On 09/01/10 22:07 +0100, Hartmut Goebel wrote:
> Hi Cedrik,
> 
> I'm currently working on https://bugs.tryton.org/roundup/issue70 (xml:
> based validation on Namespaces). Well, I took the change to refactor the
> schema and remove redundancies (eg. multiple definitions for label,
> newline, etc.).
> 
> Now I found some points which I'm unsure whether they are correctly
> described in the schema:
> 
> * For newline the attribute id is optional. Most <newline> tags in the
>   .xmls do not have an id attribute, so this matches the implementation.
>   But for all other elements, id is required (or not allowed).

That is correct. I don't think it is required to have id on newline at least
we have not yet need it.

> 
> * For board, graph and tree, <data> is defined as
>        element data { data.attlist, xpath+ }  # at least one xpath
>   whereas for form, <data> is defined as
>        element data { data.attlist, xpath* }  # could be empty
>   Is this meant this way?

It could be empty for all.

> 
> * IMHO this <data> element could be removed at all, since it only ever
>   contains xpath.

Except that XML requires a unique tag at root.

> 
> * According tho the current relaxng schema, in this construct::
>      <form> ... <graph> <x> <field>
>   the <field> could be a field as defined in form.rnc. I wonder whether
>   this is correct or this sould only allow fileds as defined in
>   graph.rnc?

It is correct. All attributes from form are not used in graph.


-- 
Cédric Krier

B2CK SPRL
Rue de Rotterdam, 4
4000 Liège
Belgium
Tel: +32 472 54 46 59
Email: [email protected]
Jabber: [email protected]
Website: http://www.b2ck.com/
twitter: http://twitter.com/cedrickrier
identi.ca: http://identi.ca/cedrickrier

Attachment: pgpqlyZGn3smr.pgp
Description: PGP signature

Reply via email to