| Re: Preserving coordinate transformation [message #50812] |
Tue, 17 October 2006 07:14  |
David Fanning
Messages: 11724
Registered: August 2001
|
Senior Member |
|
|
Sverre Solberg writes:
> And thanks for the help, I wasn't aware that wset reestablished !D.
I have the advantage of having recently spent several
glorious days in the thrall of coordinate conversion
mechanisms. :-)
I did learn something else of interest in the three
excruciating days I spent considering all of this.
I was working with a window that had several images
in it. Each image has associated with it its own
data coordinate system. My draw widget is able
to "select" which image I clicked in, and tells
that image to establish its data coordinate system.
Now, I usually need to convert the window location
(event.x and event.y) into the data coordinate system
of the image. No problem with CONVERT_COORD, in the
manor I outlined previously:
d = Convert_Coord(event.x, event.y, /Device, /To_Data)
BUT, on occasion I need to convert a *distance* in device
coordinates into a distance in data coordinates. For example,
if I want to draw a station plot at the location where I
clicked, but I want to add a station name just *above* the
location, I may want to move the text 10 pixels toward
the top of the image.
Naively, I was doing the conversion like this:
d = Convert_Coord(10, 10, /Device, /To_Data)
Then adding the "length" to the y data coordinate:
y_data = y_data + d[1]
This worked great with some images, and not at all with
others! It turns out it worked great when I had one
image in the window and the image coordinate system
overlapped more or less with the device coordinate
system. (Or, when the image I selected was near the
origin of the device coordinate system.) But if an
image was far away from the origin of the device
coordinate system my text could be WAY off the mark.
I realized, after much futzing around and thinking
through an unhappy dinner with my wife ("Are you
paying attention to what I'm talking about!?") that
this was the wrong way to calculate a distance or
length in my data coordinate system.
What I needed to know, was how many data coordinate
"units" were in 10 "units" of device coordinates. This
depended on my data coordinate range and is something
CONVERT_COORD cannot compute for me. In fact, I realized
I had to do the conversion myself, like this:
d = Abs((yrange[1] - yrange[0])) / !D.Y_Size * 10
y_data = y_data + d
Clearing up this little misunderstanding solved a LOT
of problems for me. :-)
Cheers,
David
--
David Fanning, Ph.D.
Fanning Software Consulting, Inc.
Coyote's Guide to IDL Programming: http://www.dfanning.com/
Sepore ma de ni thui. ("Perhaps thou speakest truth.")
|
|
|
|
comp.lang.idl-pvwave archive
Members
Search
Help
Login
Home
