It would seem that you could transform Roger's original document
(miserable though it is) into yours with a previous xslt transformation.
That's if its pattern and ordering were to continue as the problem
goes to higher and higher N.
TomP
On 4/1/2017 6:49 PM, David Carlisle wrote:
> If you use an easier markup, for example
>
> <n-queens>
> <q c="1" r="3"/>
> <q c="2" r="1"/>
> <q c="3" r="4"/>
> <q c="4" r="2"/>
> </n-queens>
>
>
> which denotes your board with 4 queens in the specified row and
> column, then I think the following single xpath works for any size
> board and evaluates to true if every column has a queen, every row
> has a queen and the number of distinct falling and rising diagonals
> with a queen equals the number of queens.
>
> <xsl:stylesheet version="2.0" xmlns:xsl="http://www.w3.org/1999/XSL/Transform">
>
> <xsl:template match="n-queens">
> <xsl:sequence select="
> (every $c in 1 to count(q) satisfies exists(q[@c=$c])) and
> (every $r in 1 to count(q) satisfies exists(q[@r=$r])) and
> count(q)=count(distinct-values(q/(number(@r) - number(@c)))) and
> count(q)=count(distinct-values(q/(number(@r) + number(@c))))
> "/>
> </xsl:template>
> </xsl:stylesheet>
>
>
> you could easily make that into a schematron assertion.
>
>
> On 1 April 2017 at 23:19, David Carlisle <> wrote:
>>
>>> <column1><row>3</row></column1>
>>
>> why the horrible asymmetry in the markup, with the column number in the
>> element name and the row number in element content?
>> That defeats most ways of shortening the schema by exploiting the symmetry
>> in the problem.
>>
>> David
