[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

19991103: NAWIPS compile busts on udunits...f77 eprintf message

Chris,

You may have to use gcc to link the programs since gcc compiled the
udunits library. eprintf is a c library routine and not in
the utlib.c source.

In general, the only programs you likely need are dcacars and dcncprof
which do not use udunits, and are not part of that build.

You can build those programs with:

cd $NAWIPS/unidata/ldmbridge/dcacars
make clean
make all
make install
make clean

and
 
cd $NAWIPS/unidata/ldmbridge/dcncprof
make clean
make all
make install
make clean

The programs built with WANTNETCDF=yes are the grid programs for
reading NetCDF grid files produced by gribtonc. Unfortunately,
most other gridded netcdf files don't follow those conventions,
so I haven't had much call for those programs.

I did include a conversion program:
$NAWIPS/unidata/ldmbridge/dcreanal
which converts a NCAR/NCEP reanalysis netcdf file to a Gempak file
(does not require udunits either).
Dcreanal will create monthly GEMPAK files from the reanalysis NetCDF files.
Note that Gempak does not handle Gaussian (variable spaced rows) grids
which some of the reanalysis data is on.

Steve Chiswell



>From: "C. Vandersip" <address@hidden>
>Organization: .
>Keywords: 199911040228.TAA14011

>Hey Steve,
>
>I'm finally getting around to building the latest Nawips and have run into
>a snag.  
>
>The OS is Solaris 7 using SC5.0.  I'm incorporating (or trying to :))
>netcdf and udunits, so I've got WANTCDF=YES.  I'm getting a symbol
>referncing error under f77 when it resources udunits in
>netcdf/programs/gd/gdcntr. I've checked the archives and seen a few older
>relevant Qs about this, but I can't believe that with the latest OS and
>compilers that make would complain.  BTW, I did have to build udunits
>using gcc (2.8.1) because SC5.0 cc did not like the perl section of the
>make.  Might this make a difference?  Thanks for the help.  Here is the
>relevant output from make.out:
>
>f77 -O -xildoff GDCNTR.f gdcntr.a
>/u1/users/gempak/nawips/unidata/lib/sol/*.o
>/u1/users/gempak/nawips/unidata/lib/sol/gemcdf.a
>/u1/users/gempak/nawips/lib/sol/gemlib.a
>/u1/users/gempak/nawips/lib/sol/appl.a
>/u1/users/gempak/nawips/lib/sol/syslib.a
>/u1/users/gempak/nawips/unidata/lib/sol/gnet.a
>/u1/users/gempak/nawips/unidata/lib/sol/nnet.a
>/u1/users/gempak/udunits-1.11.7/lib/libudunits.a
>/u1/users/gempak/netcdf-3.4/lib/libnetcdf.a
>/u1/users/gempak/nawips/lib/sol/gemlib.a -lm -lsocket -lnsl -lgen -o
>GDCNTR
>GDCNTR.f:
> MAIN gdcntr:
>"GEMINC:GEMPRM.PRM", line 20: Warning: local variable "ermiss" never used
>Undefined                       first referenced
> symbol                             in file
>__eprintf
>/u1/users/gempak/udunits-1.11.7/lib/libudunits.a(utlib.o)
>ld: fatal: Symbol referencing errors. No output written to GDCNTR
>*** Error code 1
>make: Fatal error: Command failed for target `GDCNTR'
>Current working directory
>/u1/users/gempak/nawips-5.4/unidata/netcdf/programs/gd/gdcntr
>*** Error code 1
>make: Fatal error: Command failed for target `all'
>Current working directory
>/u1/users/gempak/nawips-5.4/unidata/netcdf/programs/gd
>*** Error code 1
>make: Fatal error: Command failed for target `all'
>Current working directory
>/u1/users/gempak/nawips-5.4/unidata/netcdf/programs
>*** Error code 1
>make: Fatal error: Command failed for target `all'
>Current working directory /u1/users/gempak/nawips-5.4/unidata/netcdf
>*** Error code 1
>make: Fatal error: Command failed for target `all'
>Current working directory /u1/users/gempak/nawips-5.4/unidata
>*** Error code 1
>make: Fatal error: Command failed for target `all'
>
>Regards,
>
>Chris
>
>        ###############################################################
>        #                      Chris Vandersip                        #
>        #        Computer Research Specialist/Dept. Sysadmin          #
>        #  Rm. 024, Dept. of Meteorology, Florida State University    #
>        #          address@hidden   (850)644-2522                     #
>        ###############################################################
>