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

Re: netCDF for Win32

> To: address@hidden ('Russ Rew')
> Subject: netCDF for Win32
> Organization: Varian Chromatography Systems

Hi Jean-Louis,

> Since you guys obviously did a lot of work on netCDF, I've thrown away my   
> old 2.x code kludged for Windows 3.x and restarted anew, rather than   
> attempting a 32 port of obsolete code.
> 
> I have downloaded the sources for netCDF 3.3.1, applied the MAKS.ZIP   
> patch for Win32, and started rebuilding the code.
> 
> It's working quite well, with the following exceptions, that I have not   
> seen documented on the netCDF Web site (Did I miss it?)
> 
> 
> 1)  Most (if not all) of the MSOFT.MAK make files have the same minor   
> typo:
> 
> LIB = ..\libsrc:LIB  should be LIB = ..\libsrc;LIB   (semicolon instead   
> of colon)

I've forwarded the above to Glenn Davis, since he's the author of MAKS.ZIP.

> 2)  NCGEN can't be built because NCGENTAB,C refers to VALUES.H, which   
> does not exist in MSVC 5.0 - as far as I can see. I removed the reference   
> to values.h in NCGENTAB.C and it seems to compile and work just fine, so   
> may be I don't need it at all ?  What do you think ?

The "#include <values.h>" is generated by yacc, in creating ncgentab.c
from ncgen.y.  I see know way to suppress its generation, but you're
right, it doesn't seem to be needed.  When I delete it from ncgentab.c
and rebuild under Solaris 2.6, everything works fine without it.  If
it's not needed on other systems either, I'll remove it from ncgentab.c
for the next distribution.

--Russ

_____________________________________________________________________

Russ Rew                                         UCAR Unidata Program
address@hidden                     http://www.unidata.ucar.edu