Re: [xsl] problem with xsl:import and namespaces

["G. Ken Holman" <gkholman@xxxxxxxxxxxxxxxxxxxx>]

At 2010-08-17 14:19 +0200, Jostein Austvik Jacobsen wrote:
> I've got five stylesheets, where test.xsl imports A.xsl and B.xsl,
> A.xsl imports C1.xsl, and B.xsl imports C2.xsl. C1.xsl and C2.xsl is
> in the same namespace, and defining the same function (i.e. C2.xsl is
> a newer version of C1.xsl). Now, I wouldn't think that this would
> cause a problem, as C1.xsl and C2.xsl is not imported directly into
> the same stylesheet. test.xsl only makes use of C1.xsl and C2.xsl
> indirectly via A.xsl and B.xsl, not knowing that there are multiple
> versions of C (or that A.xsl and B.xsl uses C for that matter.)
>
> However, when importing B.xsl after A.xsl in test.xsl, then the
> function in C1.xsl is overridden by the one in C2.xsl, even when
> called from A.xsl...

Yes, by definition.  There is only *one* collection of top-level 
names when all of the importation and including is done.

> I guess there's something I've missed about xsl:import and namespaces

All top-level names can be qualified with a namespace URI.

At the time of executing an instruction, there is a single set of 
names of top-level constructs.  Which top-level construct can be seen 
is based on the instruction ... for example, <xsl:apply-templates/> 
will see one of every top-level construct name considering every tree 
in the import chain.  When using <xsl:apply-imports/> the instruction 
will see one of every top-level construct name considering every tree 
in only imported stylesheets and their imports.

You are using function calls and function calls are resolved across 
the entire importation tree, not just a subset.  So, the function 
call with the highest importance in the entire tree is going to be 
the one that is always called.

Latter imports in a stylesheet confer higher importance on imported 
stylesheet constructs than earlier imports.  In the spec this concept 
of importance is referred to officially as the "import precedence".

I hope this helps.

. . . . . . . . . .  Ken

--
XSLT/XQuery training:   after http://XMLPrague.cz 2011-03-28/04-01
Vote for your XML training:   http://www.CraneSoftwrights.com/s/i/
Crane Softwrights Ltd.          http://www.CraneSoftwrights.com/s/
G. Ken Holman                 mailto:gkholman@xxxxxxxxxxxxxxxxxxxx
Male Cancer Awareness Nov'07  http://www.CraneSoftwrights.com/s/bc
Legal business disclaimers:  http://www.CraneSoftwrights.com/legal