OASIS Mailing List ArchivesView the OASIS mailing list archive below
or browse/search using MarkMail.


Help: OASIS Mailing Lists Help | MarkMail Help

office message

[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]

Subject: [OASIS Issue Tracker] Commented: (OFFICE-3317) ODF 1.2 Part 1 5.5.2<text:changed-region> ID attribute inconsisently available

    [ http://tools.oasis-open.org/issues/browse/OFFICE-3317?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=21776#action_21776 ] 

Dennis Hamilton commented on OFFICE-3317:

My proposal would be to deprecate (not forbid) the non-xml:id ID attribute.

Presumably the reason it is still allowed at all (in ODF 1.2) is so that ODF 1.2 consumers can successfully consume ODF 1.0/1.1 documents as if they are ODF 1.2 documents in this respect.  (This should be explained in a note.)

If that is the use case, the standalone non-xml:id case needs to be handled.  

If the intention is to produce down-level understandable ODF 1.2 documents, then the recommendation is to provide both an xml:id and the non-xml:id NCName attribute with exactly the same attribute value.  (This should be explained in a note.)

My proposal implements the "treat as if an xml:id" case in the schema itself.

By deprecating the non-xml:id, the xml:id case is encouraged in ODF 1.2 producers, yes?

The difference is because of differences in conforming consumer and producer behavior.  We can even say SHOULD/SHALL NOT on the standalone non-xml:id for ODF 1.2 producers, but we can't rule it out for consumers if the uses cases are as I understand them.  We could also say that if an ID attribute is to be produced the element, it SHALL be with an xml:id.  We just need to get the language right and have the schema and the language aligned.

> ODF 1.2 Part 1 5.5.2 <text:changed-region> ID attribute inconsisently available
> -------------------------------------------------------------------------------
>                 Key: OFFICE-3317
>                 URL: http://tools.oasis-open.org/issues/browse/OFFICE-3317
>             Project: OASIS Open Document Format for Office Applications (OpenDocument) TC
>          Issue Type: Bug
>          Components: Needs Discussion, Part 1 (Schema), Public Review, Text
>    Affects Versions: ODF 1.2 CD 05
>         Environment: This defect applies since at least ODF 1.2 Part 1 CD04.  This issue is presented in terms of ODF 1.2 CD05 Part 1.
>            Reporter: Dennis Hamilton
>            Assignee: Michael Brauer
>            Priority: Blocker
>             Fix For: ODF 1.2 CD 06
>  1. Every <table:changed-region> attribute requires an attribute of type ID in order to be referenced by the IDREF text:change-ide attribute in one or more change marks elements.
> However, with the current RNG Schema, there is no <table:changed-region> attribute with a value of type ID when there is no xml:id attribute.  That is because the text:id attribute, which must be present when there is no xml:id attribute, has a value of type NCName, not type ID.
>  2. The definition for text:id and xml:id usage in this case does not correspond to what is required to satisfy the requirements for use of ID in accordance with the [XML-ID].  Those conditions surely have precedence over any supplementary conditions in the text of ODF 1.2.  In particular, the text of section 19.811.5 specifies that there must always be an xml:id when there is a text:id, but that contradicts the schema, which makes explidit allowance for a standalone text:id attribute on a <text:changed-region> attribute.
>  3. It is not clear how backward compatibility is served by the peculiar requirement to have xml:id and text:id have the same value when both are present.  This presumes that a down-level consumer will ignore the xml:id and not consider there to be a violation of [XML-ID] by the apparent appearance of to ID-valued attributes with the same value (according to the down-level schema plus [XML-ID]).
>  [NOTE: There are other places where the same device appears to be attempted with regard to introduction of an xml:id attribute that pre-empts the type of another attribute formerly (and in isolation) having a value of type ID.  Similar measures may be appropriate for those cases.  This issue does not address those potential cases.]

This message is automatically generated by JIRA.
If you think it was sent incorrectly contact one of the administrators: http://tools.oasis-open.org/issues/secure/Administrators.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira


[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]