| Object Graphics when using 'stop' [message #30555] |
Thu, 02 May 2002 08:12  |
Randall Skelton
Messages: 169
Registered: October 2000
|
Senior Member |
|
|
Hi all,
I'm using IDL 5.3 under linux and I have a curious question. For
perspective, I tend to do most of my coding from the command line and not
the gui (yes, I do use the 'stop' command a lot). I routinely check the
contents of variables with 'print' and 'plot' commands to determine if my
calculations look reasonable. My problem is, once I have issued a 'stop'
command I cannot use object graphics to plot anything. All I get is a
black background window that is usually dissociated from xmanager. I'm
sure there is a good reason for this but I'm stymied. Any thoughts?
Cheers,
Randall
|
|
|
|
|
|
| Re: Object Graphics when using 'stop' [message #30634 is a reply to message #30555] |
Mon, 13 May 2002 14:36  |
Paul Sorenson
Messages: 48
Registered: May 2002
|
Member |
|
|
David Fanning's post advising /BLOCK is brilliant! There are two bodies of
code
that implement the main event loop in IDL. The one that runs when there is
a blocking
widget is written in .pro (xmanger.pro). We know that .pro code can be run
when IDL
is stopped. That is why the dist rountine, for example, can be invoked.
When we ask
for a blocking widget, we cause IDL to use the .pro version of its main
event loop.
That code, just like any other .pro code, runs, and we can interact with
widgets, rotate,
etc. as usual.
Ignore my post about "dbov", which was a wrapper to xObjView. Just invoke
xObjView
with /BLOCK instead. Much better than a cartoon flip-book!
-Paul
"Paul Sorenson" <aardvark62@msn.com> wrote in message
news:3cdc2db1$1_3@corp-goliath.newsgroups.com...
> The offending widget programs are written in such a way that they never
draw
> their graphics until they receive an expose event. xObjView currently has
> the same weakness. Perhaps a clever fix could be put into these programs.
> Until then here is a lame hack I use. Its a wrapper to xObjView.
>
> pro dbov, oObj, _extra=e
> ;
> ;Procedure dbov: "debug objview". Provide an invocation of xobjview
> ;that is useful when IDL is stopped (e.g. at a breakpoint). Widget
> ;programs don't run (i.e. respond to events) when IDL is stopped.
> ;xObjView is a widget program. If it is invoked when IDL is stopped,
> ;its graphic window remains blank because it is waiting for an expose
> ;event which never comes. Procedure dbov forces an extra draw to
> ;guarantee that the initial view of oObj is drawn.
> ;
> xobjview, oObj, tlb=tlb, _extra=e
> xobjview, refresh=tlb
> end
>
> Paul Sorenson
> Gun For Hire
> aardvark62@msn.com
>
> "Rick Towler" <rtowler@u.washington.edu> wrote in message
> news:abgtdg$1kks$1@nntp6.u.washington.edu...
>> Ahhh, widgets....
>>
>> Issuing the STOP command stops xmanager's ability to process events.
> Since
>> OG widget programs rely on events to draw the view to the window this
> never
>> happens and you get the black window of despair. Direct graphics widget
>> programs will possibly draw the initial plot but will not be able to
> process
>> events either.
>>
>> I can't explain the details but I'm sure someone else in the group can.
>>
>> A non widget program will run as expected after issuing a stop. I
created
> a
>> simple non-widget OG program that I could pass data to plot after
issuing
> a
>> stop. You could craft yourself some nice little tools to do line and
>> surface plots which automagically scale and rotate objects. You lose
the
>> ability to manipulate them and cleanup is awkward. You would probably
> have
>> to use READ to wait for a keypress before destroying the objects.
>>
>> Or maybe you just look at your data using DG. So that's why they are
>> keeping DG around....
>>
>> -Rick
>>
>>
>> "Randall Skelton" <rhskelto@atm.ox.ac.uk> wrote in message
>> news:Pine.LNX.4.33.0205101048540.16883-100000@mulligan.atm.o x.ac.uk...
>>> With regard to more context, imagine a procedure aptly named 'test':
>>>
>>> PRO TEST
>>> DATA = BESELJ(SHIFT(DIST(40),20,20)/2,0)
>>> STOP
>>> END
>>>
>>> When I run this in IDL I get:
>>>
>>> IDL> test
>>> % Compiled module: TEST.
>>> % Compiled module: DIST.
>>> % Stop encountered: TEST 5 /blah/.../blah/test.pro
>>>
>>> If I try and plot 'data' with an object graphics routine like
>>> 'fsc_surface', or any of the RSI example code 'show3_track',
>>> 'test_surface' all I get is a black window? Moreover, all of these
>>> routines have built-in data (in the event that the user doesn't pass
an
>>> array) and even this fails to display when my routine is stopped. Of
>>> course, once I continue the program and allow it to finish, everything
> is
>>> fine and the object-graphics window fills with the expected surface.
>>>
>>> Any help would be greatly appreciated!
>>>
>>> Cheers,
>>> Randall
>>>
>>> IDL 5.3 (soon to be 5.5); RH Linux 7.x; XFree 4.0.3
>>>
>>> On Fri, 3 May 2002, Rick Towler wrote:
>>>
>>>> Hi Randall,
>>>>
>>>> Can you provide some more context? I just stopped an OG program,
>>>> manipulated objects and drew the window. Is that what you are
trying
> to
>> do?
>>>>
>>>> -Rick
>>>>
>>>>
>>>> "Randall Skelton" <rhskelto@atm.ox.ac.uk> wrote in message
>>>>
news:Pine.LNX.4.33.0205021605500.27169-100000@mulligan.atm.o x.ac.uk...
>>>> > Hi all,
>>>> >
>>>> > I'm using IDL 5.3 under linux and I have a curious question. For
>>>> > perspective, I tend to do most of my coding from the command line
> and
>> not
>>>> > the gui (yes, I do use the 'stop' command a lot). I routinely
check
>> the
>>>> > contents of variables with 'print' and 'plot' commands to
determine
> if
>> my
>>>> > calculations look reasonable. My problem is, once I have issued a
>> 'stop'
>>>> > command I cannot use object graphics to plot anything. All I get
is
> a
>>>> > black background window that is usually dissociated from xmanager.
>> I'm
>>>> > sure there is a good reason for this but I'm stymied. Any
thoughts?
>>>> >
>>>> > Cheers,
>>>> > Randall
>>>> >
>>>>
>>>>
>>>>
>>>
>>
>>
>
>
>
>
> -----------== Posted via Newsgroups.Com - Uncensored Usenet News
==----------
> http://www.newsgroups.com The #1 Newsgroup Service in the World!
> -----= Over 100,000 Newsgroups - Ulimited Fast Downloads - 19 Servers
=-----
-----------== Posted via Newsgroups.Com - Uncensored Usenet News ==----------
http://www.newsgroups.com The #1 Newsgroup Service in the World!
-----= Over 100,000 Newsgroups - Ulimited Fast Downloads - 19 Servers =-----
|
|
|
|
| Re: Object Graphics when using 'stop' [message #30662 is a reply to message #30555] |
Sun, 12 May 2002 12:42 |
Rick Towler
Messages: 821
Registered: August 1998
|
Senior Member |
|
|
Heck, ignore what I am saying. Check out David's solution using his
blocking fsc_surface program. Very cool.
-R
"Rick Towler" <tsehai@attbi.com> wrote
> The suggested work arounds don't really address Randall's main need for
> manipulating his data sets. You can force an application to draw before
it
> goes into the event loop but it still won't accept input or redraw. Maybe
> an approach like this would work:
>
> PRO TEST
> DATA = BESELJ(SHIFT(DIST(40),20,20)/2,0)
>
> not_ok:
> stop
> FSC_SURFACE, data, /block, ok=ok
> if (not ok) then goto, not_ok
>
> END
>
> You could issue a stop, play with your data, then issue a .cont where you
> would jump into your OG viewer. Depending on how you choose to proceed,
you
> could stop again and play with the data or you could continue on. Some
> minor modifications to your favorite OG program will be in order. Not
> pretty but functional.
>
> -Rick
>
>
>
|
|
|
|
| Re: Object Graphics when using 'stop' [message #30664 is a reply to message #30555] |
Sat, 11 May 2002 12:43 |
David Fanning
Messages: 11724
Registered: August 2001
|
Senior Member |
|
|
David Fanning (david@dfanning.com) writes:
> In your little example test program, I can stop the program,
> then call FSC_SURFACE like this:
>
> IDL> FSC_SURFACE, data
Whoops! Of course, this should be:
IDL> FSC_SURFACE, data, /Block
Cheers,
David
--
David W. Fanning, Ph.D.
Fanning Software Consulting
Phone: 970-221-0438, E-mail: david@dfanning.com
Coyote's Guide to IDL Programming: http://www.dfanning.com/
Toll-Free IDL Book Orders: 1-888-461-0155
|
|
|
|
| Re: Object Graphics when using 'stop' [message #30665 is a reply to message #30555] |
Sat, 11 May 2002 12:39 |
David Fanning
Messages: 11724
Registered: August 2001
|
Senior Member |
|
|
Randall Skelton (rhskelto@atm.ox.ac.uk) writes:
> Well this is certainly not good news... The whole point of me using OG was
> so that I could actively pan-and-scan through a very complex data array.
> Personally, I need to be able to do this when I use STOP or set
> breakpoints.
>
> Surely all those die-hard proponents of IDL OG will be hard at work this
> weekend trying to come up with a fix for this ;)
I have long thought of myself as Cassandra, crying in the
wilderness, desperate to get across my message of the power
of prayer. But my words continue to fall on deaf ears.
(We even have a new call for a "wish list" this week.)
People! Wake up! The gods listen to our prayers and answer
them. The result is always chaos and confusion.
This whole business, I'm afraid, is because some people (I won't
name them, but you would recognize the names, I'm sure) would
not be content running widget programs as the gods intended them to
be run: one at a time. No, they wanted to run multiple widget
programs AND type at the IDL command line, all at the same time.
So, we were given this, and it has all been downhill since.
So, whatever it is that STOPS working with a STOP command,
it is the thing that makes widgets work while the interpreter
is working, too. (I intend to wave my hands here a bit, because
I don't understand the details, or even the fundamentals. I'm
just going to tell you how to work around whatever it is.)
Let's just say there are loops within loops, and leave it at
that. When you STOP whatever it is, call it the "initial interpreter",
you stop the ability to send widget events to widget programs
from that same "event loop". (This was all discussed the other
day.)
What we would like to do at the moment we are stopped, is start
up *another* event loop that could work on our widget program,
while the initial interpreter is momentarily stopped. We can do
this, if we make the new widget program a BLOCKING widget program.
Now, I don't want to go far into the details of this, but recall
that only ONE program can BLOCK. If you try to run two blocking
programs, only the first one blocks the command line.
So, if the program you are running is a non-blocking program.
(It sounds like it to me.) Then you can run an object graphics
widget program, IF you can get this widget program to run as
a blocking widget program.
I've added a BLOCK keyword to FSC_SURFACE so you can run
it as a blocking program if this keyword is set.
http://www.dfanning.com/programs/fsc_surface.pro
In your little example test program, I can stop the program,
then call FSC_SURFACE like this:
IDL> FSC_SURFACE, data
Not only does the surface show up, but I can rotate the surface,
print, make JPEG files, etc. I have to close the program before
I can type .CONTINUE in the stopped program.
> PS: David, on what page of your book is this bizarre IDL widget behavior
> described?
Oh, I'm sure it's described somewhere. I just can't find it. :-(
> Surely you have an insightful comment to add?
Of course.
The secret to a happy life is not getting what you
want. It is wanting what you get.
Cheers,
David
--
David W. Fanning, Ph.D.
Fanning Software Consulting
Phone: 970-221-0438, E-mail: david@dfanning.com
Coyote's Guide to IDL Programming: http://www.dfanning.com/
Toll-Free IDL Book Orders: 1-888-461-0155
|
|
|
|
| Re: Object Graphics when using 'stop' [message #30666 is a reply to message #30555] |
Sat, 11 May 2002 12:23 |
Rick Towler
Messages: 821
Registered: August 1998
|
Senior Member |
|
|
[more stuff deleted]
> [stuff deleted]
>
> I recall an interesting thread from last year regarding XMANAGER
> internals, and whether you even needed XMANAGER to run a widget program.
> A search on Google Groups
>
> http://groups.google.com/groups?group=comp.lang.idl-pvwave
>
> for XMANAGER yielded the thread "Re: Xmanager R US" as the first hit. In
> the thread, Pavel Romashkin and J.D. Smith make some interesting
> observations about the workings of XMANAGER which may be relevant.
>
> Cheers,
> Liam.
> Practical IDL Programming
> http://www.gumley.com/
This is worth investigating but I wonder how deep this goes. IDL still
needs to process events whether it be xmanager or whatever is *really*
managing events and I am not convinced that xmanager is the real problem.
The suggested work arounds don't really address Randall's main need for
manipulating his data sets. You can force an application to draw before it
goes into the event loop but it still won't accept input or redraw. Maybe
an approach like this would work:
PRO TEST
DATA = BESELJ(SHIFT(DIST(40),20,20)/2,0)
not_ok:
stop
FSC_SURFACE, data, /block, ok=ok
if (not ok) then goto, not_ok
END
You could issue a stop, play with your data, then issue a .cont where you
would jump into your OG viewer. Depending on how you choose to proceed, you
could stop again and play with the data or you could continue on. Some
minor modifications to your favorite OG program will be in order. Not
pretty but functional.
-Rick
|
|
|
|
| Re: Object Graphics when using 'stop' [message #30667 is a reply to message #30555] |
Sat, 11 May 2002 09:26 |
Liam E. Gumley
Messages: 378
Registered: January 2000
|
Senior Member |
|
|
Rick Towler wrote:
>
> Ahhh, widgets....
>
> Issuing the STOP command stops xmanager's ability to process events. Since
> OG widget programs rely on events to draw the view to the window this never
> happens and you get the black window of despair. Direct graphics widget
> programs will possibly draw the initial plot but will not be able to process
> events either.
>
> I can't explain the details but I'm sure someone else in the group can.
>
[stuff deleted]
I recall an interesting thread from last year regarding XMANAGER
internals, and whether you even needed XMANAGER to run a widget program.
A search on Google Groups
http://groups.google.com/groups?group=comp.lang.idl-pvwave
for XMANAGER yielded the thread "Re: Xmanager R US" as the first hit. In
the thread, Pavel Romashkin and J.D. Smith make some interesting
observations about the workings of XMANAGER which may be relevant.
Cheers,
Liam.
Practical IDL Programming
http://www.gumley.com/
|
|
|
|
| Re: Object Graphics when using 'stop' [message #30668 is a reply to message #30555] |
Sat, 11 May 2002 08:45 |
David Fanning
Messages: 11724
Registered: August 2001
|
Senior Member |
|
|
Randall Skelton (rhskelto@atm.ox.ac.uk) writes:
> Surely all those die-hard proponents of IDL OG will be hard at work this
> weekend trying to come up with a fix for this ;)
OK I have a partial fix, but if I want to save my
marriage, I am going to have to write it up later.
If this lawn doesn't get mowed NOW I am in HOT water. :-(
I'll set my alarm for 3AM and write it up then.
Cheers,
David
--
David W. Fanning, Ph.D.
Fanning Software Consulting
Phone: 970-221-0438, E-mail: david@dfanning.com
Coyote's Guide to IDL Programming: http://www.dfanning.com/
Toll-Free IDL Book Orders: 1-888-461-0155
|
|
|
|
| Re: Object Graphics when using 'stop' [message #30669 is a reply to message #30555] |
Fri, 10 May 2002 23:52 |
Randall Skelton
Messages: 169
Registered: October 2000
|
Senior Member |
|
|
Well this is certainly not good news... The whole point of me using OG was
so that I could actively pan-and-scan through a very complex data array.
Personally, I need to be able to do this when I use STOP or set
breakpoints.
Surely all those die-hard proponents of IDL OG will be hard at work this
weekend trying to come up with a fix for this ;) In the meantime, I'll be
forwarding my complaint to RSI UK!
Cheers,
Randall
PS: David, on what page of your book is this bizarre IDL widget behavior
described? Surely you have an insightful comment to add? After all, it
was _your_ book that convinced me that widget-based viewers using object
graphics was the way forward. Note, of course, that I do not hold _you_
responsible for my duress as I try and view matrices using shad_surf and
for-loops. Surprisingly, the 'window refresh' actually reminds me of
making little cartoon flip-books in grade school. I am now beginning to
long for the days when my only concern was how to best dispose of the 30
celery sticks my mom packed in my lunch...
On Fri, 10 May 2002, Rick Towler wrote:
> Ahhh, widgets....
>
> Issuing the STOP command stops xmanager's ability to process events. Since
> OG widget programs rely on events to draw the view to the window this never
> happens and you get the black window of despair. Direct graphics widget
> programs will possibly draw the initial plot but will not be able to process
> events either.
>
> I can't explain the details but I'm sure someone else in the group can.
>
> A non widget program will run as expected after issuing a stop. I created a
> simple non-widget OG program that I could pass data to plot after issuing a
> stop. You could craft yourself some nice little tools to do line and
> surface plots which automagically scale and rotate objects. You lose the
> ability to manipulate them and cleanup is awkward. You would probably have
> to use READ to wait for a keypress before destroying the objects.
>
> Or maybe you just look at your data using DG. So that's why they are
> keeping DG around....
>
> -Rick
>
>
> "Randall Skelton" <rhskelto@atm.ox.ac.uk> wrote in message
> news:Pine.LNX.4.33.0205101048540.16883-100000@mulligan.atm.o x.ac.uk...
>> With regard to more context, imagine a procedure aptly named 'test':
>>
>> PRO TEST
>> DATA = BESELJ(SHIFT(DIST(40),20,20)/2,0)
>> STOP
>> END
>>
>> When I run this in IDL I get:
>>
>> IDL> test
>> % Compiled module: TEST.
>> % Compiled module: DIST.
>> % Stop encountered: TEST 5 /blah/.../blah/test.pro
>>
>> If I try and plot 'data' with an object graphics routine like
>> 'fsc_surface', or any of the RSI example code 'show3_track',
>> 'test_surface' all I get is a black window? Moreover, all of these
>> routines have built-in data (in the event that the user doesn't pass an
>> array) and even this fails to display when my routine is stopped. Of
>> course, once I continue the program and allow it to finish, everything is
>> fine and the object-graphics window fills with the expected surface.
>>
>> Any help would be greatly appreciated!
>>
>> Cheers,
>> Randall
>>
>> IDL 5.3 (soon to be 5.5); RH Linux 7.x; XFree 4.0.3
>>
>> On Fri, 3 May 2002, Rick Towler wrote:
>>
>>> Hi Randall,
>>>
>>> Can you provide some more context? I just stopped an OG program,
>>> manipulated objects and drew the window. Is that what you are trying to
> do?
>>>
>>> -Rick
>>>
>>>
>>> "Randall Skelton" <rhskelto@atm.ox.ac.uk> wrote in message
>>> news:Pine.LNX.4.33.0205021605500.27169-100000@mulligan.atm.o x.ac.uk...
>>>> Hi all,
>>>>
>>>> I'm using IDL 5.3 under linux and I have a curious question. For
>>>> perspective, I tend to do most of my coding from the command line and
> not
>>>> the gui (yes, I do use the 'stop' command a lot). I routinely check
> the
>>>> contents of variables with 'print' and 'plot' commands to determine if
> my
>>>> calculations look reasonable. My problem is, once I have issued a
> 'stop'
>>>> command I cannot use object graphics to plot anything. All I get is a
>>>> black background window that is usually dissociated from xmanager.
> I'm
>>>> sure there is a good reason for this but I'm stymied. Any thoughts?
>>>>
>>>> Cheers,
>>>> Randall
>>>>
>>>
>>>
>>>
>>
>
>
>
|
|
|
|
| Re: Object Graphics when using 'stop' [message #30677 is a reply to message #30555] |
Fri, 10 May 2002 13:39 |
Paul Sorenson
Messages: 48
Registered: May 2002
|
Member |
|
|
The offending widget programs are written in such a way that they never draw
their graphics until they receive an expose event. xObjView currently has
the same weakness. Perhaps a clever fix could be put into these programs.
Until then here is a lame hack I use. Its a wrapper to xObjView.
pro dbov, oObj, _extra=e
;
;Procedure dbov: "debug objview". Provide an invocation of xobjview
;that is useful when IDL is stopped (e.g. at a breakpoint). Widget
;programs don't run (i.e. respond to events) when IDL is stopped.
;xObjView is a widget program. If it is invoked when IDL is stopped,
;its graphic window remains blank because it is waiting for an expose
;event which never comes. Procedure dbov forces an extra draw to
;guarantee that the initial view of oObj is drawn.
;
xobjview, oObj, tlb=tlb, _extra=e
xobjview, refresh=tlb
end
Paul Sorenson
Gun For Hire
aardvark62@msn.com
"Rick Towler" <rtowler@u.washington.edu> wrote in message
news:abgtdg$1kks$1@nntp6.u.washington.edu...
> Ahhh, widgets....
>
> Issuing the STOP command stops xmanager's ability to process events.
Since
> OG widget programs rely on events to draw the view to the window this
never
> happens and you get the black window of despair. Direct graphics widget
> programs will possibly draw the initial plot but will not be able to
process
> events either.
>
> I can't explain the details but I'm sure someone else in the group can.
>
> A non widget program will run as expected after issuing a stop. I created
a
> simple non-widget OG program that I could pass data to plot after issuing
a
> stop. You could craft yourself some nice little tools to do line and
> surface plots which automagically scale and rotate objects. You lose the
> ability to manipulate them and cleanup is awkward. You would probably
have
> to use READ to wait for a keypress before destroying the objects.
>
> Or maybe you just look at your data using DG. So that's why they are
> keeping DG around....
>
> -Rick
>
>
> "Randall Skelton" <rhskelto@atm.ox.ac.uk> wrote in message
> news:Pine.LNX.4.33.0205101048540.16883-100000@mulligan.atm.o x.ac.uk...
>> With regard to more context, imagine a procedure aptly named 'test':
>>
>> PRO TEST
>> DATA = BESELJ(SHIFT(DIST(40),20,20)/2,0)
>> STOP
>> END
>>
>> When I run this in IDL I get:
>>
>> IDL> test
>> % Compiled module: TEST.
>> % Compiled module: DIST.
>> % Stop encountered: TEST 5 /blah/.../blah/test.pro
>>
>> If I try and plot 'data' with an object graphics routine like
>> 'fsc_surface', or any of the RSI example code 'show3_track',
>> 'test_surface' all I get is a black window? Moreover, all of these
>> routines have built-in data (in the event that the user doesn't pass an
>> array) and even this fails to display when my routine is stopped. Of
>> course, once I continue the program and allow it to finish, everything
is
>> fine and the object-graphics window fills with the expected surface.
>>
>> Any help would be greatly appreciated!
>>
>> Cheers,
>> Randall
>>
>> IDL 5.3 (soon to be 5.5); RH Linux 7.x; XFree 4.0.3
>>
>> On Fri, 3 May 2002, Rick Towler wrote:
>>
>>> Hi Randall,
>>>
>>> Can you provide some more context? I just stopped an OG program,
>>> manipulated objects and drew the window. Is that what you are trying
to
> do?
>>>
>>> -Rick
>>>
>>>
>>> "Randall Skelton" <rhskelto@atm.ox.ac.uk> wrote in message
>>> news:Pine.LNX.4.33.0205021605500.27169-100000@mulligan.atm.o x.ac.uk...
>>>> Hi all,
>>>>
>>>> I'm using IDL 5.3 under linux and I have a curious question. For
>>>> perspective, I tend to do most of my coding from the command line
and
> not
>>>> the gui (yes, I do use the 'stop' command a lot). I routinely check
> the
>>>> contents of variables with 'print' and 'plot' commands to determine
if
> my
>>>> calculations look reasonable. My problem is, once I have issued a
> 'stop'
>>>> command I cannot use object graphics to plot anything. All I get is
a
>>>> black background window that is usually dissociated from xmanager.
> I'm
>>>> sure there is a good reason for this but I'm stymied. Any thoughts?
>>>>
>>>> Cheers,
>>>> Randall
>>>>
>>>
>>>
>>>
>>
>
>
-----------== Posted via Newsgroups.Com - Uncensored Usenet News ==----------
http://www.newsgroups.com The #1 Newsgroup Service in the World!
-----= Over 100,000 Newsgroups - Ulimited Fast Downloads - 19 Servers =-----
|
|
|
|
| Re: Object Graphics when using 'stop' [message #30681 is a reply to message #30555] |
Fri, 10 May 2002 09:44 |
Rick Towler
Messages: 821
Registered: August 1998
|
Senior Member |
|
|
Ahhh, widgets....
Issuing the STOP command stops xmanager's ability to process events. Since
OG widget programs rely on events to draw the view to the window this never
happens and you get the black window of despair. Direct graphics widget
programs will possibly draw the initial plot but will not be able to process
events either.
I can't explain the details but I'm sure someone else in the group can.
A non widget program will run as expected after issuing a stop. I created a
simple non-widget OG program that I could pass data to plot after issuing a
stop. You could craft yourself some nice little tools to do line and
surface plots which automagically scale and rotate objects. You lose the
ability to manipulate them and cleanup is awkward. You would probably have
to use READ to wait for a keypress before destroying the objects.
Or maybe you just look at your data using DG. So that's why they are
keeping DG around....
-Rick
"Randall Skelton" <rhskelto@atm.ox.ac.uk> wrote in message
news:Pine.LNX.4.33.0205101048540.16883-100000@mulligan.atm.o x.ac.uk...
> With regard to more context, imagine a procedure aptly named 'test':
>
> PRO TEST
> DATA = BESELJ(SHIFT(DIST(40),20,20)/2,0)
> STOP
> END
>
> When I run this in IDL I get:
>
> IDL> test
> % Compiled module: TEST.
> % Compiled module: DIST.
> % Stop encountered: TEST 5 /blah/.../blah/test.pro
>
> If I try and plot 'data' with an object graphics routine like
> 'fsc_surface', or any of the RSI example code 'show3_track',
> 'test_surface' all I get is a black window? Moreover, all of these
> routines have built-in data (in the event that the user doesn't pass an
> array) and even this fails to display when my routine is stopped. Of
> course, once I continue the program and allow it to finish, everything is
> fine and the object-graphics window fills with the expected surface.
>
> Any help would be greatly appreciated!
>
> Cheers,
> Randall
>
> IDL 5.3 (soon to be 5.5); RH Linux 7.x; XFree 4.0.3
>
> On Fri, 3 May 2002, Rick Towler wrote:
>
>> Hi Randall,
>>
>> Can you provide some more context? I just stopped an OG program,
>> manipulated objects and drew the window. Is that what you are trying to
do?
>>
>> -Rick
>>
>>
>> "Randall Skelton" <rhskelto@atm.ox.ac.uk> wrote in message
>> news:Pine.LNX.4.33.0205021605500.27169-100000@mulligan.atm.o x.ac.uk...
>>> Hi all,
>>>
>>> I'm using IDL 5.3 under linux and I have a curious question. For
>>> perspective, I tend to do most of my coding from the command line and
not
>>> the gui (yes, I do use the 'stop' command a lot). I routinely check
the
>>> contents of variables with 'print' and 'plot' commands to determine if
my
>>> calculations look reasonable. My problem is, once I have issued a
'stop'
>>> command I cannot use object graphics to plot anything. All I get is a
>>> black background window that is usually dissociated from xmanager.
I'm
>>> sure there is a good reason for this but I'm stymied. Any thoughts?
>>>
>>> Cheers,
>>> Randall
>>>
>>
>>
>>
>
|
|
|
|
comp.lang.idl-pvwave archive
Members
Search
Help
Login
Home
