Massive commit ahead

[Martin Desruisseaux <ma...@geomatys.fr>]

Hello all

In maybe 2 or 3 days, I will hopefully be able to commit the metadata 
classes that implements the org.opengis.metadata interface. However 
because of cross-dependencies, it would be tedious to commit the class 
one-by-one. Is there any objection if I commit them in one big bunch? It 
would be about 150 classes in a single commit...

     Martin

Re: Massive commit ahead

[Martin Desruisseaux <ma...@geomatys.fr>]

Hello Adam

Le 10/03/13 05:24, Adam Estrada a écrit :
> I am excited to see these commits come rolling in. I did start adding some content to the wiki last year but we didn't necessarily have a clear roadmap in mind at that time. Honestly, I didn't know what you were going to be adding at the time. Just keep us informed of what is going on so that I can include the details in the board reports and whatnot.

Sure. In fact, I committed a little bit more than what is strictly 
necessary for metadata. The org.apache.sis.measure.Range class is a good 
example. Metadata just needs a container for minimum and maximum values; 
it doesn't need all those 'union', 'intersect' and similar operations. 
But Range is going to play a much more important role in the coverage 
module, which will need those operations. I think it was easier to 
commit the complete Range class since we were on it, than filtering the 
parts needed by metadata and expand later.

As for SIS 0.3, I think that the "success metric" informing us when SIS 
is ready for a release could be to parse a relatively big ISO 19139 
compliant XML document, edit some metadata values, then write back the 
XML document with minimal information lost. Information lost that we may 
tolerate for SIS 0.3 release may be:

  * Coordinate Reference Systems (will need the sis-referencing module)
  * Temporal information (also related to sis-referencing module)
  * Geometries (will need the sis-geometry module)
  * XML comments (some said that comments must be preserved - I'm not
    sure yet how to handle that)


For everything else, information lost should be either considered as a 
bug, or be added to the above list.

     Martin

Re: Massive commit ahead

[Adam Estrada <es...@gmail.com>]

Martin,

I am excited to see these commits come rolling in. I did start adding some content to the wiki last year but we didn't necessarily have a clear roadmap in mind at that time. Honestly, I didn't know what you were going to be adding at the time. Just keep us informed of what is going on so that I can include the details in the board reports and whatnot.

Thanks again!
Adam

On Mar 8, 2013, at 6:33 PM, Mattmann, Chris A (388J) wrote:

> +1 it's fine to do it in a single commit, since this is really all part of
> the same stuff.
> 
> Also, we might want to think about (for smaller ones) leveraging Review
> Board:
> 
> http://reviews.apache.org/
> 
> Super awesome system and I love the interface (very Github like).
> 
> I'll request project access from INFRA for SIS:
> 
> https://issues.apache.org/jira/browse/INFRA-5963
> 
> 
> If anyone here on SIS besides those named in the issue want RB access just
> comment on the issue.
> 
> Thanks guys!
> 
> Cheers,
> Chris
> 
> On 3/6/13 1:00 AM, "Martin Desruisseaux" <ma...@geomatys.fr>
> wrote:
> 
>> Hello all
>> 
>> In maybe 2 or 3 days, I will hopefully be able to commit the metadata
>> classes that implements the org.opengis.metadata interface. However
>> because of cross-dependencies, it would be tedious to commit the class
>> one-by-one. Is there any objection if I commit them in one big bunch? It
>> would be about 150 classes in a single commit...
>> 
>>    Martin
>> 
> 

Re: Massive commit ahead

["Mattmann, Chris A (388J)" <ch...@jpl.nasa.gov>]

+1 it's fine to do it in a single commit, since this is really all part of
the same stuff.

Also, we might want to think about (for smaller ones) leveraging Review
Board:

http://reviews.apache.org/

Super awesome system and I love the interface (very Github like).

I'll request project access from INFRA for SIS:

https://issues.apache.org/jira/browse/INFRA-5963


If anyone here on SIS besides those named in the issue want RB access just
comment on the issue.

Thanks guys!

Cheers,
Chris

On 3/6/13 1:00 AM, "Martin Desruisseaux" <ma...@geomatys.fr>
wrote:

>Hello all
>
>In maybe 2 or 3 days, I will hopefully be able to commit the metadata
>classes that implements the org.opengis.metadata interface. However
>because of cross-dependencies, it would be tedious to commit the class
>one-by-one. Is there any objection if I commit them in one big bunch? It
>would be about 150 classes in a single commit...
>
>     Martin
>