Re: undocumented IDL built-in TVRDC [message #2292 is a reply to message #2283] |
Fri, 10 June 1994 04:33   |
thompson
Messages: 584 Registered: August 1991
|
Senior Member |
|
|
gurman@uvsp.gsfc.nasa.gov (Joseph B. Gurman) writes:
> In article <1994Jun9.182922.21061@noao.edu>,
> eharold@corona.sunspot.noao.edu (Elliotte Harold) wrote:
>> I recently ran across what appears to be an undocumented built-in
>> procedure in IDL 3.5 called TVRDC. I can't find a reference to it anywhere,
>> but it seems to return the point in the IDL window which the user
>> clicked. It's syntax is
>>
>> tvrdc,x,y,/dev
>>
>>
>> After calling this x and y now contain the x and y coordinates where the
>> mouse was clicked. Here at NSO we found it useful for a program that
>> tracks Jupiter's moons between frames. In the picture of Jupiter taken
>> with the Vacuum Tower Telescope the moons are fairly small, pretty much
>> single pixel objects so to locate them we just move the mouse and click
>> on the moon. Has anyone else used this? More to the point is there some
>> reason it's undocumented? Or is it documented somewhere that I just haven't
>> been able to find? If you're wondering how I found out about it, I saw it
>> used in the TVRD userlib procedure.
> TVRDC certainly used to be documented (at least way back in the IDL < 2
> days), but it may be (rank speculation here) that not every platform that
> now supports IDL also supports cursor readback.
> It's:
> TVRDC, x, y [, wait_for_mouse_button]
> where wait_for_mouse_button non-zero (default) means: wait 'til mouse is
> clicked to read back x and y. If wait_for_mouse_button is 0, it executes
> immediately.
> Keywords can be one of (at least): /device, /normalized, and /data.
> Anybody from RSI care to indicate why TVRDC is undocumented? Or are you
> all too busy preparing for the release of IDL 3.6? Or selling ENVI
> licenses?
I'll haphazard a guess that it's because CURSOR now does all the things that
TVRDC used to do, including taking the optional WAIT parameter. CURSOR
certainly is documented. Back in pre-version 2 days, one used TVRDC for device
coordinates and CURSOR for data coordinates, but now CURSOR can work with both
data and device coordinates, as well as normalized coordinates, by using it
with the right keyword. I wouldn't be surprised if these days TVRDC is just a
synonym for CURSOR.
By the way, this is what I have for the possible values of the optional WAIT
parameter:
Value Corresponding Keyword Action
------------------------------------------------
0 NOWAIT Return immediately
1 WAIT Return if button is down
2 CHANGE Return if a button is
pressed, released, or the
pointer is moved.
3 DOWN Return on button down
4 UP Return on button up
With the default value being 1.
Bill Thompson
|
|
|