script creating images with getADDEImage stopped working

[hproe]

Hi -

The attached method worked flawlessly in 1.3beta1 builds. It does not work any more under 1.3 and 1.4beta1, i.e. it trough the image sequence but at the end the display remains blank with a colour bar of range missing-missing and the correct number of green boxes on top.

cheers, HP

[Jon]

Hi HP,

Apologies for the bug. I'm looking into this now and am having some trouble replicating the problem, but I will keep investigating.

[ removed some useless advice…sorry! ]

Jon

[Jon]

Okay, now I have some better advice…though I will admit that I don't know if this will fix things for you.

I've attached a modified version of your code that defines a new function, "test_createLayer(...)", and took the liberty of replacing the "imDisplay.createLayer(...)" call in your function with a call to this new one. I did this because we introduced a new mechanism for scheduling calls to methods/functions like "_Display.createLayer"…but perhaps it is not as benign as we suspected. At any rate, the new function is essentially the current "_Display.createLayer" code with the scheduling thing removed--so here's hoping this does the trick!

Again, sorry for the bug.

Jon

[hproe]

Many thanks, Jon. I am sorry to say that I do not see any change when running 1.4beta1 of 26 June. Still no display.

HP

[tomw]

HP -- Running on 64-bit Win7, both McV1.3 and McV1.4b1 -- I copied your .py file into my Jython Library, then from the Jython Shell I ran:
a=sequence_singBand("2013-06-25 12:00UTC",3,1)

and it displayed a 3-image sequency just fine. I also tried Jon's modified routine and it also displays okay.

I assume you are able to load one image from the target ADDE server without a problem. Can you rename the McIDAS-V directory and start McV up again to see if there is some "environment" difference between our machines?

[hproe]

Tom & Jon -

I am running W7 64bit. However, starting with a maiden McIDAS-V and just adding the Jon's and my *.py does not display the images either. I have seen two details, though.
1. The drop-down list with the time slots shows n times the first slot (n= number of slots).
2. The script displays the image correctly when only one time slot is called.

Both items are true irrespective of using Jon's method or not. Obviously, there is a problem when having to display more than one image.

HP

[jayh]

Hi HP-

I started a 1.3 session with a new McIDAS-V directory and followed TomW's steps, and my display looked good with 3 images. Here is the Jython Shell output:

a=sequence_singBand("2013-06-25 12:00UTC",3,1)
0
url: adde://geoarc.ssec.wisc.edu/imagedata?& ... &LATLON=50 35 E&PLACE=CENTER&SIZE=900 1200&UNIT=TEMP&MAG=1 1&SPAC=4&NAV=X&AUX=YES&DOC=X&DAY=2013176&TIME=12:00:00 12:00:00 I&POS=0&TRACK=1
lines=900 elements=1200
1
url: adde://geoarc.ssec.wisc.edu/imagedata?& ... &LATLON=50 35 E&PLACE=CENTER&SIZE=900 1200&UNIT=TEMP&MAG=1 1&SPAC=4&NAV=X&AUX=YES&DOC=X&DAY=2013176&TIME=12:15:00 12:15:00 I&POS=0&TRACK=1
lines=900 elements=1200
2
url: adde://geoarc.ssec.wisc.edu/imagedata?& ... &LATLON=50 35 E&PLACE=CENTER&SIZE=900 1200&UNIT=TEMP&MAG=1 1&SPAC=4&NAV=X&AUX=YES&DOC=X&DAY=2013176&TIME=12:30:00 12:30:00 I&POS=0&TRACK=1
lines=900 elements=1200
done

I'm not sure where to check next for what might be the problem. I was wondering if you are using the Eumetcast version? That's the only other difference I could think of checking for...

Thanks, Jay