| CARVIEW |
Select Language
HTTP/2 200
date: Tue, 07 Oct 2025 23:47:06 GMT
content-type: text/html
content-encoding: gzip
last-modified: Thu, 13 Jul 2023 18:10:33 GMT
cache-control: max-age=2592000, public
expires: Thu, 06 Nov 2025 23:47:05 GMT
vary: Accept-Encoding
access-control-allow-origin: *
x-request-id: 98b148159d88c537
strict-transport-security: max-age=15552015; preload
x-frame-options: deny
x-xss-protection: 1; mode=block
cf-cache-status: EXPIRED
set-cookie: __cf_bm=3tYSl6XNe6EcmUcWjP50d0ryby6IMDf9fFZyBHmNsTo-1759880826-1.0.1.1-z0urNp3_QyeZb7XSc01QDcTH0PE7zEKvgGHb5UdKI.VTeF1F3DYG1mp9pMfyHnMK9JgQml6K1uP1zXqLTRsUaU_SpyQJJ2pG.zG5sasnyDw; path=/; expires=Wed, 08-Oct-25 00:17:06 GMT; domain=.w3.org; HttpOnly; Secure; SameSite=None
server: cloudflare
cf-ray: 98b148159d88c537-BLR
alt-svc: h3=":443"; ma=86400
Grouping experience -> tiny change needed from Phil Archer on 2008-07-18 (public-powderwg@w3.org from July 2008)
Grouping experience -> tiny change needed
- From: Phil Archer <parcher@icra.org>
- Date: Fri, 18 Jul 2008 11:06:22 +0100
- To: Public POWDER <public-powderwg@w3.org>
- Message-ID: <48806B1E.5020501@icra.org>
I've had an e-mail exchange over the last 24 hours that has thrown up a slight problem with the grouping doc - but it's easily fixed. Thinking only in terms of POWDER-S and the matchesregex property, we should be able to match against https://purl.org/dc/.*/creator which would give a match for DC creator whether the old or new namespace were used. In general, we should be able to match against https://example.org/foo# so that we can spot properties in a given vocabulary. Sounds easy enough and using the tool at [1] the DC creator one works just fine [2, 3] but the example.org/foo# doesn't - or rather, didn't until I fixed it. In dropping support for IRI set definition in terms of user info and fragment identifiers we took it out of the processing model too. In the grouping doc we quote what we lovingly call Rabin's reg ex [4] but actually it isn't - we stripped out the parts that capture user info and fragment. Consequently, when my test tool takes a URI and splits it up using the given reg ex it /cannot/ match an IRI that has those components. The fix is simple enough - we just need to quote and use the full Rabin's Reg Ex but say that we don't provide POWDER elements to directly address those components. Not a big task but a good catch before we go to LC. With the fix in place, the example works [5] I'm editing that doc today and should have it to the group a little later. Phil. [1] https://www.fosi.org/powder-group-tester/ [2] https://www.icra.org/cgi-bin/powder-group.cgi?action=iriset&includeregex=%5Ehttp%3A%2F%2Fpurl.org%2Fdc%2F.%2A%2Fcreator&iri=http%3A%2F%2Fpurl.org%2Fdc%2Felements%2F0.1%2Fcreator [3] https://www.icra.org/cgi-bin/powder-group.cgi?action=iriset&includeregex=%5Ehttp%3A%2F%2Fpurl.org%2Fdc%2F.%2A%2Fcreator&iri=http%3A%2F%2Fpurl.org%2Fdc%2Fterms%2Fcreator [4] https://www.w3.org/TR/2008/WD-powder-grouping-20080630/#rabinsRegEx [5] https://www.icra.org/cgi-bin/powder-group.cgi?action=iriset&includeregex=%5Ehttp%3A%2F%2Fexample.org%2Ffoo%23&iri=http%3A%2F%2Fexample.org%2Ffoo%23bar
Received on Friday, 18 July 2008 10:07:28 UTC