This archive contains answers to questions sent to Unidata support through mid-2025. Note that the archive is no longer being updated. We provide the archive for reference; many of the answers presented here remain technically correct, even if somewhat outdated. For the most up-to-date information on the use of NSF Unidata software and data services, please consult the Software Documentation first.
>From: Jim Koermer <address@hidden> >Organization: Plymouth State College >Keywords: 200202280553.g1S5rxx20688 McIDAS scripts MCTABLE_READ MCTABLE_WRI Jim, >I used your script verbatim, but still get the following output: > >%./mc_test > >Group Name Server IP Address >-------------------- ---------------------------------------- >DATALOC -- done > cannot locate string names - rerun setup. >imgdisp.k: Image data server unable to resolve this dataset: >NEXRCOMP/1KN0R-NAT >imgdisp.k: done > cannot locate string names - rerun setup. >Frame saved in /home/jim/nowrad.gif >% > >I get the same thing is I run "mcenv -f 512x640 mc_test". > >If I start up mcidas and try running the DATALOC in command, I never get >it setting the location of NEXRCOMP to the IP, I just get similar >messages to that above. I logged onto pscwx and verified that (with one minor change I will point out below) that I could run my example script as the user 'mcidas'. I then became you and verified your observation that the script would cause a number of 'cannot locate string names - rerun setup.' errors and go nowhere. I found that the cause of the problem was in your .cshrc file. What happens is that McIDAS commands will source the .cshrc file for users running in the C shell. Part of the setup in one's .cshrc file is to put a guard around the resetting of the MCPATH set of directories. The block of code that sets MCDATA, MCPATH, etc. in your .cshrc file should look like: if ( ! ${?MCPATH} ) then setenv MCDATA $HOME/mcidas/data setenv MCPATH $HOME/mcidas/data:/home/mcidas/data:/data/mcidas-c:/home/mcidas/help setenv MCGUI /home/mcidas/bin setenv MCTABLE_READ "$MCDATA/MCTABLE.TXT;$MCHOME/data/ADDESITE.TXT" setenv MCTABLE_WRITE "$MCHOME/data/ADDESITE.TXT" endif The 'if ( ! ${?MCPATH} ) then' construct says that if MCPATH is already defined then skip all setenv invocations in the rest of the if block. This is absolutely necessary for C shell users. After adding the guard in your .cshrc file, I copied over the updated version of mc_test from the ~mcidas directory and ran it with no problems in your account: <login as you> cd ~ ./mc_test3 -------------------- ---------------------------------------- NEXRCOMP PSCWX.PLYMOUTH.EDU <LOCAL-DATA> indicates that data will be accessed from the local data directory.DATALOC -- done Beginning Image Data transfer, bytes= 79364 IMGDISP: loaded frame 1 EG;MAP H imgdisp.k: done EU: Restoring BREF24.ET to frame(s)= 1 EU: Done Erased graphic frame(s) 1-1 MAP: Completed frame 1 Frame saved in /home/jim/nowrad.gif Ok, the small change I made in my example script was to put a wait after the imgdisp.k command. Here is the altered code: mcenv << EOF dataloc.k ADD NEXRCOMP pscwx.plymouth.edu imgdisp.k NEXRCOMP/1KN0R-NAT STA=TWX EU=BREF24 MAG=2 REFRESH='EG;MAP H' wait.k 3 frmsave.k 1 "/home/jim/nowrad.gif" EOF The reason the wait was needed is that imgdisp.k runs the commands listed in the REFRESH= keyword asynchronously: imgdisp.k exits before EG and MAP are run, and the framsave.k line is executed very fast. The upshot of this is that one gets to the EOF line and exits mcenv BEFORE the EG and MAP commands have run. When mcenv exits, the frame into which the display was made is removed and EG and MAP will then issue errors. The other way around this "timing" issue is to run the MAP command by itself: mcenv << EOF dataloc.k ADD NEXRCOMP pscwx.plymouth.edu imgdisp.k NEXRCOMP/1KN0R-NAT STA=TWX EU=BREF24 MAG=2 eg.k map.k H frmsave.k 1 "/home/jim/nowrad.gif" EOF A couple of quick comments: When loading the 1 km resolution reflectivity product, it is best to use a MAG=-5 first to get the big picture. After finding out where the action is, then you can bore down on the interesting echos using less and less blow-down. For instance, today the action is down at the Gulf Coast in LA. I would change the mc_test3 code to: mcenv << EOF dataloc.k ADD NEXRCOMP pscwx.plymouth.edu imgdisp.k NEXRCOMP/1KN0R-NAT LATLON=30 93 MAG=1 EU=BREF24 eg.k map.k H frmsave.k 1 "/home/jim/nowrad.gif" EOF to get a really nice picture of the activity down there. Tom