tag:blogger.com,1999:blog-15626356.post4647758275240873863..comments2012-10-17T10:09:12.903-07:00Comments on ecmanaut: XPath bookmarksJohan Sundströmhttp://www.blogger.com/profile/04076097346172610543noreply@blogger.comBlogger4125tag:blogger.com,1999:blog-15626356.post-60374664472762635642007-05-14T03:03:00.000-07:002007-05-14T03:03:00.000-07:00I'll happily err on the side of pragmatism; the XP...I'll happily err on the side of pragmatism; the <A HREF="http://www.w3.org/TR/xptr-framework/" REL="nofollow">XPointer framework</A> seems to have stuck its head in the clouds in not being defined neother for non-XML doctypes nor non-well-formed documents, and messy HTML being the norm on the web today, I don't see how XPointer would be the more useful approach at present. Quote:<BR/><BR/><I>"The [XPointer] framework is intended to be used as a basis for fragment identifiers for any resource whose Internet media type is one of <B>text/xml</B>, <B>application/xml</B>, <B>text/xml-external-parsed-entity</B>, or <B>application/xml-external-parsed-entity</B>."</I><BR/><BR/>The XPath bookmarks defined above quite purposefully neither comply with nor water out the <A HREF="http://www.w3.org/TR/xptr-xpointer/" REL="nofollow">XPointer standard</A>, as they would if they did.<BR/><BR/>Question back: could you give an XPointer example link that Firefox supports and manages to resolve, or any more detail of what you mean by "supports", in summary or reference? <A HREF="http://golem.ph.utexas.edu/category/2006/09/dimensional_analysis_and_coord.html#xpointer(id(%22comments%22))" REL="nofollow">this XPointer link</A> does not seem to be supported by my Firefox in any useful fashion, anyway. That page is arguably "just" well-formed <B>application/xhtml+xml</B> though, so it might not be a proper example of anything but how limiting the XPointer framework scope is (when really anally interpreted by the letter, anyway).Johan Sundströmhttps://www.blogger.com/profile/04076097346172610543noreply@blogger.comtag:blogger.com,1999:blog-15626356.post-33648565620417687172007-05-14T00:07:00.000-07:002007-05-14T00:07:00.000-07:00Please take a look at the W3C’s XPointer (supporte...Please take a look at the W3C’s XPointer (supported in Firefox), and don’t invent something that is custom :).Laurens Holsthttps://www.blogger.com/profile/12694446373092927787noreply@blogger.comtag:blogger.com,1999:blog-15626356.post-62435184799040557782007-05-04T08:05:00.000-07:002007-05-04T08:05:00.000-07:00It's a valid uri, and in theory, you could say it'...It's a valid uri, and in theory, you could say it's an invalid anchor, as (at least according to the XHTML 1.0 Strict DTD) /, for instance, is not a valid part of an <a name=""> anchor name attribute. Ergo: yes.<BR/><BR/>More interestingly though, it's very, very unlikely people would use such anchors themselves, except perhaps as inspired by this post. And then, more likely for programmatic purposes IMO.Johan Sundströmhttps://www.blogger.com/profile/04076097346172610543noreply@blogger.comtag:blogger.com,1999:blog-15626356.post-10615886982333107682007-05-04T06:58:00.000-07:002007-05-04T06:58:00.000-07:00Hi, I really like the idea. I've often thought tha...Hi, I really like the idea.<BR/> I've often thought that it'd be nice if there was some standard way of mixing xpath into a uri to make it point to a specific element in a document - it would be especially useful in Piggbank/Solvent type situations. Is it a valid uri, and is it distinguishable from a url that might happen to have that fragment identifier?Anonymousnoreply@blogger.com