Subject: Re: [xsl] GByte Transforms From: Kevin Jones <kjones@xxxxxxxxxxx> Date: Wed, 2 Jun 2004 23:24:50 +0100 |
Hi Jeff, Comments below. On Wednesday 02 June 2004 21:40, Jeff Kenton wrote: > > > > Ignore the problem, leave to stylesheet writer testing. > > Can't do this. People do the strangest things in stylesheets, as any > reader of this list knows. Your job is to take anything a customer might > throw your way, no matter how weird, and "do the right thing". Ok. I was really just trying that one on. I think performance would be too unpredictable if we did that which is why we could do with some thoughts on this. > > > Extra smarts in the compiler to warn of the use of potentially non-linear > > behaviour. E.G. Recursive templates not being tail recursive, nested > > loop/ template constructions. > > > > As above but aided by structural information for better targeting. > > Sure, the more diagnostics for the user, the better. But be prepared for > users ignoring them, and customers that set things up so that users never > see the warnings. That is a rather an obvious drawback of just warning about potential problems, although I have always liked the high level of feedback Saxon gives about stylesheet issues. > > Subset XSLT to limit the scope for non-linear transforms. > > It often comes down to that. The other way to look at the problem of > "streaming" large input files is to analyze the stylesheet and try to > decide how much of the input you need to keep during processing. For some > operations, only the current node is necessary. More often, keeping just > the path from the root to the current node will work (as another poster > suggested). Sometimes, you need the entire input tree, and you're not > really "streaming" anymore. Consider it a continuum, rather than just a > binary "can I stream this stylesheet or not" question. Nice point about it not being a binary choice although I think streamability is perhaps only a sub-plot here. I suspect trivially streamable transforms may always have a linear performance characteristic by definition but the reverse does not hold, i.e. things that a streamable via re-writing or not streamable at all may also have linear performance. I will have a closer look at that relationship to see if it holds anything we can use. Thanks, Kev
Current Thread |
---|
|
<- Previous | Index | Next -> |
---|---|---|
Re: [xsl] GByte Transforms, Jeff Kenton | Thread | Large transforms (was Re: [xsl] GBy, Kevin Jones |
Re: [xsl] encoding shift_jis into a, M. David Peterson | Date | Re: [xsl] Performance Question: Exp, M. David Peterson |
Month |