axis procedure [message #2954] |
Wed, 19 October 1994 15:40  |
tonym
Messages: 14 Registered: May 1994
|
Junior Member |
|
|
We have version version 3.6 of idl. I have trying to use
the user library bar_plot routine. There are bugs in it, but my main
finding is this: the axis procedure no longer seems to work as I expect.
When placing tickvalues on the axis, the tick coordinates are not
interpreted as data or normalized coords. It seems to work like this:
Suppose I want to place tick marks at 0.0, 0.3, and 0.5 along an
axis going from 0.0 to 0.1. I thought I could do this:
axis,xaxis=0.0,xsty=1,xticks=2,xtickname=['1','2','3'],xtick v=[0.0,0.3,0.5],$
xrange=[0.0,1.0]
BUT, this kind of thing always places the third tickmark at the very right,
the first tickmark at the very left and the second tickmark a little right
of center. In other words, some sort of scaling is going on which always
takes the range of the xtickv array and scales to fit so that the first tickmark
is always to the left and the last tickmark is to the far right.
This seems very strange! Is this the behaviour with older versions also
(I don't recall these problems in the past). In this routine now broken.
The documentation does not seem to imply the behaviour I see!
-Tony Mannucci
tonym@lurleen.jpl.nasa.gov
Tony Mannucci Jet Propulsion Laboratory
tonym@lurleen.jpl.nasa.gov 4800 Oak Grove Drive
M/S 238-600
Voice: (818)354-1699 Fax: (818)393-4965 Pasadena, CA 91109
|
|
|
Re: axis procedure [message #2962 is a reply to message #2954] |
Thu, 20 October 1994 06:42  |
thompson
Messages: 584 Registered: August 1991
|
Senior Member |
|
|
tonym@lurleen.jpl.nasa.gov (Tony Mannucci) writes:
> We have version version 3.6 of idl. I have trying to use
> the user library bar_plot routine. There are bugs in it, but my main
> finding is this: the axis procedure no longer seems to work as I expect.
> When placing tickvalues on the axis, the tick coordinates are not
> interpreted as data or normalized coords. It seems to work like this:
> Suppose I want to place tick marks at 0.0, 0.3, and 0.5 along an
> axis going from 0.0 to 0.1. I thought I could do this:
> axis,xaxis=0.0,xsty=1,xticks=2,xtickname=['1','2','3'],xtick v=[0.0,0.3,0.5],$
> xrange=[0.0,1.0]
> BUT, this kind of thing always places the third tickmark at the very right,
> the first tickmark at the very left and the second tickmark a little right
> of center. In other words, some sort of scaling is going on which always
> takes the range of the xtickv array and scales to fit so that the first tickmark
> is always to the left and the last tickmark is to the far right.
> This seems very strange! Is this the behaviour with older versions also
> (I don't recall these problems in the past). In this routine now broken.
> The documentation does not seem to imply the behaviour I see!
There's a bug in IDL 3.6 that causes this behavior. Contact RSI for the bug
fix.
Bill Thompson
|
|
|