You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@sis.apache.org by Martin Desruisseaux <ma...@geomatys.fr> on 2013/05/15 01:12:22 UTC
Current state of NetCDF - ISO mapping
Hello all
Most of the code creating ISO 19115 metadata from the content of a
NetCDF header has been ported, except coordinate reference system stuffs
(because pending sis-referencing module). I still have to port the
tests, do some adjustments, and define a public API (I will post an
other mail on this topic tomorrow). The following page lists all NetCDF
metadata attributes recognized by the current code. One can click on an
attribute for seeing the corresponding ISO 19115 property, together with
a link to the UCAR page where the attribute is defined.
https://builds.apache.org/job/sis-jdk7/site/apidocs/org/apache/sis/storage/netcdf/AttributeNames.html
One word of caution: ISO and NetCDF uses the same words for different
things. For example what NetCDF calls "coordinate system" is actually a
mix of what OGC/ISO specifications call "coordinate system", "coordinate
reference system" and "grid geometry". The NetCDF coordinate system
"range" is closer to ISO 19123 "domain", the NetCDF coordinate system
"domain" is closer to ISO 19123 "grid envelope" and the ISO 19123
"range" is rather related to the NetCDF variable's minimum and maximum
values. Trying to use OGC/ISO and NetCDF objects in the same code
appears to be *very* confusing. The sis-netcdf module uses internally an
isolation layer with all dependencies to the UCAR library regrouped in
the "org.apache.sis.internal.netcdf.ucar" package. This is done that way
both for technical reasons (allowing us to use alternative to UCAR
libraries for relatively simple NetCDF files), and for avoiding to mix
the two vocabularies in the same classes.
Martin
Re: Current state of NetCDF - ISO mapping
Posted by Adam Estrada <es...@gmail.com>.
Hi Martin! I am not as familiar with the NetCDF spec as you are so we will
leave it up to you to make the best judgement on how to proceed. Everything
looks good from where I sit!!
Adam
On Tue, May 14, 2013 at 7:12 PM, Martin Desruisseaux <
martin.desruisseaux@geomatys.fr> wrote:
> Hello all
>
> Most of the code creating ISO 19115 metadata from the content of a NetCDF
> header has been ported, except coordinate reference system stuffs (because
> pending sis-referencing module). I still have to port the tests, do some
> adjustments, and define a public API (I will post an other mail on this
> topic tomorrow). The following page lists all NetCDF metadata attributes
> recognized by the current code. One can click on an attribute for seeing
> the corresponding ISO 19115 property, together with a link to the UCAR page
> where the attribute is defined.
>
> https://builds.apache.org/job/**sis-jdk7/site/apidocs/org/**
> apache/sis/storage/netcdf/**AttributeNames.html<https://builds.apache.org/job/sis-jdk7/site/apidocs/org/apache/sis/storage/netcdf/AttributeNames.html>
>
> One word of caution: ISO and NetCDF uses the same words for different
> things. For example what NetCDF calls "coordinate system" is actually a mix
> of what OGC/ISO specifications call "coordinate system", "coordinate
> reference system" and "grid geometry". The NetCDF coordinate system "range"
> is closer to ISO 19123 "domain", the NetCDF coordinate system "domain" is
> closer to ISO 19123 "grid envelope" and the ISO 19123 "range" is rather
> related to the NetCDF variable's minimum and maximum values. Trying to use
> OGC/ISO and NetCDF objects in the same code appears to be *very* confusing.
> The sis-netcdf module uses internally an isolation layer with all
> dependencies to the UCAR library regrouped in the "org.apache.sis.internal.
> **netcdf.ucar" package. This is done that way both for technical reasons
> (allowing us to use alternative to UCAR libraries for relatively simple
> NetCDF files), and for avoiding to mix the two vocabularies in the same
> classes.
>
> Martin
>
>