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 Note that the archive is no longer being updated. We provide the archive for reference; many of the answers presented here remain technically correct, even if somewhat outdated.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.
>>To: address@hidden >>From: address@hidden >>Organization: Michigan State University >>Keywords: 199501121652.AA06934 In the above message you wrote: >... But I do not plan on using the FORTRAN interface to >netCDF, so I like to just disable that. I go into CUSTOMIZE and attempt >to set "FC" to nothing > >>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>> >CUSTOMIZE >>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>> > ># This file is for customizing the configuation process performed by ># `./configuration'. This file consists of sh(1) variable-definition lines. ># The value given to those variables by this file will override their default ># values. ># ># You can also customize the configuration process via the environment ># variables seen by ./configure. For example: ># ># In csh(1): ># % setenv CC acc ># & setenv CFLAGS -g ># % ./configure ># ># In sh(1): ># $ CC=acc CFLAGS=-g ./configure ># ># Variables in this file override the environmental ones. ># >############################################################################# ># ># Type of C compiler (default is `cc'): ># CC='gcc -traditional' ># ># C compiler flags (default is `-O'): ># CFLAGS=-g ># ># C preprocessor flags (default is `-DNDEBUG'): ># CPPFLAGS=-DDEBUG ># ># Installation prefix (default is `..'): ># prefix=/usr/local/unidata ># ># Whether to use the old floating-point fill values for backward ># compatibility (see INSTALL). Uncomment to obtain the pre-2.3 default, ># floating-point fill values. ># OLD_FILLVALUES=yes > >FC="" > >>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>> >>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>> > > >I also tried > >FC='' > >after doing "make distclean" and running configure after that > >That didn't work. You are correct: setting the FC variable to the empty string doesn't disable the FORTRAN interface in the netCDF 2.3.2 distribution. For what it's worth, it is fixed in the next release. I suggest manually editing the file fortran/Makefile and setting the FC macro to the empty string. You'll find the definition on line 82. Change it from FC = f77 to FC = Then, go to the top-level source directory and execute `make all test'. You shouldn't be bothered by anything FORTRAN. If you rerun the `configure' script, then you'll have to reset the FC variable to the empty string again. For a more permanent solution, you can zero it in the master makefile port/master.mk.in. -------- Steve Emmerson <address@hidden>