- From: Joseph Kesselman/Watson/IBM <keshlam@us.ibm.com>
- Date: Fri, 20 Oct 2000 10:30:05 -0400
- Cc: <www-dom@w3.org>
>Looking at the DOM2 docs I found that the event model does not discuss a >"selectStart" event. [...] At this time the DOM doesn't have the concept of "selection". That's considered a higher-level function, and the closest we come to it is to point out that one of many ways to represent a selection might be as one or more Ranges. That doesn't mean we can't handle such an event, since the DOM event model allows folks to create and distribute their own events, in addition to the ones we've standardized. It just means we haven't said anything about what that event might be called, or might contain. Your browser can implement selectStart on top of the DOM... but code which looks inside that event may not be portable from platform to platform until selectStart's contents are standardized. We do expect to add keyboard events in DOM Level 3; we might be able to consider this one at the same time. >However, when the user starts "dragging" the browser begins selecting the >text on the page. Can anyone recommend a "correct" way for avoiding this >while sticking to the DOM1/DOM2 programming model. If that behavior is itself being mediated through the DOM event model, then setting up a listener which blocks some of the "default" handling might work. If it's being implemented directly by the browser, there probably isn't anything the DOM can do about it. Note too that, at this time, the DOM says nothing about how events are generated -- only about how they are distributed and what a specific set of them is expected. So at some level you _are_ dependent upon higher-level code to issue the appropriate events, at the appropriate time, with the appropriate content. Other events may be standardized in future levels of the DOM. In fact, other events _will_ be; key events were explicitly deferred to Level 3 pending input from the i18n and accessibility working groups. ______________________________________ Joe Kesselman / IBM Research
Received on Friday, 20 October 2000 10:30:46 UTC