it does not matter hopw may layers you have.
If you are using layers, objects, eventlisteners and arrays, of all of
these,
you WILL have a mem-leak, more objects, bigger leak,less objects less leak..
Good object management can decrease the leak drastically though,
and good object management is what the team has been working on..
I must admin, that while the API is not 100% stable (Not's it's fault tho,
the browsers themselves are unstable), it is better than it ever whas, and
WAY better than any other javascript library in existance (and I've
looked!!)
Doug
----- Original Message -----
From: "Alex Chong" <[EMAIL PROTECTED]>
To: <[EMAIL PROTECTED]>
Sent: Monday, May 14, 2001 6:17 PM
Subject: [Dynapi-Help] Re: memory leak II
> Hi,
>
> A couple of questions to make sure I understand the situation clearly:
> 1) If I create 500+ layers, lots of nested layers, ....etc, even if I am
NOT using the DynAPI library, there will still be memory leak and the
browser will crash?
>
> 2) Am I safe to assume that if I have a page with 499 Dynlayers with a
maximum of 4 nested child and no recursive calls, no matter how many times I
refresh my browser, it wont crash?
>
> I have been following the DynAPI forum everyday but have somehow missed
the message about Jordi fixing the memory leak problem. So what exactly was
the problem? Could you kindly show me a
> short example of how a combination of DHTML codes that lead to memory
leak? Thanks, you guys are genius.
>
> > Subject: Re: [Dynapi-Help] Re: Please help
> > Date: Mon, 14 May 2001 07:38:36 +0200
> > From: "Richard Bennett" <[EMAIL PROTECTED]>
> > Reply-To: [EMAIL PROTECTED]
> > To: <[EMAIL PROTECTED]>
> > References: <[EMAIL PROTECTED]>
<[EMAIL PROTECTED]>
> >
> > No idea about your messages.
> >
> > On the leaking:
> > I define a memory leak as meaning that the ram in use for your page is
not
> > dynamically released, and keeps increasing , either when you refresh the
> > page, or during continuous animations.
> >
> > DynAPI used to have this problem, now it doesn't .
> > DHTML does not have this problem inherently, but some things might
create a
> > leak. for instance, building a lot of layers (500+), nesting them in
tables,
> > and other div's (five layers deep, or more), and having dozens of
> > setTimeout() and setIntervals, calling themselves recursively.
> >
> > What I do not consider a memory leak, is that if you create a tree with
2000
> > nodes, it takes 30 seconds to render, as long as it takes the same
amount of
> > time after 10 refreshes, that's not a leak . (in practice there is an
> > increase after the first refresh, but it stays stable from then on).
> > This is processor use, which would be improved by more efficient code.
> >
> > To recap, there used to be a memory leak in DynAPI, which was said to be
> > caused by a browser bug, but it was not, and Jordi managed to eliminate
it
> > completely. So the problem IS now solved.
> > As in any situation, you have to keep your code efficient though.
> >
> > I edited the FAQ to reflect this:
> >
http://dynapi.sourceforge.net/doccenter/index.php?FrequentlyAskedQuestions
> >
> > Cheers,
> > Richard Bennett
> >
> > [EMAIL PROTECTED]
> > www.richardinfo.com
> > (Everything running on, and ported to DynAPI2.53)
> > visit the DynAPI homepage (and FAQ) ::
> > http://dynapi.sourceforge.net/dynapi/index.php?menu=1
> > Browse (and search) the mailinglist here:
> > http://www.mail-archive.com/index.php3?hunt=dynapi
> >
> > ----- Original Message -----
> > From: "Alex Chong" <[EMAIL PROTECTED]>
> > To: <[EMAIL PROTECTED]>
> > Sent: 14 May, 2001 06:08
> > Subject: [Dynapi-Help] Re: Please help
> >
> > > Hi all,
> > >
> > > When I choose to receive Dynapi messages in Digested mode, how do I
reply
> > to an individual message within the digest? Currently, what I did was
to
> > click on "reply" and then delete off all
> > > messages but the one I need to reply to. It is difficult when the
digest
> > has large amount of content.
> > >
> > > Thank you Scott for your reply. But I wasnt referring to your claim.
> > There are quite a few people who have claimed that the memory leak is
caused
> > by the way browsers handle DHTML and NOT
> > > Dynapi. I hope to verify this. For example, if I do simple DHTML
stuffs
> > like just creating lots of layers, will this cause memory leaks? If it
> > does, then we can safely assume that there
> > > is nothing we can do in Dynapi to improve the situation. But if it
> > doesnt, that means only certain DHTML codes causes memory leak. In this
> > case, we can avoid those codes in Dynapi and the
> > > problem is solved. Please comment.
> > >
> > > > Subject: RE: [Dynapi-Help] Memory leaks
> > > > Date: Fri, 11 May 2001 05:20:33 -0700
> > > > From: Scott Smith <[EMAIL PROTECTED]>
> > > > Reply-To: [EMAIL PROTECTED]
> > > > To: "'[EMAIL PROTECTED]'"
> > > > <[EMAIL PROTECTED]>
> > > >
> > > > I you are referring to my claims of memory leaks at
> > www.richardinfo.com - I
> > > > was WRONG.
> > > >
> > > > See, I CAN admit when I'm wrong.
> > > >
> > > > Scott
> > > >
> > > > > -----Original Message-----
> > > > > From: Alex Chong [mailto:[EMAIL PROTECTED]]
> > > > > Sent: Friday, May 11, 2001 12:15 AM
> > > > > To: [EMAIL PROTECTED]
> > > > > Subject: [Dynapi-Help] Memory leaks
> > > > >
> > > > >
> > > > > Hi, I need help with two problems:
> > > > >
> > > > > First, I choose to receive Dynapi messages in Digested mode.
> > > > > But when I
> > > > > want to reply to an individual message within a digest, I
> > > > > dont know how.
> > > > >
> > > > > Secondly, there are a couple of messages that suggest that the
memory
> > > > > of IE is caused by the way IE handles DHTML as a whole and not
Dynapi.
> > > > > Has anyone verify this with concrete proof, example, can
> > > > > anyone write a
> > > > > piece of small simple DHTML code that still crashes IE?
> > > > >
> > > > > --
> > > > > Alex Chong
> > > > > School of IT,
> > > > > Murdoch University, Australia
> > > > >
> > > > >
> > > > >
> > > > > _______________________________________________
> > > > > Dynapi-Help mailing list
> > > > > [EMAIL PROTECTED]
> > > > > http://lists.sourceforge.net/lists/listinfo/dynapi-help
> > > > >
> > > >
> > >
>
------------------------------------------------------------------------
> > > >
> > >
> > > -----
> > > Alex Chong
> > > School of IT,
> > > Murdoch University, Australia
> > >
> > >
> > >
> > > _______________________________________________
> > > Dynapi-Help mailing list
> > > [EMAIL PROTECTED]
> > > http://lists.sourceforge.net/lists/listinfo/dynapi-help
> > >
>
>
> _______________________________________________
> Dynapi-Help mailing list
> [EMAIL PROTECTED]
> http://lists.sourceforge.net/lists/listinfo/dynapi-help
_______________________________________________
Dynapi-Help mailing list
[EMAIL PROTECTED]
http://lists.sourceforge.net/lists/listinfo/dynapi-help
