Re: Ordered union of sequences

Wolfgang,

Thanks for the warning/reminder! It is not planned to create a full schema
here, the complexType created is just a sequence of optional elements. At the
moment it isn't a sequence at all, they use <xs:all>, but want the sequence if
they have the method to find/correct the inconsistencies. The source data is
spread all over a 1000+ pages Reference Manual authored and changed by
humans.

- Michael

Am 08.04.2010 um 17:45 schrieb Wolfgang Laun:

> On Thu, Apr 8, 2010 at 3:28 PM, Michael M|ller-Hillebrand
> <mmh@xxxxxxxxxxxxx> wrote:
>>
>> BTW, the idea behind this is to create part of an XML Schema from
evaluating document instances.
>
> You do realize that this isn't possible. Any number of XML documents
> you analyze, you can't be sure that the XML Schema contains
> <xs:sequence>, <xs:all> or <xs:choice>, with minOccurs="0" or
> maxOccurs="x" in the "right" place.
>
> Have you considered investigating the approach of using "exclusion by
> contradiction"? Rather than trying to construct the correct ordering,
> look for contradictions to one of the aforementioned particle
> structures, and use whatever remains.