W3C home > Mailing lists > Public > public-tt@w3.org > April 2009

RE: Spec question. xml:space

From: Sean Hayes <Sean.Hayes@microsoft.com>
Date: Fri, 17 Apr 2009 10:05:40 +0100
To: "Glenn A. Adams" <gadams@xfsi.com>, Public TTWG List <public-tt@w3.org>
Message-ID: <AB3FC8E280628440B366A29DABB6B6E806CE1E4A02@EA-EXMSG-C334.europe.corp.microsoft.com>
Fair enough. But it would be helpful if readers of the timed text spec did not have to go through that level of reasoning. I think the crux of your reply is:

since Block.class (which is used to define the content model for <div/>) does not include #PCDATA, then whitespace in a div context must never be considered element content, regardless of the value of xml:space;

Which might make the basis of a helpful note somewhere. Anyway, at least I know how to write a test for this now.

Sean Hayes
Media Accessibility Strategist
Accessibility Business Unit
Microsoft

Office:  +44 118 909 5867,
Mobile: +44 7875 091385

From: Glenn A. Adams [mailto:gadams@xfsi.com]
Sent: 17 April 2009 5:20 AM
To: Sean Hayes; Public TTWG List
Subject: Re: Spec question. xml:space

i think there is no ambiguity; you need to to distinguish between whitespace at the XML level and whitespace at the application level; xml:space is used to signal authorial intention at the application level;

in particular, referring to XML 1.0 5th Ed. Section 2.10,
An XML processor must always pass all characters in a document that are not markup through to the application. A validating XML processor must also inform the application which of these characters constitute white space appearing in element content.

so in case of DFXP, all whitespace is passed from XML processor to DFXP application processor; at that point, it is up to DFXP application processor to determine which of these whitespace appear in element content; in the current spec we normatively define, by means of the XML Representation - Element Information Item definitions, this usage by means of reference to #PCDATA via the definition of Inline. class Element Vocabulary Group (Table 4 of Section 5.3.1);

since Block.class (which is used to define the content model for <div/>) does not include #PCDATA, then whitespace in a div context must never be considered element content, regardless of the value of xml:space;

xml:space only says what the intention is for treating whitespace in #PCDATA, and not whitespace in non-#PCDATA contexts;

On 4/16/09 11:30 PM, "Sean Hayes" <Sean.Hayes@microsoft.com> wrote:
Right, but that brings us back to the original starting point: Should xml:preserve create a block level item to hold the preserved PCDATA in a div context or not. I don't believe the spec is clear on that point, it doesn't say that it should, but then it doesn't say it shouldn't either. My expectation is that it should not, but I'd like some words in the spec to indicate this explicitly.


Sean Hayes
Media Accessibility Strategist
Accessibility Business Unit
Microsoft

Office:  +44 118 909 5867,
Mobile: +44 7875 091385


From: Glenn A. Adams [mailto:gadams@xfsi.com]
Sent: 16 April 2009 4:17 PM
To: Sean Hayes; Public TTWG List
Subject: Re: Spec question. xml:space


well, the same information is also obtained from the XML Representation - Element Information Item of div and p, in sections 7.1.4 and 7.1.5, respectively, which, under the Content definition includes Block.class for the former and Inline.class for the latter, where these latter two are defined in Table 4 in Section 5.3.1, where you can see that only Inline.class includes #PCDATA, i.e., having a mixed content model

On 4/16/09 11:05 PM, "Sean Hayes" <Sean.Hayes@microsoft.com> wrote:
Perhaps a note in the Section 4 abstract document type, or under xml:space  would clarify this?


Sean Hayes
Media Accessibility Strategist
Accessibility Business Unit
Microsoft

Office:  +44 118 909 5867,
Mobile: +44 7875 091385


From: Glenn A. Adams [mailto:gadams@xfsi.com]
Sent: 16 April 2009 3:28 PM
To: Sean Hayes; Public TTWG List
Subject: Re: Spec question. xml:space

It's not straightforward to extract this information from the text of the spec. From the RNG schema's Content Module (C.1.2.3), the definition of TTAF.Inline.class includes the keyword "text", which is referenced by TTAF.p.content, but not by TTAF.div.content. TTAF.div.content includes TTAF.Block.class which does not include "text". A similar logic applies for the XSD schema.

On 4/16/09 10:20 PM, "Sean Hayes" <Sean.Hayes@microsoft.com> wrote:
Thanks, that's kind of what I thought, but I don't see what combination of spec references provides that answer.

From: Glenn A. Adams [mailto:gadams@xfsi.com]
Sent: 16 April 2009 2:53 PM
To: Sean Hayes; Public TTWG List
Subject: Re: Spec question. xml:space


On 4/16/09 8:57 PM, "Sean Hayes" <Sean.Hayes@microsoft.com> wrote:
I'm  trying to determine the proper handling of xml:space in the following example fragment:

My question is, what if anything should happen to preserved space between the div and first and last p and between the p elements. Inline text is not legal here, so should anonymous <p> elements be inserted, or is the space  ignored?
[GA] It (the whitespace) is ignored since the content type for <tt:div> is not MIXED.
Received on Friday, 17 April 2009 09:04:45 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 2 November 2009 22:41:42 GMT