- From: Norman Walsh <ndw@nwalsh.com>
- Date: Sat, 05 May 2007 08:31:59 -0400
- To: public-xml-processing-model-wg@w3.org
- Message-ID: <87abwjwiww.fsf@nwalsh.com>
/ Innovimax SARL <innovimax@gmail.com> was heard to say: | [[ | 5.14 p:document Element | The parser which the p:document element employs must be conformant | Namespaces in XML. It must not perform validation. It must not perform | any other processing, such as expanding XIncludes. | ]] | | [[ | A.1.9 Load | Load attempts to read an XML document from the specified URI. If the | document does not exist, or is not well-formed, the step fails. | Otherwise, the document read is produced on the result port. I think we need 'validate' and 'namespace-support' options on load. The former should default to 'no', the latter to 'yes. Implementations should be free to throw a dynamic error if namespace-support='no' is requested and they do not support documents that aren't namespace well-formed. Alternatively, we could outlaw documents that aren't namespace well-formed altogether. I could live with that. | I see two points : | 1) DTD Validation | 2) Non namespace well-formed | | 1) could easily be an option "dtd-validation" (true/false). But we may | need to have access to a catalog. For the moment, I'm happy to leave catalog support as a quality of implementation issue. But we'll come back to the p:catalog proposal soon. | 2) is already more tricky : | --> Is there really case where a well-formed file could be converted | to a namespace-aware-well-formed file ? | The only thing I can imagine is proposing an option to replace ":" by | another char (may be with colon-as-first-char-replacement and | colon-replacement) | But is this really worth ? I don't think it makes any sense to attempt to transform a namespace not-well-formed document automatically into something else. The question is really, do we think it's worth allowing them at all. Most modern technologies require namespace well-formed documents. Be seeing you, norm -- Norman Walsh <ndw@nwalsh.com> | Truth lies within a little uncertain http://nwalsh.com/ | compass, but error is immense.
Received on Saturday, 5 May 2007 12:32:05 UTC