NOTICE: This version of the NSF Unidata web site (archive.unidata.ucar.edu) is no longer being updated.
Current content can be found at unidata.ucar.edu.

To learn about what's going on, see About the Archive Site.

Re: [netcdfgroup] does anyone mind if the fortran 77 library always gets its own file, instead of being combined with the C library?

  • To: netcdfgroup@xxxxxxxxxxxxxxxx
  • Subject: Re: [netcdfgroup] does anyone mind if the fortran 77 library always gets its own file, instead of being combined with the C library?
  • From: Magnus Hagdorn <Magnus.Hagdorn@xxxxxxxx>
  • Date: Tue, 29 Mar 2011 09:28:18 +0100
Ed Hartnett wrote:
Howdy netCDFians!

In the old days the netCDF F77 library was built into the netCDF C
library. (This seemed natural, because it was really just a bunch of C
functions anyway.) Fortran programs using the F77 API, and C programs
linked with -lnetcdf.

This idea does not work with shared libraries, so when shared libraries
are built (now the default) a separate F77 library is built. Now C
library users still link with -lnetcdf, but F77 library users must use
-lnetcdff -lnetcdf. (Note the extra "f" at the end of the F77 library
name.)

Maintaining the ability to combine the C and F77 library is expensive in
terms of programmer time. Will anyone object if I remove this option?

Users will see no impact on their code. They will have to add
"-lnetcdff" to their Makefiles for F77 programs. (This is already true
for shared library builds.)

Any comment or feedback welcome...



Hi Ed,
does that mean you cannot have mixed C/Fortran code where both languages access a netCDF library? I don't have such a beast, but I am curious. Also, this might affect mixed python/fortran codes.
Cheers
magnus

--
The University of Edinburgh is a charitable body, registered in
Scotland, with registration number SC005336.



  • 2011 messages navigation, sorted by:
    1. Thread
    2. Subject
    3. Author
    4. Date
    5. ↑ Table Of Contents
  • Search the netcdfgroup archives: