W3C home > Mailing lists > Public > public-xml-processing-model-wg@w3.org > October 2011

[closed] Re: Base URIs and Browsers

From: Norman Walsh <ndw@nwalsh.com>
Date: Thu, 13 Oct 2011 11:10:56 -0400
To: public-xml-processing-model-wg@w3.org
Message-ID: <m2ehyhq64f.fsf@nwalsh.com>
At the 13 Oct telcon, we agreed that this was an informative message, not a comment on the spec.

Alex Milowski <alex@milowski.org> writes:
> Just an FYI, but I've been testing the xml:base support in various
> browsers (Safari, Chrome, and Firefox) with some interesting results.
> In theory, according to the HTML5 specification, the "base URI"
> property of the of the element must be used to resolve URI values [1].
>  For example, that means an 'img' element with an xml:base attribute
> should load the image relative to the value of the base URI derived
> from the xml:base attribute.
> Firefox does this properly when an 'img' element either has an
> xml:base attribute in the original markup or the attribute was set via
> Javascript on the element before it is added to the DOM.  Firefox does
> not do this properly when the xml:base attribute is added or changed
> once the image is in the DOM.  The HTML5 specification says nothing
> about this kind of behavior and I suspect it an oversight that should
> be corrected.
> Neither of the WebKit based browsers (Safari or Chrome) that I tested
> do this correctly.  They derive the base URI properly but this base
> URI has no affect on the image location.  This is a bug according to
> the HTML5 specification.
> This all becomes important if you expected content retrieved via
> XInclude to work properly.  In theory, a chunk of XHTML included
> should be able to have relative references to ancillary resources.
> This will work during the initial include in Firefox but not in any
> WebKit-based browser.  I haven't tested IE yet but I will do so.
> What does this have to do with the XML Processing Model?  Well, I
> think we need to be clear about the expectations we are putting on
> browser vendors.  If xml:base works properly, XInclude will work quite
> well.  Then, regardless of whether a XInclude is invoked automatically
> by the browser or via an application in Javascript, the correct thing
> will happen with intrinsic vocabularies like XHTML or SVG.
> [1] http://dev.w3.org/html5/spec/Overview.html#resolving-urls
> [2] http://dev.w3.org/html5/spec/Overview.html#the-img-element

                                        Be seeing you,

Norman Walsh
Lead Engineer
MarkLogic Corporation
Phone: +1 413 624 6676

Received on Thursday, 13 October 2011 15:11:41 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 21:32:49 UTC