[Geoserver-devel] Geoserver WCS

On Tue, 22 Mar 2005, Alessio Fabiani wrote:

Chris, a little notice ... if you recompile the package replacing the
gt2-arcgrid and gt2-image, the WCS probably will not work because
Martin has not updated the gt2 trunk with the Simone's patches yet.
However I put the patches in a branch subdirectory.

Hey, first off use the geoserver devel list for things like this. Other
people on the list may be checking out your work - Indeed I am probably
not going to check up on you much until a review for merge. This is a
development list, so don't feel shy about posting . if people want less
traffic they can go to the users list.

  Second, Simone does not yet have commit access? We should get that for
him, as he's doing a bunch of good work, and shouldn't have to wait on
Martin. Has anyone else reviewed his code? I'll send an email to the
list.

Chris

I think the gt2 will be updated soon.

Cheers,
          Alessio.
> On Tue, 22 Mar 2005 13:01:12 +0000, cholmes@anonymised.com
<cholmes@anonymised.com> wrote:
> No worries. I just deleted it. And I announced the branch on the list.
> And do encourage/help out the Norman Baker guy - I think he did some
> good work on the deegree wcs - if you show enough potential with your
> implementation he'll probably be able to help out.
>
> best regards,
>
> Chris
>
>
> Quoting Alessio Fabiani <alessio.fabiani@anonymised.com>:
>
> > Paul, Chris ... I made a mistake, I have updated the trunk. Don't
> > worry, I was aligned with the last version, all working ... but there
> > is the wcs inside. So please, modify the trunk with the old version
> > and leave the wcs version only to the branch.
> >
> > branches/wcs-branch created.
> >
> > Thanks and sorry,
> > Alessio.
> >
> >

Quoting Alessio Fabiani <alessio.fabiani@anonymised.com>:

Sorry for the post, I didn't see geoserver-devel in the CC ... I'll
pay more attention in the future.

No, I said _use_ the devel list. I was the one who added the cc.
People like knowing what work is being done, seeing developers
communicate in an open way.

Simone has not yet right commits. I think that Martin reviews his
code
when ready.

Ok, I just sent an email asking for commit rights for him. It's silly
with the amount of work that he is doing to hold off. If/when GeoTools
has tons of people working on the same code, and gets into trouble
coordinating, we can go with more of the patch style. But since we are
a bit undermanned relative to the size of the codebase, it makes sense
to give good people commit rights relatively quickly (at least that's
my opinion - and I think most of geotools does agree).

Until the new Grid Formats are not ready, I'm working on rendering
GridCoverages in WMS ... my idea was to create a DataStore for
GridCoverages, so it would be possible to create new Featrues (Grid
based) inside GeoServer. Do you think it's the right way? Consider
that, as Martin said, for the new ISO specs the GridCoverages "are"
Features, and Geotools will extend the interfaces for this.

I really don't know coverages well enough to comment. Jody has talked
about a new api for coverages, perhaps he has some ideas. DataStore
for GridCoverages makes sense to me. They are a bit different, so
perhaps they should descend from a common super class - DataStore, and
then we could have CoverageStore and VectorStore or something? It
would be very good to have a coverage api that is easier to work with
than GridCoverageExchange - perhaps after working with that class you
have some ideas to do things better? Maybe a simplified api that can
return features that is based on GridCoverageExchange. Though actually
you could kill that class, I don't think anyone would be very sad to
see it go. One other thing to remember - DataStore was based on WFS,
so similarly CoverageStore or whatever maybe could use WCS as its
inspiration? Also probably worth checking out the classes that
Stephane Fallah did for geoapi.

best regards,

Chris

On Wed, 23 Mar 2005 05:34:58 -0500 (EST), Chris Holmes
<cholmes@anonymised.com> wrote:
> On Tue, 22 Mar 2005, Alessio Fabiani wrote:
> > Chris, a little notice ... if you recompile the package replacing
the
> > gt2-arcgrid and gt2-image, the WCS probably will not work because
> > Martin has not updated the gt2 trunk with the Simone's patches
yet.
> > However I put the patches in a branch subdirectory.
> >
> Hey, first off use the geoserver devel list for things like this.
Other
> people on the list may be checking out your work - Indeed I am
probably
> not going to check up on you much until a review for merge. This
is a
> development list, so don't feel shy about posting . if people want
less
> traffic they can go to the users list.
>
> Second, Simone does not yet have commit access? We should get
that for
> him, as he's doing a bunch of good work, and shouldn't have to wait
on
> Martin. Has anyone else reviewed his code? I'll send an email to
the
> list.
>
> Chris
>
> > I think the gt2 will be updated soon.
> >
> > Cheers,
> > Alessio.
> > > On Tue, 22 Mar 2005 13:01:12 +0000, cholmes@anonymised.com
> > <cholmes@anonymised.com> wrote:
> > > No worries. I just deleted it. And I announced the branch on
the list.
> > > And do encourage/help out the Norman Baker guy - I think he did
some
> > > good work on the deegree wcs - if you show enough potential
with your
> > > implementation he'll probably be able to help out.
> > >
> > > best regards,
> > >
> > > Chris
> > >
> > >
> > > Quoting Alessio Fabiani <alessio.fabiani@anonymised.com>:
> > >
> > > > Paul, Chris ... I made a mistake, I have updated the trunk.
Don't
> > > > worry, I was aligned with the last version, all working ...
but there
> > > > is the wcs inside. So please, modify the trunk with the old
version
> > > > and leave the wcs version only to the branch.
> > > >
> > > > branches/wcs-branch created.
> > > >
> > > > Thanks and sorry,
> > > > Alessio.
> > > >
> > > >
>
>

----------------------------------------------------------
This mail sent through IMP: https://webmail.limegroup.com/

Argh, I keep writing messages to you, and keep forgetting to thank you
for rolling back the WCS stuff properly. I realized that I had done it
wrong right after I left the internet. Thanks for catching it for me,
and saving work for me today.

Also, if the email to geotools devel for commit rights doesn't show up
let me know - the lists seem slow today.

thanks again,

Chris

Quoting Chris Holmes <cholmes@anonymised.com>:

Quoting Alessio Fabiani <alessio.fabiani@anonymised.com>:

> Sorry for the post, I didn't see geoserver-devel in the CC ... I'll
> pay more attention in the future.
No, I said _use_ the devel list. I was the one who added the cc.
People like knowing what work is being done, seeing developers
communicate in an open way.

>
> Simone has not yet right commits. I think that Martin reviews his
> code
> when ready.
Ok, I just sent an email asking for commit rights for him. It's
silly
with the amount of work that he is doing to hold off. If/when
GeoTools
has tons of people working on the same code, and gets into trouble
coordinating, we can go with more of the patch style. But since we
are
a bit undermanned relative to the size of the codebase, it makes
sense
to give good people commit rights relatively quickly (at least that's
my opinion - and I think most of geotools does agree).

>
> Until the new Grid Formats are not ready, I'm working on rendering
> GridCoverages in WMS ... my idea was to create a DataStore for
> GridCoverages, so it would be possible to create new Featrues (Grid
> based) inside GeoServer. Do you think it's the right way? Consider
> that, as Martin said, for the new ISO specs the GridCoverages "are"
> Features, and Geotools will extend the interfaces for this.
I really don't know coverages well enough to comment. Jody has
talked
about a new api for coverages, perhaps he has some ideas. DataStore
for GridCoverages makes sense to me. They are a bit different, so
perhaps they should descend from a common super class - DataStore,
and
then we could have CoverageStore and VectorStore or something? It
would be very good to have a coverage api that is easier to work with
than GridCoverageExchange - perhaps after working with that class you
have some ideas to do things better? Maybe a simplified api that can
return features that is based on GridCoverageExchange. Though
actually
you could kill that class, I don't think anyone would be very sad to
see it go. One other thing to remember - DataStore was based on WFS,
so similarly CoverageStore or whatever maybe could use WCS as its
inspiration? Also probably worth checking out the classes that
Stephane Fallah did for geoapi.

best regards,

Chris

>
>
> On Wed, 23 Mar 2005 05:34:58 -0500 (EST), Chris Holmes
> <cholmes@anonymised.com> wrote:
> > On Tue, 22 Mar 2005, Alessio Fabiani wrote:
> > > Chris, a little notice ... if you recompile the package
replacing
> the
> > > gt2-arcgrid and gt2-image, the WCS probably will not work
because
> > > Martin has not updated the gt2 trunk with the Simone's patches
> yet.
> > > However I put the patches in a branch subdirectory.
> > >
> > Hey, first off use the geoserver devel list for things like this.
> Other
> > people on the list may be checking out your work - Indeed I am
> probably
> > not going to check up on you much until a review for merge. This
> is a
> > development list, so don't feel shy about posting . if people
want
> less
> > traffic they can go to the users list.
> >
> > Second, Simone does not yet have commit access? We should get
> that for
> > him, as he's doing a bunch of good work, and shouldn't have to
wait
> on
> > Martin. Has anyone else reviewed his code? I'll send an email
to
> the
> > list.
> >
> > Chris
> >
> > > I think the gt2 will be updated soon.
> > >
> > > Cheers,
> > > Alessio.
> > > > On Tue, 22 Mar 2005 13:01:12 +0000, cholmes@anonymised.com
> > > <cholmes@anonymised.com> wrote:
> > > > No worries. I just deleted it. And I announced the branch
on
> the list.
> > > > And do encourage/help out the Norman Baker guy - I think he
did
> some
> > > > good work on the deegree wcs - if you show enough potential
> with your
> > > > implementation he'll probably be able to help out.
> > > >
> > > > best regards,
> > > >
> > > > Chris
> > > >
> > > >
> > > > Quoting Alessio Fabiani <alessio.fabiani@anonymised.com>:
> > > >
> > > > > Paul, Chris ... I made a mistake, I have updated the trunk.
> Don't
> > > > > worry, I was aligned with the last version, all working ...
> but there
> > > > > is the wcs inside. So please, modify the trunk with the old
> version
> > > > > and leave the wcs version only to the branch.
> > > > >
> > > > > branches/wcs-branch created.
> > > > >
> > > > > Thanks and sorry,
> > > > > Alessio.
> > > > >
> > > > >
> >
> >
>

----------------------------------------------------------
This mail sent through IMP: https://webmail.limegroup.com/

----------------------------------------------------------
This mail sent through IMP: https://webmail.limegroup.com/