> Hi Charlie ,
> I see you have removed the nc_long's and the NC_LONG's
Yes, I changed NC_LONG to NC_INT since NC_LONG is deprecated.
I changed nc_long to nco_long for reasons explained in the ChangeLog.
Note that since we store NC_INTs as intrinsic longs internally to NCO, the
get_ and put_ functions still need to call the _long versions,
it's just that the on-disk storage type is NC_INT instead of NC_LONG.
I know that sounds confusing!
In the future netCDF4.x will presumably have a 64 bit integer type
called NC_LONG so it hasn't gone away forever.
> Prior to realise I think the new version needs serious testing -
> I'm sure my changes have introduced a few subtle bugs which the
> test scripts have not picked up on.--- Is there a group of heavy users
> we could
> introduce the new version to ? .
That's probably a good idea but I do not have the energy
to organize beta testing. This is why some software packages
have 1.9.x releases. But our project is so small I'm not sure
it's worth the hassle. Buyer Beware!
> Is the increase in speed with netCDF3 noticeable ?
I'm very curious to find out, and I think it's on the TODO
list. But again, I'm looking for volunteers to do the
benchmarks.
> Prior to release I could tidy up the calls to nco_inq_var() and
> nco_inq_dim()--Use the
> more specific calls eg nco_inq_var_type , nco_inq_var_ndim etc ???
>
This does not need to be done prior to release,
but would certainly be welcome in 2.0.1 etc.
> Regards Henry
>
> ----- Original Message -----
> From: Charlie Zender <zender@uci.edu>
> To: Henry Butowsky <thin@thin.screaming.net>
> Sent: Monday, April 02, 2001 3:30 PM
> Subject: nco 2.0
>
>
>
>> Hi Henry,
>>
>> Do you have any mods you want to go into the next nco release?
>>
>> I've decided that the next NCO release should be named NCO 2.0
>> instead of 1.4.
>> I want to cleanly separately the versions which support netcdf2.x
>> from netcdf3.x, and this is a good time to do it. I'm almost ready
>> to tag NCO 2.0.
>>
>> Charlie
>> --
>> Charlie Zender zender@uci.edu (949) 824-2987/FAX-3256, Department of
>> Earth System Science, University of California, Irvine CA 92697-3100
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Henry M Butowsky wrote:
> Hi Charlie ,
> I see you have removed the nc_long's and the NC_LONG's
Yes, I changed NC_LONG to NC_INT since NC_LONG is deprecated.
I changed nc_long to nco_long for reasons explained in the ChangeLog.
Note that since we store NC_INTs as intrinsic longs internally to NCO, the
get_ and put_ functions still need to call the _long versions,
it's just that the on-disk storage type is NC_INT instead of NC_LONG.
I know that sounds confusing!
In the future netCDF4.x will presumably have a 64 bit integer type
called NC_LONG so it hasn't gone away forever.
> Prior to realise I think the new version needs serious testing -
> I'm sure my changes have introduced a few subtle bugs which the
> test scripts have not picked up on.--- Is there a group of heavy users
> we could
> introduce the new version to ? .
That's probably a good idea but I do not have the energy
to organize beta testing. This is why some software packages
have 1.9.x releases. But our project is so small I'm not sure
it's worth the hassle. Buyer Beware!
> Is the increase in speed with netCDF3 noticeable ?
I'm very curious to find out, and I think it's on the TODO
list. But again, I'm looking for volunteers to do the
benchmarks.
> Prior to release I could tidy up the calls to nco_inq_var() and
> nco_inq_dim()--Use the
> more specific calls eg nco_inq_var_type , nco_inq_var_ndim etc ???
>
This does not need to be done prior to release,
but would certainly be welcome in 2.0.1 etc.
> Regards Henry
>
> ----- Original Message -----
> From: Charlie Zender <zender@uci.edu>
> To: Henry Butowsky <thin@thin.screaming.net>
> Sent: Monday, April 02, 2001 3:30 PM
> Subject: nco 2.0
>
>
>
>> Hi Henry,
>>
>> Do you have any mods you want to go into the next nco release?
>>
>> I've decided that the next NCO release should be named NCO 2.0
>> instead of 1.4.
>> I want to cleanly separately the versions which support netcdf2.x
>> from netcdf3.x, and this is a good time to do it. I'm almost ready
>> to tag NCO 2.0.
>>
>> Charlie
>> --
>> Charlie Zender zender@uci.edu (949) 824-2987/FAX-3256, Department of
>> Earth System Science, University of California, Irvine CA 92697-3100