Firstly, it's not at all clear that one of these rules is "more specific". If
you draw the Venn diagram, the two patterns define two sets of elements,
neither of which is a subset of the other.

Secondly, XSLT does not say that the more specific rule wins. It says (see
https://www.w3.org/TR/xslt-30/#default-priority)

<quote>
In many cases this means that highly selective patterns have higher priority
than less selective patterns...

However, it is not invariably true that a more selective pattern has higher
priority than a less selective pattern. For example, the priority of the
pattern node()[self::*] is higher than that of the pattern salary. Similarly,
the patterns attribute(*, xs:decimal) and attribute(*, xs:short) have the same
priority, despite the fact that the latter pattern matches a subset of the
nodes matched by the former.
</quote>

Michael Kay
Saxonica

> On 23 Aug 2022, at 17:42, David Carlisle d.p.carlisle@xxxxxxxxx
<xsl-list-service@xxxxxxxxxxxxxxxxxxxxxx> wrote:
>
>
>
> On Tue, 23 Aug 2022 at 17:35, Roger L Costello costello@xxxxxxxxx
<xsl-list-service@xxxxxxxxxxxxxxxxxxxxxx> wrote:
> Hi Folks,
>
> I have these two template rules:
>
> <xsl:template match="procedureLeg[ancestor::airport]">
> <xsl:template match="*[child::*]">
>
> When the <procedureLeg> element is encountered in the instance document,
Saxon throws a warning, saying that it is ambiguous which rule to use.
>
> But, but, but, ...
>
> Clearly the first rule is more specific.
>
> "specific" isn't a defined term here though. They both have the same default
priority.
>
> Default priority is based on some simple rules on the structure (and both of
these have [..])
> it explicitly does not do logical analysis to determine "anything that
matches the first pattern must match the second", which is what  I guess you
mean by specific here.
>
> David
>
>
>
> I thought the XSLT specification says: the more specific rule wins. Right?
So why is Saxon throwing a warning message?
>
> I know that I can add "priority" to the first rule and then the warning will
go away, but I shouldn't have to do that, right?
>
> /Roger
>
>
> XSL-List info and archive
> EasyUnsubscribe (by email)