W3C home > Mailing lists > Public > public-cwm-talk@w3.org > April to June 2012

Re: WebIDs and Content-Location

From: Henry Story <henry.story@bblfish.net>
Date: Fri, 13 Apr 2012 19:02:35 +0200
Cc: Linked Data community <public-lod@w3.org>, public-webid <public-webid@w3.org>, public-cwm-talk@w3.org, Owen Sacco <owen.sacco@deri.org>
Message-Id: <3FDB5428-2238-46F1-B509-15D42A13B2E6@bblfish.net>
To: Tim Berners-Lee <timbl@w3.org>

On 13 Apr 2012, at 17:35, Tim Berners-Lee wrote:

> 
> 1) The base address used for parsing an RDF document should be the request URI,
> not the Content-Location: value.   Otherwise randomly clients who can accept
> n3 and rdf/xml will get <hada.rdf#me> and <hada.n3#me> which is clearly a bad idea.
> (Imagining that there is a hada.n3 option).

My thinking was that if a GET on <hada> has a Content-Location of <hada.rdf> then
if one follows that procÚdure then the graphs for a GET on <hada> and the one returned
by GETing <hada.rdf> will be different:
one of them will have as foaf:primaryTopic <hada#me> and the other <hada.rdf#me> .
That seems odd.


> 
> Is there is a 302 Moved  redirect, then that new Location:  URI should be used
> as the URI for the document and the base URI for parsing it.  (But NOT for 301).
> 
> 2) You say that rapper is string outputting things as relative URIs as  but i would
> support that, as often the absolute URI of a bit of RDF system is actually 
> been mapped through proxying, or from looking at the files on a server in file:// space, 
> and life is much easier if things default to 
> 
> I also like it that, anyone can do
>         $ echo '<#a> <#p> 123 .' | cwm --quiet
> and get
>         @prefix : <#> .
>         :a     :p 123 .
> 
> without being cluttered with  lots of references to the working directory.
> 
> On 2012-04 -13, at 08:54, Henry Story wrote:
> 
>> I have an issue about canonicalisation (de-relativisation?) of URLs. cwm and rapper
>> don't return the same results, though cwm agrees with http://www.w3.org/RDF/Validator/
>> 
>> What is the full URL for the rdf:ID="me" in the XML returned below?
>> Is it 
>> 
>>  - <http://vmuss13.deri.ie/foafprofiles/hada#me> as cwm  ( cwm.py,v 1.198 2012-01-30) and the w3 validator state?
>>  or is it
>>  - <hada.rdf#me> as raptor 2.0.6 returns (bizarrely as a relative url though) and as I thought it should be.
>> 
>> 
>> $ curl -i http://vmuss13.deri.ie/foafprofiles/hada
>> HTTP/1.1 200 OK
>> Date: Fri, 13 Apr 2012 12:04:11 GMT
>> Server: Apache/2.2.9 (Debian) PHP/5.2.6-1+lenny12 with Suhosin-Patch
>> Content-Location: hada.rdf
>> Vary: negotiate
>> TCN: choice
>> Last-Modified: Fri, 13 Apr 2012 11:26:38 GMT
>> ETag: "8080-6af-4bd8dbeebb780;4bd8dbeebb780"
>> Accept-Ranges: bytes
>> Content-Length: 1711
>> Content-Type: application/rdf+xml
>> 
>> <?xml version="1.0" encoding="ISO-8859-1"?>
>> <rdf:RDF xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#"
>>       xmlns:rdfs="http://www.w3.org/2000/01/rdf-schema#"
>>       xmlns:foaf="http://xmlns.com/foaf/0.1/"
>>       xmlns:rsa="http://www.w3.org/ns/auth/rsa#"
>>       xmlns:cert="http://www.w3.org/ns/auth/cert#"
>>       xmlns:admin="http://webns.net/mvcb/">
>> <foaf:PersonalProfileDocument rdf:about="">
>>     <foaf:maker rdf:resource="#me"/>
>>     <foaf:primaryTopic rdf:resource="#me"/>
>> </foaf:PersonalProfileDocument>
>> <foaf:Person rdf:ID="me">
>>     <foaf:nick>HADAUser1</foaf:nick>
>>     <foaf:givenName>Jane</foaf:givenName>
>>     <foaf:familyName>Smith</foaf:familyName>
>>     <foaf:workplaceHomepage rdf:resource="http://hhs.gov"/>	
>>     <foaf:topic_interest rdf:resource="HEAR"/>
>>     <foaf:topic_interest rdf:resource="Accounting"/>
>> 
>>     <cert:key>
>>       <cert:RSAPublicKey rdf:ID="key1">
>> 	<rdfs:label>HADA Admin</rdfs:label>
>>         <cert:modulus rdf:datatype="http://www.w3.org/2001/XMLSchema#hexBinary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cert:modulus>
>>         <cert:exponent rdf:datatype="http://www.w3.org/2001/XMLSchema#integer">65537</cert:exponent>
>>       </cert:RSAPublicKey>
>>     </cert:key>
>> </foaf:Person>
>> </rdf:RDF>
>> 
>> In my code  on read-write-web I use the returned Content-Location to form the base URL
>> 
>>    148           val loc = headers("Content-Location").headOption match {
>>    149             case Some(loc) =>  new URL(u,loc)
>>    150             case None => new URL(u.getProtocol,u.getAuthority,u.getPort,u.getPath)
>>    151           }
>>    152           res>>{ in=> modelFromInputStream(in,loc,encoding) }
>>    153 
>> ( https://dvcs.w3.org/hg/read-write-web/file/c6520ef80d5c/src/main/scala/GraphCache.scala#l148 )
>> 
>> Where modelFromInputStream uses the Jena libraries like this:
>> 
>>     12   def modelFromInputStream(
>>     13       is: InputStream,
>>     14       base: URL,
>>     15       lang: Lang): Validation[Throwable, Model] =
>>     16     try {
>>     17       val m = ModelFactory.createDefaultModel()
>>     18       m.getReader(lang.jenaLang).read(m, is, base.toString)
>>     19       m.success
>>     20     } catch {
>>     21       case t =>  {
>>     22         logger.info("cought exception turning stream into model ",t)
>>     23         t.fail
>>     24       }
>>     25     }
>> ( https://dvcs.w3.org/hg/read-write-web/file/c6520ef80d5c/src/main/scala/util/package.scala )
>> 
>> 
>> Social Web Architect
>> http://bblfish.net/
>> 
> 

Social Web Architect
http://bblfish.net/
Received on Friday, 13 April 2012 17:03:24 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 20:01:06 UTC