Re: Proposal for DataStore API

[johann sorel <jo...@geomatys.com>]

Ok, so we will have :

+ DataStore :
  - Metadata getMetadata()
  - Resource getRootResource()
  - Resource findResource(String name)

+ Resource :
  - Metadata getMetadata()
  - Envelope getEnvelope()

+ Aggregate extends Resource :
  - Collection<Resource> getResources();

+ DataSet extends Resource

+ FeatureSet extends DataSet :
  - Stream<Feature> read()
  - FeatureSet subset(Query)

+CoverageSet extends FeatureSet :
  - to be defined

We are just mimicking the metadata model, ISO-19115 is not made to 
manage datas.
it is missing of lot of aspects anyway, queries, filters, creation, 
deletion, transactions...

A WMS would likely be a WMSDataStore with a root resource of type 
Aggregate and children
resources of type CoverageSet for each layer of the getCapabilities.
The SV_Service would be only in the getMetadata().


Johann Sorel


On 31/07/2017 23:33, Martin Desruisseaux wrote:
> I would like to bring some notes for though, but I'm not sure yet what
> would be the implication for the Resource hierarchy.
>
> Le 31/07/2017 à 14:47, johann sorel a écrit :
>
>> A Resource can be of multiple types, so far we can imagine several
>> subtypes :
>> - FeatureResource (or FeatureCollection?) for features obviously
>> - CoverageResource for images and OGC map services, WMS,WMTS,WCS ...
>> - MetadataResource (or Catalog) for services such as OGC CSW
>> - SensorResource for datas like NMEA or OGC SOS
> This proposal defines a hierarchy based on the kind of objects (Feature,
> Coverage or Metadata) returned by the Resource. But ISO 19115 section
> 6.1 and 6.2 rather defines the Resource hierarchy based on whether the
> data were collected by the same sensor, or adhere to a common production
> specification, etc. For example one "DS_Resource" subtype is
> "SV_Service". In my understanding, the same "SV_Service" type would be
> used for WMS, WCS, WFS, CSW, etc. regardless if the service return
> Feature, Coverage or Metadata.
>
> ISO 19115 seems to consider Feature and Coverage as DataSets components
> instead than Resource sub-types. ISO 19115 does not said much about how
> the relationship is established, except "The method for relating
> metadata to feature and attribute instance is defined in ISO 19109". I
> will check tomorrow if I can find more information.
>
> We could add to GeoAPI the Resource hierarchy as defined by ISO 19115.
> But the way to relate those resources to Feature and Coverage may have
> to be an Apache SIS extension, since I didn't found a clear guidance in
> OGC/ISO specifications on this topic yet.
>
>      Martin
>
>