Re: [Moderator Action] Possible Errata

Moderator: sent to www-forms-editor@w3.org <www-forms-editor@w3.org>

----- Original Message -----
From: "Grant Husbands" <w3c-xforms@grant.x43.net>
To: <www-forms-editor@w3.org>
Sent: Wednesday, September 04, 2002 6:10 PM
Subject: [Moderator Action] Possible Errata


>
>
> Hi,
>
> As regards the latest XForms spec, I've compiled a list of things I
believe
> to be errors (I've used XPath expressions, or something similar, in
> explaining the elements/attributes I mean):
>
> In the schema, //submission/@method can have the values:
>
("post"|"put"|"get"|"form-data-post"|"urlencoded-post"|qname-but-not-ncname)
> In the table in 3.3, the list is given as:
> ("post"|"get"|"put"|qname-but-not-ncname)
>
> In the schema, //bind/@p3ptype exists.
> The table in 3.3 doesn't mention it.
>
> The table in 3.3 mentions //submission/@ref, but not any of the other
> Single-Node Binding Attributes. It should probably reference "3.2.3
> Single-Node Binding Attributes" for the 'submission' element.
>
> Similarly, it mentions //bind/@nodeset, but not any of the other Node-Set
> Binding Attributes. It should probably reference "3.2.4 Node-Set Binding
> Attributes" for the 'bind' element.
>
> //Case/@id is mandatory in the schema, but 9.2 and 9.2.2 don't mention
that.
>
> The schema has //switch using Single-Node Binding Attributes and Common UI
> Attributes, but 9.2 and 9.2.1 don't mention these.
>
> The schema has //group using Common UI Attributes, but 9.1 and 9.1.1 don't
> mention them.
>
> In 9.3, the last row of the table seems to be unparsable. The first four
> items appear in square brackets, with the fourth one being later repeated
> outside of the square brackets, and a fifth attribute appearing after
that.
> "Node Set Binding attributes" appears in brackets and not as a link.
>
> In 10.1, for //message/@level, it says the possible values are
("ephemeral"
> | "modeless" | "modal"); the schema does not specify this restriction.
>
> 10.1.12 and 10.1 claim that //message has "Linking Attributes", but the
> schema does not reflect that.
>
> The table in 9.3 doesn't mention "Events" for insert, delete and setindex,
> even though the schema and sections 9.3.5, 9.3.6 and 9.3.7 all do.
>
> 10.1.10 states that //send/@submission is required, but the schema says
it's
> optional.
>
> 10.1.8 says either the single node binding attributes or the linking
> attributes are required, but the schema says src is always required.
>
> 10.1.8 only gives descriptions for the 'new' and 'replace' values for
> //load/@show, but the table at 10.1 gives 'embed', 'other' and 'none'.
These
> should probably be explained.
>
> The values given for //load/@show (in the table at 10.1) aren't reflected
in
> the schema (which only restricts it to being a string).
>
> 10.1.3, 10.1.4, 10.1.5 and 10.1.6 all fail to mention the 'model'
attribute
> directly, instead they say "a specific XForms Model", without saying how
to
> specify one.
>
> In 10.1.2, //dispatch/@bubbles and //dispatch/@cancelable are described as
> being optional, but their default (which the schema says is 'true') isn't
> specified.
>
> http://www.w3.org/TR/xforms/slice10.html#action-xmlevents specifies
> (action|dispatch|...|setindex)* as being of content set "Action". The
schema
> doesn't include 'action' as part of that set, and additionally includes
> 'message' as part of that set. (This is probably a side-effect of the fact
> that the spec uses the same term "Action" for the xsd:group "actionGroup"
> and the xsd:group "actions")
>
> http://www.w3.org/TR/xforms/slice10.html#action-xmlevents is missing a
> vertical bar between the two lines in
> "(action|dispatch|rebuild|refresh|recalculate|revalidate|setfocus
> load|setvalue|send|reset|toggle|insert|delete|setindex)*"
>
> http://www.w3.org/TR/xforms/slice8.html#attrs-ui-common doesn't mention
the
> 'extension' element as part of "UI Common", even though the schema does.
>
> In Appendix E.5, the bottom example has "Comments user predict"; it should
> be "Comments user predictOn" if the table further up the page is to be
> believed.
>
> Grant.

Received on Wednesday, 4 September 2002 12:11:48 UTC