- From: <Svgdeveloper@aol.com>
- Date: Fri, 2 May 2003 02:16:41 EDT
- To: public-qt-comments@w3.org
I and a friend have been independently exploring the date-time functions in XPath 2.0 / XQuery 1.0. They are very tedious to hand code. Looking at this from a coder's point of view, it would be much nicer to have something like, extract(someDesiredDateComponent, someDateTimeItem) rather than a mulitiplicity of get-somethingInteresting-from-gHorribleKludge() functions. So, would it be possible to replace the panoply of component extraction functions with, for example, extract(month, myItemName) the first argument being all legal components (seconds, minutes, timezone etc) and the second argument being simply an item name (not even requiring a coder to master all the gHorribleKludge types), with the item having to be a date-time type? So I guess I am asking several questions: 1. Is there a good reason for having the multiplicity of dateTime component extraction functions as separate functions? 2. Is there a compelling technical reason why extract(month, myDateTimeItem) couldn't work or is perceived as technically inferior? 3. Depending on the answers to 1. and 2. is the WG willing to consider a more compact syntax requiring fewer date-time component extraction functions? I appreciate that we are now 5.5 months into the nominal "3 month" WD cycle, so I guess a new batch of WDs will arrive soon, so even if found worth exploring I accept that any acceptance of the idea won't be immediately expressed in a WD. I would be interested in the WG's comments. Andrew Watt
Received on Friday, 2 May 2003 02:16:53 UTC