http://englich.wordpress.com/2006/10/31/implementing-xpaths-builtin-functions/ A well-defined blog. Thu, 24 Sep 2009 10:58:13 +0000 http://wordpress.com/ hourly 1 http://englich.wordpress.com/2006/10/31/implementing-xpaths-builtin-functions/#comment-5272 Idetrorce Sat, 15 Dec 2007 13:12:49 +0000 http://englich.wordpress.com/2006/10/31/implementing-xpaths-builtin-functions/#comment-5272 very interesting, but I don't agree with you Idetrorce very interesting, but I don’t agree with you
Idetrorce

]]> http://englich.wordpress.com/2006/10/31/implementing-xpaths-builtin-functions/#comment-474 dedicated hosting Thu, 09 Nov 2006 14:36:43 +0000 http://englich.wordpress.com/2006/10/31/implementing-xpaths-builtin-functions/#comment-474 Nice post indeed, its quite informative. Nice post indeed, its quite informative.

]]> http://englich.wordpress.com/2006/10/31/implementing-xpaths-builtin-functions/#comment-422 Michael Kay Wed, 01 Nov 2006 11:32:29 +0000 http://englich.wordpress.com/2006/10/31/implementing-xpaths-builtin-functions/#comment-422 Yes, it's an interesting idea - not one that I've pursued. In XPath 1.0 Saxon started off with custom Java implementations of all the 25 or so functions - which was probably necessary as there was very little redundancy present. I've stuck to that approach in 2.0 even though there's much more redundancy now. It would certainly be nice to find better ways of doing it; but it's difficult in practice because there are so many opportunities for special-casing, e.g. in the static type inferencing. I just fall back to using inheritance and delegation and table-driven code as much as possible to get reuse of common logic where possible. Yes, it’s an interesting idea – not one that I’ve pursued. In XPath 1.0 Saxon started off with custom Java implementations of all the 25 or so functions – which was probably necessary as there was very little redundancy present. I’ve stuck to that approach in 2.0 even though there’s much more redundancy now. It would certainly be nice to find better ways of doing it; but it’s difficult in practice because there are so many opportunities for special-casing, e.g. in the static type inferencing. I just fall back to using inheritance and delegation and table-driven code as much as possible to get reuse of common logic where possible.

]]> http://englich.wordpress.com/2006/10/31/implementing-xpaths-builtin-functions/#comment-421 Aron Stansvik Wed, 01 Nov 2006 09:43:59 +0000 http://englich.wordpress.com/2006/10/31/implementing-xpaths-builtin-functions/#comment-421 As always :) As always

]]> http://englich.wordpress.com/2006/10/31/implementing-xpaths-builtin-functions/#comment-420 Vincent Wed, 01 Nov 2006 07:51:47 +0000 http://englich.wordpress.com/2006/10/31/implementing-xpaths-builtin-functions/#comment-420 Very interesting entry... Very interesting entry…

]]>