> I have having a problem with some of the AWIPS grids. With 5.9.4 I was > getting names for grids like: > > model/ffg/2007091312_ktab255.gem > > now with 5.10.3 built with Intel fortran and C on a linux box I am getting > > model/ffg/2007091312_subc162255.gem > > From doing traces and looking at the code, the problem lies with the > recoding of the routines in gempak/source/gemlib/na into C, especially > nartbl. It is writing into a structure called _nacent. The dcgrib2 > program has a routine called tbsubc.f which is using a Fortran common > block called NACENT and reading an array called mcabb. In this > environment _nacent and NACENT are not the same, thus the mcabb array is > empty. Eventually dcgrib2 tries to read a file called "subcenters.tbl" > instead of "nwssubcenters.tbl". Do you have a fix for this, or should I > do my own (which would be to rewrite tbsubc.f in C)? > > Harry, I did a similary update of the Fortran to C transition of na_rtbl in the dcreanal2 code but missed this instance in dcgrib2. I just updated the Makefile and tblist.c files in dcgrib2 and removed the tbsubc.f routine. Attached are the modified routines. Thanks for pointing this bug out. Steve Chiswell Unidata User Support Ticket Details =================== Ticket ID: AXZ-622793 Department: Support GEMPAK Priority: Normal Status: Closed
Attachment:
Makefile
Description: Binary data
Attachment:
tblist.c
Description: Binary data