W3C home > Mailing lists > Public > www-dom@w3.org > July to September 2006

Selectors API

From: Joao Eiras <joao.eiras@gmail.com>
Date: Mon, 25 Sep 2006 04:23:09 +0100
To: "www-dom@w3c.org" <www-dom@w3c.org>
Message-ID: <op.tgezovcgxl3at9@dragast>

The Selectors API defines 2 methods for the DocumentSelector interface  
which Documents will then inherit.
My point is.. do  you really want to call those 2 methods 'match' and  
'matchAll' ? Is this API flexible ?
The names are very redundant.

Imagine a future scenario where we want to provide a API for using a 3rd  
and a 4th language for selecting DOM nodes?
At will these be called ? find, findAll ? searchNode?
The words evaluate (in xpath), match and matchAll say nothing about their  
implementations.

I find the names 'match' and 'matchAll' hardly something useful, or  
understandable.
The names of the methods should have something that binds them with their  
meaning, like matchCSSSelector or matchAllCSSSelector.

Plus, I'd like to have the possibility of having a live node-set or a  
snapshot only (currently supported).

Eventually with the growth of the web standards, more selection languages  
will appear, so there should be a common API to use several selection  
languages,
more or less like IE's setProperty/selectNodes/selectSingleNode.

Goodbye.
Received on Monday, 25 September 2006 03:23:17 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Friday, 22 June 2012 06:13:58 GMT