[ 
https://issues.apache.org/jira/browse/CLEREZZA-525?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
 ]

Reto Bachmann-Gmür closed CLEREZZA-525.
---------------------------------------

    Resolution: Fixed

I replied on the list for the platform related usecase, what you think is bug 
turns out to be a feature as it allows for example an uploaded turtle document 
to be used in a sparql query (while the turtle document is still available at 
its uri including orginal formatting ans comments).

but even if this wasn't case, the patch is a solution for the issue at hand 
which does not concern the platform layers but the SCB layer. SCB TcManager is 
an APi that returns graphs given a URI naming them. Per se (i.e. rdf.core) has 
nothing do neither with serving graphs nor with retrieving them, the modular 
architecture allows multiple system to be connected. The storage provider 
realized in this issue connects the Semantic Web to TcManager. The semantic can 
be descibed as a huge set of named graphs, thanks to the resolution of this 
issue, these graphs are now available with TcManager.

Because of the different provider having different weights and this provider 
having weight 0 this ensures that it is only used when no other provider 
provides a graph for a specified name.

But again, if naive usage of this proxy was to create problems on the platform 
layer, this would be no reason to deny this useful client library on this 
layer. Rather one should address this problems on the layer they show up, but 
as per now I don't see a situation in the platform where using the proxy 
creates fundamental problems.

> Create caching storageprovider retrieven triple-collections from the web
> ------------------------------------------------------------------------
>
>                 Key: CLEREZZA-525
>                 URL: https://issues.apache.org/jira/browse/CLEREZZA-525
>             Project: Clerezza
>          Issue Type: Improvement
>            Reporter: Reto Bachmann-Gmür
>            Assignee: Reto Bachmann-Gmür
>
> The current api has many undocumented public methods and it seems to make an 
> unsharp distinction between the caching (and respective access) of graphs and 
> of graphnodes.

--
This message is automatically generated by JIRA.
For more information on JIRA, see: http://www.atlassian.com/software/jira

Reply via email to