Hi Alexey See the answers inline
On Sun, Dec 23, 2012 at 8:43 PM, Alexey Kudinov <[email protected]> wrote: > Being more specific: > 1. Is it a good idea to use the schema.rdfs.org for a custom managed site? If the schema fits your domain I would recommend to use it. Especially if you plan to use RDFa/Microdata to embed such data in Webpages. > 2. Since I'm going to serialize my entities into schema.rdfs.org schema, is > there any field mapping that should be done in order to index properly with > SolrYard? Such mappings are optional. However the following mappings might be of interest # rdfs:label is the default label field used by a lot of RDF tools (including Stanbol). So typically it is a good idea to have rdfs:label values (if you do not care to store labels twice) schema:name > rdfs:label #schema.org supports adding additional types. While those should be converted automatically to rdf:type when converting data to RDF there might still be some left. So having a explicit mapping might not be a bad idea schema:additionalType | d=xsd:anyURI > rdf:type # The following mappings will improve how the Stanbol Enhancer UI present you Entities. So they might be useful during testing schema:description > rdfs:comment schema:image > foaf:depiction The default mappings are mainly for SKOS and therefore will not be interesting for you. So I would recommend to delete them. > 3. What property is mapped to the " _ref" field defined in SOLR schema? All URI values of your Entities. Its like the full text field (_text) but for URIs > 4. I'd like to alter some SOLR configuration - can I simply change the > default or should I create a custom solrindex.zip? If I understand correctly, > the latter is simply the SOLR conf and data directories, am I right? (I'm > familiar with SOLR, so the question is more about the best practices of > working in conjunction with Stanbol) I would suggest you create your own configuration. The best is to extract the default one (available at [1]). Than extract it and change it as you need. After that zip it again and name it e.g. "mymanaged.solrindex.zip". The next step is to copy this file to the datafiles directory ({stanbol-working-dir}/stanbol/datafiles). Then when you create the SolrYard set the value of the "Solr Index/Core" (org.apache.stanbol.entityhub.yard.solr.solrUri) to "mymanaged". If you want to use your custom configuration for all SolrYard instances you can also override the default configuration by naming you custom config "default.solrindex.zip". Files in the datafile directory will always override files with the same name. If you do that you will need to enable the "Use default SolrCore configuration" when creating the SolrYard. [1] http://svn.apache.org/repos/asf/stanbol/trunk/entityhub/yard/solr/src/main/resources/solr/core/default.solrindex.zip > 5. Can I submit any format other than rdf+xml to EntityHub to create an > entity? Apache Stanbol uses Apache Clerezza for parsing RDF files. Therefore the list of supported formats is the same for all services. * application/rdf+xml * text/turtle * application/x-turtle * text/rdf+nt * text/rdf+n3 * application/rdf+json > 6. I would like to leverage the latest development of SOLR 4, like the cloud > mode etc. - what version is currently supported by Stanbol? Solr 3.6.1. I had not yet time to look into Solr 4, but it is on the TODO list after the next Stanbol release (should be out in the next 2 weeks or so). Merry Christmas Rupert > > Thanks, > Alexey > > > > -----Original Message----- > From: Rupert Westenthaler [mailto:[email protected]] > Sent: Sunday, December 23, 2012 12:11 PM > To: [email protected] > Subject: Re: managed site walkthrough > > Hi Alexey, > > On Sun, Dec 23, 2012 at 10:10 AM, Alexey Kudinov <[email protected]> > wrote: >> Hi, >> i'd like to create a managed site with custom vocabulary content. is >> there any detailed walkthrough for this? I've read the wiki, but still >> have some questions. > > See [1] for information on how to configure and use managed sites. A detailed > description of the RESTful services is available via the WebUI of the > ManagedSite. As soon as you have configured your ManagedSite simple open > "http://{stanbol-instance}/entityhub/site/{managed-site-id}";. > > If you have specific questions feel free to ask. > >> say,i want to use schema.org schema with my entities, and i don't have >> a ready list of entities (example with generic indexer doesn't apply). >> That is, I'm going to use the REST api to insert all the entities and >> use a Solr Yard. how can i accomplish this?what should i configure? > > Yes you will have to use the RESTful api provided by > "http://{stanbol-instance}/entityhub/site/{managed-site-id}/entity"; to > perform CRUD operations with the entities you want to manage. > > Regarding the Yard: > > If you plan to use your Entities with the Stanbol Enhancer, than I would > strongly recommend to use the SolrYard. If you want to have a SPARQL endpoint > for your Entities, than the ClerezzaYard would be the right Yard for you. If > you are unsure I would recommend to start with the SolrYard. > > Hope this helps > best > Rupert > > > [1] http://stanbol.apache.org/docs/trunk/components/entityhub/managedsite.html > > -- > | Rupert Westenthaler [email protected] > | Bodenlehenstraße 11 ++43-699-11108907 > | A-5500 Bischofshofen > -- | Rupert Westenthaler [email protected] | Bodenlehenstraße 11 ++43-699-11108907 | A-5500 Bischofshofen
