- From: Martin J. Dürst <duerst@it.aoyama.ac.jp>
- Date: Wed, 24 Jun 2009 15:43:12 +0900
- To: Erik Wilde <dret@berkeley.edu>
- CC: uri@w3.org, Anne van Kesteren <annevk@opera.com>, Dan Brickley <danbri@danbri.org>, Michael Schneider <schneid@fzi.de>
On 2009/06/24 0:40, Erik Wilde wrote: > Anne van Kesteren wrote: >> On Tue, 23 Jun 2009 17:23:24 +0200, Dan Brickley <danbri@danbri.org> >> wrote: >>> Fwd'ing to uri@w3.org from the SW IG list. Can anyone here advise? It >>> would also be interesting to hear about the state of URI validation >>> libraries, even if they're not currently exposed as Web Services. >> Validator.nu does IRI reference validation for attributes that take >> them as value. > > http://uriparser.sourceforge.net/ and other URI parser implementations > are an option. submitting a URI would be easy, but the question is what > to return: just a (URI|IRI|invalid) response, or some sort of parsing > result that can be used to process the URI? the latter option might be > interesting beyond mere validation, but then again URI parsing is such a > low-level task that in most cases, people will probably prefer to have > it as a local library, instead of using a web service. What is more, the general grammar for URIs (or IRIs) is extremely loose, because it has to accommodate many different schemes. Validating by scheme can be more strict, but this means that data for each scheme has to be collected; this data may not be available easily and uniformly. Regards, Martin. -- #-# Martin J. Dürst, Professor, Aoyama Gakuin University #-# http://www.sw.it.aoyama.ac.jp mailto:duerst@it.aoyama.ac.jp
Received on Wednesday, 24 June 2009 06:44:09 UTC