set the yrange to a value I want. This of course solved my issue, but it's very strange that I have to use three lines to achieve this.
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>> Xin
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>> On Saturday, June 8, 2013 12:03:26 PM UTC+8, Chris Torrence wrote:
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>> > On Friday, June 7, 2013 8:39:57 PM UTC-6, Xin Tao wrote:
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>> >
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>> > > I don't know whether this is a bug or not, but the behaviour of the following two lines is not what I expected
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>> >
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>> > >
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>> >
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>> > >
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>> >
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>> > >
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>> >
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>> > > IDL> a=plot([0,1,3],yrange=[0,3])
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>> >
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>> > >
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>> >
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>> > > IDL> b = plot([-1,1,2],/overplot, 'g')
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>> >
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>> > >
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>> >
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>> > >
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>> >
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>> > >
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>> >
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>> > > The line of b will go outside the plotting box, because it tried to get to -1, but the yrange is limited to be [0,3] by a.
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>> >
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>> > >
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>> >
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>> > >
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>> >
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>> > >
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>> >
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>> > > Here is my IDL version:
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>> >
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>> > >
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>> >
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>> > >
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>> >
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>> > >
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>> >
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>> > > { x86_64 darwin unix Mac OS X 8.2.3 May 2 2013 64 64}
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>> >
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>> > >
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>> >
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>> > >
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>> >
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>> > >
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>> >
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>> > > Please let me know whether there is anything I can do about it.
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>> >
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>> > >
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>> >
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>> > >
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>> >
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>> > >
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>> >
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>> > > Thanks,
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>> >
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>> > >
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>> >
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>> > >
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>> >
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>> > >
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>> >
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>> > > Xin
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>> >
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>> >
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>> >
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>> > Hi Xin,
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>> >
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>> > I would recommend either not setting the YRANGE, or, just setting the Yrange after you add the second plot. Either way should get you what you want.
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>> >
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>> > Cheers,
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>> >
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>> > Chris
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>> >
>
>>
>
>>>
>
>>
>
>>>>
>
>>
>
>>>
>
>>
>
>>>> > ExelisVIS
>
>>
>
>>>
>
>>
>
>>>
>
>>
>
>>>
>
>>
>
>>>
>
>>
>
>>>
>
>>
>
>>> My understanding is that OVERPLOT keyword in NG *does not* work like the OPLOT of DG. Indeed, by using OVERPLOT=gr (or 1), you will mix the added plot with the 'gr' (or current) one, but without necessarily retaining the initial plot axes.
>
>>
>
>>>
>
>>
>
>>> To get what you want, you must do a true overlay as follows:
>
>>
>
>>>
>
>>
>
>>>
>
>>
>
>>>
>
>>
>
>>> IDL> a = plot([0,1,3],yrange=[0,3])
>
>>
>
>>>
>
>>
>
>>> IDL> b = plot([-1,1,2], POSITION=a.POSITION, YRANGE=a.YRANGE, /CURRENT, 'g')
>
>>
>
>>>
>
>>
>
>>>
>
>>
>
>>>
>
>>
>
>>> alx.
>
>
>
> I agree with Xin that the new behavior is less desirable that the old. If I set the range in the original plot, then I probably want to keep that range as I build up the graphic. I wonder if this new behavior is related to the issue of plotting speed for complicated graphics that arose over 8.2.2.
>
>
>
> David
Indeed the behaviour of keyword OVERPLOT looks like to have changed in 8.2.3.
In 8.2.2 and before, when using /OVERPLOT, the axis ranges were modified as needed in order to show all of the two curves in a single window. In 8.2.2, the second plot seems to be just shown unclipped within the initial window. The new behaviour appears to me a bit more consistent, but still not ideal.
I shall continue to prefer the combination of POSITION and RANGE keywords in order to safely overplot several curves.
alx.
|
|
|
|
| Re: A possible bug in IDL 8.2.3 [message #84849 is a reply to message #84768] |
Thu, 13 June 2013 13:24 |
Haje Korth
Messages: 651
Registered: May 1997
|
Senior Member |
|
|
This bugs is so bad, it virtually breaks all my codes. So I am doing what I have never done in almost 30 years of using IDL, which is to roll back to an older version. So v8.2.2 it is until the next release.
On Friday, June 7, 2013 10:39:57 PM UTC-4, Xin Tao wrote:
> I don't know whether this is a bug or not, but the behaviour of the following two lines is not what I expected
>
>
>
> IDL> a=plot([0,1,3],yrange=[0,3])
>
> IDL> b = plot([-1,1,2],/overplot, 'g')
>
>
>
> The line of b will go outside the plotting box, because it tried to get to -1, but the yrange is limited to be [0,3] by a.
>
>
>
> Here is my IDL version:
>
>
>
> { x86_64 darwin unix Mac OS X 8.2.3 May 2 2013 64 64}
>
>
>
> Please let me know whether there is anything I can do about it.
>
>
>
> Thanks,
>
>
>
> Xin
|
|
|
|
|
|
| Re: A possible bug in IDL 8.2.3 [message #84851 is a reply to message #84850] |
Thu, 13 June 2013 13:53 |
Haje Korth
Messages: 651
Registered: May 1997
|
Senior Member |
|
|
Congratulations on your prediction, wish you were wrong!
On Thursday, June 13, 2013 4:31:39 PM UTC-4, David Fanning wrote:
> Haje Korth writes:
>
>
>
>> This bugs is so bad, it virtually breaks all my codes. So I am doing what I have never done in almost 30 years of using IDL, which is to roll back to an older version. So v8.2.2 it is until the next release.
>
>
>
> Let's see, I think I predicted that the function graphics system is so
>
> complicated that fixing one thing would inevitably have to break
>
> something else. Let's hope whatever it is in next release doesn't affect
>
> your work. ;-)
>
>
>
> Cheers,
>
>
>
> David
>
>
>
> --
>
> David Fanning, Ph.D.
>
> Fanning Software Consulting, Inc.
>
> Coyote's Guide to IDL Programming: http://www.idlcoyote.com/
>
> Sepore ma de ni thue. ("Perhaps thou speakest truth.")
|
|
|
|
|
|
| Re: A possible bug in IDL 8.2.3 [message #84859 is a reply to message #84852] |
Fri, 14 June 2013 10:43 |
chris_torrence@NOSPAM
Messages: 528
Registered: March 2007
|
Senior Member |
|
|
Hi all,
This is indeed a bug in IDL 8.2.3. When I tried out Xin's initial code, I didn't see the problem because it had already been fixed, so I assumed it was just a misunderstanding of how overplot worked. My bad!
Anyway, it looks like it is just a couple of files that need to get changed. Mark Piper is going to post them to our website for download.
In the meantime, if you need a quick fix for the problem, one workaround is to disable and then re-enable clipping for the plot. Like this:
a=plot([0,1,3],yrange=[0,3])
b = plot([-1,1,2],/overplot, 'g',clip=0)
b.clip = 1
Sorry for the bug, and thanks for reporting it!
Cheers,
Chris
ExelisVIS
|
|
|
|
|
|
|
|
|
|
| Re: A possible bug in IDL 8.2.3 [message #84879 is a reply to message #84878] |
Sun, 16 June 2013 07:42 |
David Fanning
Messages: 11724
Registered: August 2001
|
Senior Member |
|
|
David Fanning writes:
>
> Mark Piper writes:
>
>> Here's Chris' fix. It requires two PRO files. Drop this file:
>>
>> ftp://download.exelisvis.com/groups/esg/dropoff/idlitvisdata space__define.pro
>>
>> into the lib/itools/framework directory, and this file:
>>
>> ftp://download.exelisvis.com/groups/esg/dropoff/idlitvisplot __define.pro
>>
>> into lib/itools/component. Reset your IDL session to ensure these autocompile when called.
>>
>> I've checked this with Xin's code on Linux and Windows.
>
> I just tried this on my Windows machine and it doesn't work. The reason,
> I *think* it doesn't work is that I don't have permission to write into
> those directories on my machine. (Although I don't get any errors when I
> try to save the files to those directories.) I have the same problem
> when I make any change in a lib file. I have to move the file to a
> directory I own to make the change.
>
> I presume this has to do with permissions on these directories, and I
> presume I have permission to change those permissions. But, I have no
> idea how to go about this. Any ideas?
>
> Cheers,
>
> David
>
> P.S. I just confirmed that even though it appeared I was able to write
> into those two directories, nothing in the two directories (as
> determined by the data modified property) was changed.
To make this work on my Windows computer, I had to right-click on the
/lib/itools folder and choose the Properties selection from the menu.
Then, I had to chose the Security tab on the Properties dialog. I
selected the Group CREATOR OWNER and hit the Edit tab below the Group
selector and click the Allow box on the Full Control selection pop-up.
Then, I dragged the files from another location where I had saved them
into the appropriate directories. I got a dialog indicating I needed
Administrator privilege to move a file into the directory. I clicked OK,
and they were moved into the directories successfully.
I don't know if any or all of this is necessary. I just know I can't
save them into IDL-owned directories directly.
Cheers,
David
--
David Fanning, Ph.D.
Fanning Software Consulting, Inc.
Coyote's Guide to IDL Programming: http://www.idlcoyote.com/
Sepore ma de ni thue. ("Perhaps thou speakest truth.")
|
|
|
|
|
|
|
|
comp.lang.idl-pvwave archive
Members
Search
Help
Login
Home
