Re: PURL WebIDs (was: WebID Frustration)

On Wed, 2013-08-07 at 19:14 +0100, Norman Gray wrote:
> Greetings.
> 
> Replying to myself, here...
> 
> On 2013 Aug 7, at 13:47, Norman Gray wrote:
> 
> > (I don't think this would _quite_ work with purl.org right now, because of conneg intricacies, but it's near-as-dammit, and one can imagine a very similar service which did).
> 
> This doesn't _quite_ work (I thought I might as well try, rather than merely speculate), but I suspect it's not far off.
> 
> 1. I chose an alternative WebID <http://purl.org/nxg/webid>.
> 
> 2. I created a new certificate using Nicholas Humfrey's script at <https://gist.github.com/njh/2432427>, entering the above PURL, and a suitably distinct name "Norman Gray (purl webid)", I think.  I let it import the new certificate into my (OS X) keychain.
> 
> 3. I copied the script's generated RDF/XML to a Dropbox file,...
> 
> 4. ...and created a link to it
> 
> 5. I went to purl.org and created the above WebID as a 'See other URL (303)', giving the Dropbox link as the SeeAlso url.
> 
> So: at this point, I have a WebID URL at purl.org which 303-redirects to a Dropbox URL which 302-redirects to another URL which returns RDF/XML which appears to me to conform to the WebID spec (I'm quietly impressed that this comes back down the wire with the correct application/rdf+xml content-type -- well done, Dropbox).
> 
> If, however, I now go to <https://webid.turnguard.com/WebIDTestServer/debug> and choose this associated certificate, I get a blank page.  I suppose it's possible that webid.turnguard.com is presuming Turtle or RDFa, but ...ooh, Jürgen Jakobitsch is in this thread -- any ideas, Jürgen?
> 
> ----

hi, norman.

webIDRealm's dereference method accepts any format it has a parser for
(inclusive of rdfa, for which i modified openrdf's version to work on a
local catalog to keep w3c servers' peace)

thing is that in the end of all redirections the httpconnection
determines the content type of the response to be text/html for which
there's no parser of course.

here's the code of the dereference method :

http://sourceforge.net/p/webidrealm/code/17/tree/trunk/src/main/java/com/turnguard/webid/utils/WebIDUtils.java#l159

you can see the exception if you use
http://webid.turnguard.com/WebIDTestServer/utils/parser
paste your webID and hit Parse URI, it will use HttpUtils.dereference in
WebIDUtils.java (see above).

i'll still need to figure why text/html is the determined contentType..
i'll let you know.

wkr jürgen

> 
> Interestingly, none of steps 1, 2 and 3 involved the web (once I'd downloaded the script, that is), and specifically they did not involve a certificate-generating service); step 4 involved a service (Dropbox) I'm already familiar with; as did step 5, though admittedly that's a more exotic interest than Dropbox.
> 
> Steps 1, 2 and 3 did involve a gist, running a shell-script, and cut-and-pasting a block of text into a file, but making that nicer is just a matter of UI design and the right scripting language.
> 
> Hey -- this stuff is easy! (and nearly works)
> 
> All the best,
> 
> Norman
> 
> 

-- 
| Jürgen Jakobitsch, 
| Software Developer
| Semantic Web Company GmbH
| Mariahilfer Straße 70 / Neubaugasse 1, Top 8
| A - 1070 Wien, Austria
| Mob +43 676 62 12 710 | Fax +43.1.402 12 35 - 22

COMPANY INFORMATION
| web       : http://www.semantic-web.at/
| foaf      : http://company.semantic-web.at/person/juergen_jakobitsch
PERSONAL INFORMATION
| web       : http://www.turnguard.com
| foaf      : http://www.turnguard.com/turnguard
| g+        : https://plus.google.com/111233759991616358206/posts
| skype     : jakobitsch-punkt
| xmlns:tg  = "http://www.turnguard.com/turnguard#"

Received on Wednesday, 7 August 2013 19:47:48 UTC