Accessing Fcurve key 's via scripting 2004-03-19 - By kevin korngut
Back
I was wondering if anyone knew of a way to check to see if a key's Slope
Length/Orientation was Broken or Unified and a way to set that via
scripting.
anyone have any ideas?
Thanks,
Kevin
Kevin Korngut
Systems Administrator
Janimation
www.janimation.com
-- --Original Message-- --
From: Matt Lind [mailto:speye_21@(protected)]
Sent: Thursday, September 11, 2003 3:33 PM
To: SI_XSI
Subject: Re: script: get jscripts 'for in' to work on selection
> On the linux implementaion of JScript, we dont have built in push and
> unshift methods (they don't show up until, JScript 5.X, I believe).
> Therefore we are left to create our own. If you prepend this to the
> head of your example, voila, a predictable creation order is returned.
>
> The built in push and unshift methods are producing the unreliable
results,
> not the for/in loop.
[Matt]
The built-in push() and unshift() methods are working just fine. If
they weren't, the standard for(;;) loop would also be broken.
With each call of push() or unshift() a new strip of memory is allocated
to hold the new data. This means the array, while a single entity in
script code, is actually fragmented across several memory spaces. The
standard
for(;;) loop is intelligent in that it checks the location of the
specified element in the array when being called, but the for/in loop
simply runs through the allocated memory on a first-come, first-serve
basis and returns whatever's there. Because the user cannot know or
control the layout of memory, the returned value of for/in is random as
illustrated in my previous example.
for/in 'appears' to work with your code because your custom unshift and
push methods rewrite the array as a series of single-index memory
spaces. Because there's only one way to read a memory space that
contains a single element, for/in 'appears' to work properly. However,
that doesn't fix for/in as you're just working around/hiding it's
faults. for/in still returns elements randomly from the memory spaces.
If Jscript goes through a garbage collection phase or your array picks
up data held in continguous blocks, memory layout can change and for/in
will exhibit the same problems all over again.
You still haven't proven for/in returns elements in their creation
order.
Matt
Date : Thu, 11 Sep 2003 11:02:44 -0700
To : XSI(at)Softimage.COM
>From : Andy Buecker <abuecker(at)ilm.com>
Subject : Re: script: get jscripts 'for in' to work on selection
On the linux implementaion of JScript, we dont have built in push and
unshift methods (they don't show up until, JScript 5.X, I believe).
Therefore we are left to create our own. If you prepend this to the head
of your example, voila, a predictable creation order is returned.
-- ---- ---- ---- ---- ---- ---- ---- ---- ---- ---- ---- ---
function push() {
for (var i=0; i<arguments.length; i++) {
this[this.length] = arguments[i];
}
}
function unshift() {
var temp = new Array();
for (var i=0; i<arguments.length; i++) {
temp[temp.length] = arguments[i];
}
for (var i=0; i<this.length; i++) {
temp[temp.length] = this[i];
}
for (var i=0; i<temp.length; i++) {
this[i] = temp[i];
}
}
Array.prototype.push = push;
Array.prototype.unshift = unshift;
-- ---- ---- ---- ---- ---- ---- ---- ---- ---- ---- ---- ---
The built in push and unshift methods are producing the unreliable
results, not the for/in loop.
It's been fun,
-A
---
Unsubscribe? Mail Majordomo@(protected) with the following text in
body: unsubscribe xsi
---
Unsubscribe? Mail Majordomo@(protected) with the following text in body:
unsubscribe xsi
|
|
