NOTICE: This version of the NSF Unidata web site (archive.unidata.ucar.edu) is no longer being updated.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.
Well it was a bit of a battle, most of it due to my unfamiliarity with the pecularities of Irix. My recommendations for a building on a SGI are; 1) add a variable called SHELL=/bin/sh to the root Makefile and add SHELL to the MYFLAGS variable. (otherwise make won't interpret shell commands). 2) SYS V install is incompatible, I used the install from my ncarg3.0 source. 3) No ranlib in SYS V, use $(AR) m instead 4) I couldn't be bothered wrestling with shared libraries, so define LIBTYPE nonsharable 5) If you want a fortran interface just specify OS to be sunos, which seems to work. All tests were passed except for one formatted test of xdr i.e. root@aqueous {xdr} ? make test 2c2 < bytes: 254 1 0 1 2 0 0 0 --- > bytes: -2 1 0 1 2 0 0 0 *** XDR fails formatted test *** The above differences are OK if small *** XDR passes binary test *** I tried this with both the supplied SGI xdr library ( -lsun ) and the distributed xdr stuff with netcdf. Both gave the same result, and isn't -2 just the 2's compliment of 254 ? Would anybody care to comment on the seriousness of this? Well, time to go and tackle X11R5 on the mongrel... Cheers, ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Bill Hart Internet : hart@xxxxxxxxxxxxxxxxx CSIRO Division of Oceanography Phone : +61 02 206 446 Hobart, Tas., 7000 Australia Fax : +61 02 240 530 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
netcdfgroup
archives: