Re: Help moving away from named templates

On 5/2/07, David Carlisle <davidc@xxxxxxxxx> wrote:
> you presumably mean select not match (which is a syntax error) and it's
> generally a bad idea (slower and not namespace-aware0 to text eleemnt
> names using name() why not simply
>
> <xsl:apply-templates select="Records/Record/firstName"/>
>
> > and in most cases this would work, but of course this would conflict
> > with any other customer info that also contained 'firstName'
>
> I can't guess what kind of conflict you see here, could you post a
> slightly larger XML that also has elements that you wouldn't want to
> select.

Let's say I want to select XSL which displays the emergency contacts
of a customer, rather than the customer's info. That XSL would look
quite similar to XSL

<Records>
   <Record>
   <custID>
   <firstName>Customer's Mommy</firstName>
   </Record>
</Records>

I happen to know this is the customer's emergency contact because this
XSL was generated from the "Customer_Emergency_Contacts" table. But to
XSL they're quite similar!

> In particular in what way do you avoid this conflict using named
> templates, you still need to decide on which nodes to call the named
> template which should be an exactly equivalent problem, shouldn't it?

A few ways which seem clunky/limited:

1) Generally, I just have the script call emergencyContacts.xsl, which
contains one template--the emergency contacts template. But doesn't
this circumvent the magical property of XSL to pick its own darn
templates according to the data? Also, there's something seemingly
naive to me about having a bunch of xsl files with one template.

2) If I need to display BOTH the consumer's info and their emergency
contacts on the same page, I include (INCLUDE HELL) the
emergencyContact XSL and then use <xsl:call-template name="eContacts"
/>