But in that case how do you speficy 'do this always'? There are some things you just plain always want to apply, no matter what the style is, they simply aren't optional. So I would much rather have it that alternates are applied if and only if there is a match for them of some sort. Yes, that does perhaps imply that you always have to specify some alternative or other, but even that isn't NECESSARILY the case, as the default may product the 'basic' document, and options may simply filter out parts you didn't want.
There has indeed been MUCH debate on the subject, and personally I think the whole PI model is a bit broken, esp when applied to stylesheets. Really they should be either mandatory or optional, and then if they are optional all sheets that constitute one alternate branch of processing would have the same label. Any stylesheet could specify multiple labels. The way it is now is simply ambiguous and sub-optimal for all cases. Seems like good ole' design by commitee... On Friday 05 September 2003 01:09 pm, Chris Strom wrote: > On Fri, Sep 05, 2003 at 11:38:08AM -0400, Tod Harter wrote: > > Yeah, thats sensible behaviour. a.xsl is a NON ALTERNATE stylesheet, so > > it ALWAYS gets applied. If these stylesheets didn't get applied then > > imaging an XSP page combined with your setup, you'd loose the XSP > > processing step! > > I guess what I'd like to see happen is that the result of my XSL > transformation includes an XSP processing instruction. That or the > XPATH document() function calls the XSP processing step. > > The specification apparently isn't clear on how xml-stylesheet should be > applied, saying only that it should be the same as HTML 4.0's <link > rel="stylesheet">. If that's the case, then I'd expect the named > stylesheet's <xsl:template match="/"> to take precedence. > > This is the most authoritative discussion that I could find in regards > to this: > > http://www.biglist.com/lists/xsl-list/archives/200004/msg01078.html > > That URL says that an alternate stylesheet takes precedence over a > preferred stylesheet when it it requested by title. > > > If you want A or B then they BOTH have to be alternate and you have to > > choose one or the other. > > > > The same debate arose recently WRT the interaction of default and non > > default stylesheets specified in server config directives too. The upshot > > of that I believe was that basically the same theory applies, if a > > stylesheet has no criteria defining when its invoked, then its ALWAYS > > invoked. > > I don't think this is how it's currently working in AxKit (and I'm > screwed if it is), nor do I think that it is correct behavior. If I'm > navigating sections of a large document, I'd like to access the document > directly (without PATHINFO of query params) and be shown the beginning > of the document. I'd then use PATHINFO to navigate the rest of the > document. > > I'm exploiting this behavior in my current project - the default > stylesheet displays the first section of the document. I'll have to > double check to make sure that nothing else is going on there... > > If I'm forced to specify the first section of the document, then all > applications referencing that document are going to have to have > pre-knowledge of the first section to be displayed. I guess that's not > a terrible hardship, but I'd much prefer to keep that info close to > the XML document. > > Thanks for the response and I hope you're wrong :) > > Chris > > --------------------------------------------------------------------- > To unsubscribe, e-mail: [EMAIL PROTECTED] > For additional commands, e-mail: [EMAIL PROTECTED] -- Tod Harter Giant Electronic Brain http://www.giantelectronicbrain.com --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
