September 23, 2021, 12:06:32 AM

News:

Due to SPAM attacks, new members must be approved before posting.  Please email jclough@warrenpinnacle.com when registering and your account will be approved.


Issues with screen input/output

Started by kfreeman, April 19, 2010, 11:04:39 AM

previous topic - next topic
Go Down

kfreeman

I have come across 3 issues with screen input and output.  A scenario with 5 year time steps until 2040 was run and saved to a .SLAMM6 file.  I used Load Simulation to load it again, and then changed the time step to 25 years and end date to 2100.  My eye happened to catch the fact that the 2100 got changed back to 2040!  I tried a few more times to change it to 2100, but it  would not stay.  I let the run continue (25 year time step to 2040).   It did process to year 2025, and then got  a range check error.  I have gotten range check error  before on previously  working simulation runs, and I am wondering if this was the problem.

When SLAMM tries to paste a map to Word (2007), it opens Word, then gets an error "Runtime error copying map to Microsoft Word for year X.  Retry writing map to Word? Yes or No?"  Yes would get the error  message again.  No would produce no map.  I found that if you paste a screen shot (or probably any object)  into Word and THEN answer yes, the maps paste without any further problem.

In my model input, I have a  large area coded as Open Ocean.   However, on both the maps pasted into Word and on the screen,  the Open Ocean appears white.  The default Legend has it symbolized as dark blue.  Why is that happening?

Jonathan S. Clough

Will try to reproduce problem 1 and will check the code to see if the "run until" date is being properly loaded and saved to the SLAMM6 file.

With respect to problem 2, sometimes Word is not in a receptive state for pasting.  Sometimes this is because a dialog box is open or another instance of word is actively using the   OLE automation can be finicky, unfortunately.  I would stick with your workaround.  If you have a problem that is reproducible I can try to reproduce it here and see if there is a programmatic workaround that would be more useful.

With respect to problem 3 (open ocean) the model will not load in categories that have "no-data" for elevation and marks these as "blank" cells.  This can be helpful when there is spotty LiDAR data inland, for example. 

Jonathan S. Clough

April 27, 2010, 08:28:32 AM #2 Last Edit: April 27, 2010, 08:33:25 AM by Jonathan S. Clough
I found one bug that could cause this range check error so I have fixed that in the latest version which I can distribute upon request to people having this problem (an official release will require me to do some documentation updates first).  Please let me know if further range-check errors persist and how you arrived at them and I'll fix those too.

Also, I checked the saving of the "last date" on the execute window and it seems to be working fine.  However, if you click the "x" at the upper right of the screen instead of "save and exit" all changes made are lost.  This is not intuitive so for now, I have gotten rid of the "x" forcing users to either choose OK or cancel...

Also, the "save and exit" button never saved to disk, it was just intended to be the opposite of cancelling changes.  So I've changed to a more intuitive OK and Cancel setup.  You still need to save to disk after exiting that screen, though.

-- Jonathan

Jonathan S. Clough


Go Up