- From: Innovimax SARL <innovimax@gmail.com>
- Date: Sat, 3 May 2008 11:52:37 +0200
- To: "XProc WG" <public-xml-processing-model-wg@w3.org>
- Message-ID: <546c6c1c0805030252o28290179s14d8531f2149adf0@mail.gmail.com>
One of the use case, I find frequently is to have a rough idea of how much is the sequence long p:count could help, but to respect the spec, it must read all the sequence for that Most of the time, I can content myself with just simple information like : * is the sequence empty ? * has the sequence only one document ? * has the sequence more than one document ? * has the sequence more than x document ? For this I propose to have an "limit" option that would be defaulted to -1 (or 0) -1 (or 0) means give a strict count of how many there are 1 would mean give 0 as result if empty, 1 if there is a document (which means it can answer very quickly) 2 would mean give 0 as result if empty, 1 if there is a document and 2 if there is more than one (would need to buffer only one document to answer) and so on One way to do that with todays tool but I don't like is to use <p:split-sequence test="position() <= $limit"/> and then do a <p:count> on the result 1) but it means that p:split-sequence needs to be optimized to know that we don't use the alternate port and once we reach the $limit we can stop processing 2) more than that it means you have to buffer one more document than my proposal Mohamed -- Innovimax SARL Consulting, Training & XML Development 9, impasse des Orteaux 75020 Paris Tel : +33 9 52 475787 Fax : +33 1 4356 1746 http://www.innovimax.fr RCS Paris 488.018.631 SARL au capital de 10.000 €
Received on Saturday, 3 May 2008 09:53:13 UTC