| the "real" screen size [message #73042] |
Sun, 24 October 2010 15:02  |
wlandsman
Messages: 743
Registered: June 2000
|
Senior Member |
|
|
I often like to view images as big as possible on my monitor. So
I get the screen size using device,get_screen_size=winsize, and open a
window of this size using WINDOW, XSIZE=, YSIZE=. I then CONGRID()
my (bigger) image down to this size and display it with TV.
After all these years, I now realize that this method results in
significant truncation of the image on my Linux (Redhat) box (and
smaller truncation on my Mac). Device,get_screen_size reports a
screen size of 1600 x 1200 on my LInux box, but when I open a 1200 x
1200 window, I am actually only viewing the first 1115 pixels of the
image in the Y direction, so I am missing more than 7% of the image,
presumably due to pixels taken up by the window and taskbar margins
There was a thread a while back ( http://tinyurl.com/2bssnfe ) on
using the exclude_Taskbar keyword in the IDLsysMonitorInfo object to
get the "free" screen size. But this method seem more relevant to
determining the available size for a widget GUI, and in any case, the
exclude_Taskbar keyword is only available for Windows OS.
I can always introduce a fudge factor (i.e. subtract 85 pixels from
the reported screen size) but does anyone know any IDL or X-window
settings that might help?
Thanks, --Wayne
|
|
|
|
| Re: the "real" screen size [message #73111 is a reply to message #73042] |
Tue, 26 October 2010 13:41  |
David Fanning
Messages: 11724
Registered: August 2001
|
Senior Member |
|
|
Mike Galloy writes:
> My screen resolution is 1680 x 1050:
>
> IDL> print, get_screen_size()
> 1680 1028
>
>> I'm I correct that:
>>
>> Device, Get_Screen_Size=theSize
>>
>> On a Mac, actually gives the screen or monitor size?
>
> This, as well as using IDLsysMonitorInfo::getRectangles(), returns the
> same thing:
>
> IDL> Device, Get_Screen_Size=theSize
> IDL> print, thesize
> 1680 1028
OK, thanks. Back to the drawing board. :-)
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.")
|
|
|
|
| Re: the "real" screen size [message #73113 is a reply to message #73042] |
Tue, 26 October 2010 13:39 |
Michael Galloy
Messages: 1114
Registered: April 2006
|
Senior Member |
|
|
On 10/26/10 2:03 PM, David Fanning wrote:
> Mike Galloy writes:
>
>> If you want to create a maximally sized graphics window so that the
>> above example code works, then on Mac I think it has to use the UNIX
>> solution.
>
> Well, this is *extremely* confusing, which is why I'm trying
> to write it down. So, what do you think the Mac is reporting,
> then, when you do Get_Screen_Size()? You think this is the
> window *without* a title bar, but not taking into account the
> task bar, is that right?
It's the window without the menubar (the top one that is always there on
Mac OS X, sort of like the Windows taskbar that is always at the
bottom), but not accounting for the menubar on individual X windows.
It's just the "usable space" for the monitor. If you could create a
graphics window without a menubar, you could use all that space.
My screen resolution is 1680 x 1050:
IDL> print, get_screen_size()
1680 1028
> I'm I correct that:
>
> Device, Get_Screen_Size=theSize
>
> On a Mac, actually gives the screen or monitor size?
This, as well as using IDLsysMonitorInfo::getRectangles(), returns the
same thing:
IDL> Device, Get_Screen_Size=theSize
IDL> print, thesize
1680 1028
>> Does this work on Windows? IDLsysMonitor::getRectangles with
>> EXCLUDE_TASKBAR excludes the windows menubar too?
>
> It would appear so, yes.
>
> Thanks for your help with this. Not having a Mac, and
> not knowing all these machine-specific nuances, is what
> makes writing machine independent code so very, very
> difficult!
Mike
--
www.michaelgalloy.com
Research Mathematician
Tech-X Corporation
|
|
|
|
| Re: the "real" screen size [message #73116 is a reply to message #73042] |
Tue, 26 October 2010 13:03 |
David Fanning
Messages: 11724
Registered: August 2001
|
Senior Member |
|
|
Mike Galloy writes:
> If you want to create a maximally sized graphics window so that the
> above example code works, then on Mac I think it has to use the UNIX
> solution.
Well, this is *extremely* confusing, which is why I'm trying
to write it down. So, what do you think the Mac is reporting,
then, when you do Get_Screen_Size()? You think this is the
window *without* a title bar, but not taking into account the
task bar, is that right?
I'm I correct that:
Device, Get_Screen_Size=theSize
On a Mac, actually gives the screen or monitor size?
> Does this work on Windows? IDLsysMonitor::getRectangles with
> EXCLUDE_TASKBAR excludes the windows menubar too?
It would appear so, yes.
Thanks for your help with this. Not having a Mac, and
not knowing all these machine-specific nuances, is what
makes writing machine independent code so very, very
difficult!
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.")
|
|
|
|
| Re: the "real" screen size [message #73117 is a reply to message #73042] |
Tue, 26 October 2010 12:41 |
Michael Galloy
Messages: 1114
Registered: April 2006
|
Senior Member |
|
|
On 10/25/10 12:09 PM, wlandsman wrote:
> On Oct 25, 1:32 pm, mgalloy<mgal...@gmail.com> wrote:
>>
>>
>> The EXCLUDE_TASKBAR is ignored on Macs, BUT it always returns the size
>> *without* the menu bar (basically its the equivalent of
>> EXCLUDE_TASKBAR=1 always):
>
> I find this also -- so it looks like for Macs we can avoid the
> flashing to get the maximum useable image area, and that a program
> GetRealScreenSize() should have separate branches for Mac, Windows,
> and Linux.
>
> A couple of other notes:
>
> 1. The value of !D.Y_SIZE on my Linux box is intelligent, in that if
> I hide the toolbar, I will get a larger value of !D.Y_SIZE, i.e. I
> will be able to write a larger IDL image.
> 2. An alternative to device,get_screen_size = win is to use the
> GET_SCREEN_SIZE() function in the ITTVIS library. Although the
> change is not documented, this function was completely rewritten in
> IDL 8.0 to use the IDLsysMonitorInfo object. However, it does not
> accept the EXCLUDE_TASKBAR keyword. --Wayne
>
> IDL> print,get_screen_size()
> 1600 1200
>
I think we are getting slightly different things with the Windows/Mac
vs. UNIX solutions (at least as of David's current MaxWindowSize
routine): Mac is returning the size of the available space to put a
window (the full size of a maximally sized window including the menubar)
and UNIX is returning the size of the graphics part of a maximally sized
window.
So the following isn't true for Macs (I don't know about Windows):
; To create a window of maximum size::
; maxsize = MaxWindowSize()
; Window, XSize=maxsize[0], YSize=maxsize[1], /Free
This creates a window that you can't see the bottom of because the
menubar of the graphics window is "pushing" the window down a bit extra
(remember there are two menubars when dealing with an X window on Mac OS
X: the normal Mac menubar that is always there *and* a Windows-like
menubar on each window).
If you want to create a maximally sized graphics window so that the
above example code works, then on Mac I think it has to use the UNIX
solution. Does this work on Windows? IDLsysMonitor::getRectangles with
EXCLUDE_TASKBAR excludes the windows menubar too?
Mike
--
www.michaelgalloy.com
Research Mathematician
Tech-X Corporation
|
|
|
|
| Re: the "real" screen size [message #73151 is a reply to message #73042] |
Mon, 25 October 2010 11:09 |
wlandsman
Messages: 743
Registered: June 2000
|
Senior Member |
|
|
On Oct 25, 1:32 pm, mgalloy <mgal...@gmail.com> wrote:
>
>
> The EXCLUDE_TASKBAR is ignored on Macs, BUT it always returns the size
> *without* the menu bar (basically its the equivalent of
> EXCLUDE_TASKBAR=1 always):
I find this also -- so it looks like for Macs we can avoid the
flashing to get the maximum useable image area, and that a program
GetRealScreenSize() should have separate branches for Mac, Windows,
and Linux.
A couple of other notes:
1. The value of !D.Y_SIZE on my Linux box is intelligent, in that if
I hide the toolbar, I will get a larger value of !D.Y_SIZE, i.e. I
will be able to write a larger IDL image.
2. An alternative to device,get_screen_size = win is to use the
GET_SCREEN_SIZE() function in the ITTVIS library. Although the
change is not documented, this function was completely rewritten in
IDL 8.0 to use the IDLsysMonitorInfo object. However, it does not
accept the EXCLUDE_TASKBAR keyword. --Wayne
IDL> print,get_screen_size()
1600 1200
|
|
|
|
| Re: the "real" screen size [message #73153 is a reply to message #73042] |
Mon, 25 October 2010 10:32 |
Michael Galloy
Messages: 1114
Registered: April 2006
|
Senior Member |
|
|
On 10/25/10 10:15 AM, David Fanning wrote:
> David Fanning writes:
>
>> Why can't you use a pixmap?
>
> Oh, never mind. I see. Darn, I hate flashing
> windows about as much as I hate anything. :-(
>
> Nothing speaks "non-commercial" more than these
> kinds of solutions.
>
> Are the LINUX and Mac numbers all over the board,
> or could you do a reasonable "guess" and just
> incorporate those into our ever-present fudge
> factors?
The EXCLUDE_TASKBAR is ignored on Macs, BUT it always returns the size
*without* the menu bar (basically its the equivalent of
EXCLUDE_TASKBAR=1 always):
For my work monitor (Mac OS X, 1680 x 1050), I get:
IDL> print, getPrimaryScreenSize()
1680 1028
IDL> print, getPrimaryScreenSize(/exclude)
1680 1028
This seems consistent from my home laptop: the menu bar is 22 pixels tall.
Mike
--
www.michaelgalloy.com
Research Mathematician
Tech-X Corporation
|
|
|
|
|
|
|
|
| Re: the "real" screen size [message #73177 is a reply to message #73042] |
Thu, 28 October 2010 07:54 |
Kenneth P. Bowman
Messages: 585
Registered: May 2000
|
Senior Member |
|
|
In article <MPG.2731e0a4a3ee52b998980f@news.giganews.com>,
David Fanning <news@dfanning.com> wrote:
> Mike Galloy writes:
>
>>>> If you want to create a maximally sized graphics window so that the
>>>> above example code works, then on Mac I think it has to use the UNIX
>>>> solution.
>
> After further investigation overnight and this morning, I've
> finally come to the conclusion that everyone using IDL on
> a Mac should be shot. You're screwing up our lives! :-(
I think the solution is to run X-Windows on the Mac in full screen mode.
And, no, I am not serious, although there are times when I have resorted
to doing that.
Ken
|
|
|
|
| Re: the "real" screen size [message #73185 is a reply to message #73042] |
Wed, 27 October 2010 15:11 |
David Fanning
Messages: 11724
Registered: August 2001
|
Senior Member |
|
|
ronn kling writes:
> I may be missing something, but this slight modification to Dick's program
> works on my Mac....
>
> -Ronn
>
> FUNCTION GetPrimaryScreenSize, Exclude_Taskbar=exclude_Taskbar
>
> oMonInfo = Obj_New('IDLsysMonitorInfo')
> rects = oMonInfo -> GetRectangles(Exclude_Taskbar=exclude_Taskbar)
> pmi = oMonInfo -> GetPrimaryMonitorIndex()
> Obj_Destroy, oMonInfo
> if !d.name eq 'WIN' then begin
> Return, rects[[2, 3], pmi] ; w & h of primary monitor avbl.
> space
> endif else begin
> return, rects[[2, 3], pmi] - [0,rects[1]]
> endelse
> END
I don't have a Macintosh, obviously, or I would have
a better idea of what I was talking about, probably.
I think this might indeed work, but for the wrong reasons. :-)
That is to say, the monitor size is wrong, by about 22 pixels,
and then when you subtract another 22 pixels from that wrong
value, you get (surprise!!) exactly the fudge factor I've
hardcoded for the "dock". So, in effect, you found the right
value by subtracting the a decent guess at the fudge factor
from the wrong monitor size value.
Humm. Yes, I guess that algorithm is as good as any! :-)
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.")
|
|
|
|
| Re: the "real" screen size [message #73186 is a reply to message #73113] |
Wed, 27 October 2010 14:38 |
ronn
Messages: 123
Registered: April 1999
|
Senior Member |
|
|
Hello All,
I may be missing something, but this slight modification to Dick's program
works on my Mac....
-Ronn
FUNCTION GetPrimaryScreenSize, Exclude_Taskbar=exclude_Taskbar
oMonInfo = Obj_New('IDLsysMonitorInfo')
rects = oMonInfo -> GetRectangles(Exclude_Taskbar=exclude_Taskbar)
pmi = oMonInfo -> GetPrimaryMonitorIndex()
Obj_Destroy, oMonInfo
if !d.name eq 'WIN' then begin
Return, rects[[2, 3], pmi] ; w & h of primary monitor avbl.
space
endif else begin
return, rects[[2, 3], pmi] - [0,rects[1]]
endelse
END
|
|
|
|
| Re: the "real" screen size [message #73197 is a reply to message #73113] |
Wed, 27 October 2010 07:10 |
David Fanning
Messages: 11724
Registered: August 2001
|
Senior Member |
|
|
Mike Galloy writes:
>>> If you want to create a maximally sized graphics window so that the
>>> above example code works, then on Mac I think it has to use the UNIX
>>> solution.
After further investigation overnight and this morning, I've
finally come to the conclusion that everyone using IDL on
a Mac should be shot. You're screwing up our lives! :-(
It turns out that our "UNIX" solution to finding the
"maximum" size window doesn't work for Macs. In fact,
NO solution works for Macs, except the tried and true
fudge factor solution.
Unless anyone can offer me hope, I'll change my article
appropriately. Maybe I'll program my MaxWindowSize program
to turn the fire alarm on if you call it from a Mac. :-)
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
