- From: Chris Lilley <chris@w3.org>
- Date: Wed, 8 Dec 2004 20:19:20 +0100
- To: Karl Dubost <karl@w3.org>
- Cc: public-xml-id@w3.org
On Wednesday, December 8, 2004, 5:46:35 PM, Karl wrote: KD> Hi, KD> http://www.w3.org/TR/2004/WD-xml-id-20041109/#processing KD> [[[ KD> 1. The attribute's value is normalized according to the rules for KD> attribute-value normalization on attributes of type ID. For xml:id KD> processors operating on an infoset or some other output from an XML KD> parser, the value will already be normalized, but unless the parser KD> normalized it as a value of type ID it will still be necessary for the KD> processor to trim leading and trailing space (#x20) characters and KD> replace sequences of space characters by a single space. KD> ]]] KD> I encourage Daniel Veillard to translate this sentence in French to see KD> the difficulties of understanding it. Doesn't French have dependent clauses? :) But yes, it needs to be split up a little to improve clarity - there is a parenthetical reminder of ID normalization rules, getting in the way of the statement about possibly needing to re-normalize - and this detracts, unfortunately, from the clarity which might otherwise be attained. (ducks at using convoluted sentences in email) How about this (reworded, added example) 1. The attribute's value is normalized according to the rules for attribute-value normalization on attributes of type ID (trim leading and trailing space (#x20) characters and replace sequences of space characters by a single space). Note: For xml:id processors operating on an infoset or some other output from an XML parser, the value will already be normalized, but unless it is already recognized as type ID, it will need to be normalized again prior to ID assignment. 2. as per spec Example: <!DOCTYPE toto [ <!ATTLIST foo xml:id ID #IMPLIED> ]> <toto xml:space="preserve" xml:id=" one "> <foo xml:id=" two "/> </toto> In this example, the XML parser already knows that xml:id attribute on the foo element is of type ID and thus, in the infoset, the value is normalized to "one". However, it has no information about the xml:id attribute on the element toto and will preserve leading and trailing spaces in the infoset. The xml:id processor will thus need to renormalize the value to "two" before performing ID assignment. KD> Please make clear, short and simple statements. The sentence is too KD> long and we loose the meaning of it in the middle. Is the suggested rewording plus example better? -- Chris Lilley mailto:chris@w3.org Chair, W3C SVG Working Group Member, W3C Technical Architecture Group
Received on Wednesday, 8 December 2004 19:19:20 UTC