Jason Grout wrote:
> From playing with putting alerts into the javascript code, the jump
> seems to occur exactly when the innerHTML is replaced in set_output_text
> javascript function, in the following lines:
>
>
> if (new_interact_output.indexOf('__SAGE_INTERACT_RESTART__') !=
> -1) {
> evaluate_cell(id, 0);
> } else {
> cell_interact.innerHTML = new_interact_output; # <-- HERE
> if (contains_jsmath(new_interact_output)) {
> jsMath.ProcessBeforeShowing(cell_interact);
> }
> }
>
>
> My guess is that it has to do with the browser all of sudden losing the
> element it was focused on or something, so the browser jumps up. Does
> anyone know what if this is the case?
From some more experimentation, it seems like this jumping happens when
the interact is the last thing in the document (or nearly the last
thing). As near as I can tell, what is happening is this:
1. @interact requests an update, based on a changed control
2. after a while, Sage sends back the new output to put in the interact cell
3. Using the above code, the old output is replaced with the new output.
This does:
(a) Delete the old output. The page is then shorter, so the browser
scrolls up so that the interact with the *deleted* output is at the
bottom of the window
(b) The new interact output is put below the controls. This
lengthens the page, but the browser does not scroll back down.
A workaround is to make the page long enough after the interact that the
browser does not have to scroll when the interact is deleted. You can
do this, for example, by just inserting a bunch of blank cells which
take up at least as much room as the interact output did.
I'm trying to figure out a way to fix this.
Thanks,
Jason
--~--~---------~--~----~------------~-------~--~----~
To post to this group, send email to sage-devel@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at http://groups.google.com/group/sage-devel
URLs: http://www.sagemath.org
-~----------~----~----~----~------~----~------~--~---
