[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [Elist Home]
Subject: [ubl-ndrsc] Containership issue with current LCSC samples
Folks: I really hate to raise the whip over this particular part of the donkey again (my arm is tired from flogging) but I'm going to anyway... If I'm right in my understanding of how XSLT works, the lack of nesting in the sample document is a total nightmare. If I have a document divided neatly into a header and a body, then any XPath expression that must be evaluated by the XSLT processor will be able to look at two nodes, and immediately know not to bother any further with one of them (which, in the case of the body of a document, may be huge - hundreds or thousands of line items, for example). If I am writing an XPath to address some header information, the processor need only iterate over the relatively limited set of child elements in the header to find all matches. The way it is now, the processor has to iterate over every child each time - including every single line item in the order to find things that are strictly header information. This is going to make XSLT processors, never known for their blinding speed, even slower, right? My transforms work OK on the samples, but they're not very large. Some real-world orders, etc., are very large indeed. This same principle will be in operation throughout the document, although it is at the top level that the current schemas seem to be most diifficult. The only other place we will have trouble is in not having explicit containers around lists. We had agreed to re-examine containership only if the resulting LCSC documents had some actual problem. I believe they do, although this is strictly a problem with the model-to-XML mapping, and not with the business modelling at all. I know some of you (Ken? Eduardo?) know a lot more about this than I do. Am I wrong about this? Should we not be worried? Please tell me I'm wrong... I'm happy to put aside the whip and let the poor donkey be, especially since I'm generally impressed with what LCSC has accomplished here [looks a lot like xCBL... ;-)], but we did agree to postpone discussions until we had an actual problem, and I'm afraid I've found one. Cheers, Arofan
[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [Elist Home]
Powered by eList eXpress LLC