McIDAS-V Support Forums

Hi,

Could you verify that you can still list adde server information in scripting with accounting being extracted from the server manager rather than being explicitly typed in the script? I am using Version: 1.7beta1 Build-date: 2017-12-13 16:30, but had this problem with a nightly from 2017-11-01.

Joleen

Hi Joleen -

I quick checked this using today's nightly on Windows 7 and OS X. I was able to run scripts with loadADDEImage, listADDEImages, and listADDEImageTimes pointing to datasets on satbuf1.ssec.wisc.edu and msg.ssec.wisc.edu. I didn't include accounting information in the scripts and they worked as expected since the datasets were defined in my server manager.

Are you seeing this problem with all servers/datasets, or only certain ones?

Thanks -
Bob Carp

Hi Bob,

I am seeing this with all the servers, but I am seeing some strange messages when I try to run a script. I recently tried initializing a git repo in my jython directory. It seemed okay at the time, but I am wondering if this is the problem. I thought that it was okay to do this now (an inquiry was addressed which would allow the users to do this).

That being said, I set up a new account on a different machine, cloned my jython repository and that account seems to be running scripts without error. I am doing the same thing on that machine, using the same code, with directories appropriate to that machine.

Joleen

Hi Bob,

I tried the scripting option again with tempuserpath (but not filling yet with my jython code). This eliminates the problem I saw above. I also tried with userpath. I found that I was accidentally using the scripting flags incorrectly, when I typed:



I see an error processing arguments in my mcidasv.log

when I typed:



A McIDAS-V session opens with no error in the log.

When I type the line correctly, the script runs for



The script runs for userpath (no python code yet in temp path), but I get JythonManager errors. I am going to have to dig further into my code. I think that the errors are coming from recent changes and I will just have to revert back to an older version. I will keep you updated if that is not the case.

Thanks,
Joleen

Hi Bob,

The problem was the section header from my config file. If I remove the space (test_recent rather than test recent, McIDAS-V works as expected with the code.

Joleen

Thanks for following up, Joleen. I'm replicating the problem of using both -tempuserpath and -script at startup on OS X (though this does work on Windows). I wrote this up as Inquiry 2615.

- Bob

Hi Bob,

Another consideration for that inquiry might be a warning when the user is trying to add a path when using the -tempuserpath flag. I was using the flag incorrectly, but there is no warning that a problem might occur. I wonder if it would be a good idea to issue a warning if the user is trying to enter a path value. I don't think this is a big deal if only the -tempuserflag is being used, but could be an issue if more than one flag is being used.

Joleen

Thanks, Joleen. This is a great point. On OS X, there's no error when trying to start up with -tempuserpath and a directory. In the terminal, you'll see that McV is starting up in a temporary directory. For example, it may say:

"Using randomly generated userpath: /var/folders/py/fcjksaofwiwof/T/mcidasv.00aptgv6"

However, most users probably wouldn't notice this since there is other text in the terminal window at startup as well. On Windows 7, you actually get an error at startup if -tempuserpath and a path are both defined. This is what should probably be happening on OS X as well to help avoid confusion.

I updated Inquiry 2615 to include this new bug.

Thanks again -
Bob