I ran a continuous test in both IE6 and NS7 and found calling
addChild as designed loses memory.
The test was continuous with 100th second delay. In IE memory
usage gradually increased a few bytes each minute in NS7 Megs were
consumed. After removing the destroy method, memory remained constant with IE
and increased very slowly with NS (bytes).
There were also significant speed gains from both browsers. These
finding are consistent with work I did with communicator about 6 years ago where
I couldn't find a way to destroy layers instead I opted to create a reusable
layer pool for discarded layers.
The Image loading issue I discussed was my system. I now know after a long
painful process that XP-Pro running IIS has a built in limit of 10
connections.
----- Original Message -----
Sent: Monday, June 23, 2003 6:23 AM
Subject: Re: [Dynapi-Dev] removing
layers
Hi Chris,
--- Chris Greener <[EMAIL PROTECTED]>
wrote:
> I'm new to the product so please bear with me.
>
Environment: 3.0 Beta/IE6
Welcome to the DynAPI dev group. Glad
to have you
aboard.
> When using the addChild() method to move a
dynlayer
> to another dynlayer parent the elm and it child
>
elements are first destroyed and then re-created. Is
> there a reason
for destroying the element layers?
> Also during this process child
layers with images
> intermittently stopped loading causing IE to
spin
> until the image load timed-out. Is this a known
> issue?
>
> I modified the _remove method to stop the elm from
>
being destroyed and modified _create method to
> re-use the same elm.
> something like the following:
> if(this.elm)
{
> parentElement =
this.parent.elm;
>
parentElement.appendChild(this.elm);
>
DynLayer._assignElement(this,this.elm);
>
DynElement._flagCreate(this);
> }
>
> This
considerably speeded up the re-draw process and
> seems to have fixed
the image load issue.
>
> Comments welcome?
>
How
does this fix affects the other browsers (such as
IE4, Mozilla, opera,
etc)? The reason for destroying
the elm was to release memory, etc. I'm not
sure it
DOM will automatically remove the elm from one parent
to the
next. But if this works fine then we can look
into it. To submit patches
you can 1) make your
change, 2) Zip the modified file into a .zip file,
3)
upload them through the patch service on DynAPI's
website.
http://sourceforge.net/tracker/?group_id=5757&atid=305757
>
Bugs found: Is this the right place to post them?
> dynlayer_ie:
setLocation and setPageLocation methods
> are broken. setClass needs
overriding
Well, normal you could post them to the bugs section
of
the site
(http://sourceforge.net/tracker/?group_id=5757&atid=105757)
or
you could send an email to the dev-help group, etc.
But since you've made
mention of it we can deal with
it as we speak.
Please provide a
working example of the bugs and send
it to this group.
Best
regards,
--
Raymond Irving
>
> Thx
>
ChrisG
>
>
>
>
>
>
__________________________________
Do you Yahoo!?
SBC Yahoo!
DSL - Now only $29.95 per month!
http://sbc.yahoo.com
-------------------------------------------------------
This
SF.Net email is sponsored by: INetU
Attention Web Developers &
Consultants: Become An INetU Hosting Partner.
Refer Dedicated Servers. We
Manage Them. You Get 10% Monthly Commission!
INetU Dedicated Managed
Hosting http://www.inetu.net/partner/index.php
_______________________________________________
Dynapi-Dev
mailing list
[EMAIL PROTECTED]
http://www.mail-archive.com/[EMAIL PROTECTED]/
|
