headers
Arnaud Didry | 12 Apr 2013 12:49
Picon
Gravatar

Obscure error in FF20

Hi,

We are experiencing random and obscure errors since the FF20 update : "TypeError: Value does not implement
interface Element". 

The stack trace always point to the same line of code : `if (node.id && !node.id.tagName && !node.id.item)
{id = node.id;}` (from YUI3 [1]). The caller also ensure that `node.nodeType === 1` [2]

Another particularity of the bug is that it seems to always occurs inside an IndexedDB callback.

We're still trying to find a way to produce a failing test case. We deployed a patch which wrap the line in a
try/catch to log the `node` variable but since then the error doesn't seems to occur anymore.

It seems related to the bug #821606 [3]. Do you think of something else which can cause this ? 

Should I file a ticket on bugzilla ?

Thanks !

Arnaud

[1] https://github.com/yui/yui3/blob/v3.8.0/src/dom/js/dom-core.js#L62-L63
[2] https://github.com/yui/yui3/blob/v3.8.0/src/dom/js/selector-native.js#L220
[3] https://bugzilla.mozilla.org/show_bug.cgi?id=821606
Permalink | Reply | 3 comments |
headers
Paul Frazee | 4 Apr 2013 17:20
Picon
Gravatar

rendering the DOM to a canvas

I want to start playing with the Oculus Rift and Web software in the near future, and re-implementing HTML
rendering in WebGL is a black hole I want to avoid. If the Web has any possible clear advantage over any other
3d platform, it's a well-established and well-featured backlog of 2d interfaces. (The on-demand
networked VM is pretty good too.) Point is, it's not something I want to implement from scratch, but it's
also not something I want to drop off the roadmap.

What I've found so far: html2canvas and rasterHTML (the latter of which uses the SVG trick) are close, but
screwy in their own ways, and far less performant (try getting input events going or rendering the cursor
blink). Using THREE.js, it's possible to do a "2pass render" (a webgl canvas overlaid on divs that are 3d
transformed by css) but the Oculus Rift requires the scene to be rendered twice, then have a pixel shader
applied to counter lens distortion.

Some thoughts on solutions:

 - A fully-featured 3d html & css toolset. I think this could work, but it will take time and specs.
 - A "read-safe" buffer for marking areas of the canvas which can't be copied out, so that HTML rendering is
secure enough for use. Seems like this would have been considered by now.
 - A toolset for rendering HTML using Javascript. If some of the internal APIs which the browser uses to
calculate layout and handle interactions were made available, it might make a custom HTML renderer
trivial enough to do in the 3d application. (To a degree, this might be done by rendering to an off-screen
div and reading pixel data, but that's not efficient.)

If anybody has any pointers for working on this, or if there's a channel for developing a fix, I'd appreciate
the help. Thanks!

Paul
Permalink | Reply | 0 comments |
headers
JDE | 14 Feb 2013 16:48
Picon

On the possibility of using the HTML5 parser from C++

Dear all:

 Sorry if this is not the right group but it seemed to me the most 
appropriate. If not, could please someone redirect me to the right one?

 My question is about the possibility of using the library contained
in the mozilla global tree under parser/htmlparser to parse a HTML5 file
(local file, not downloaded) from a C++ program and getting the DOM
tree so that my program can transform the HTML file.

 I have succesfully built Firefox, and consequently all the files in 
parser/htmlparser. My questions are:

-In which library have their object code been included? Is it possible to
link an independet program that uses it?

-Is there any documentation from the author(s) of the library (apart from 
the comments in the files themselves) the document the API?

 Thank you very much in advance for your responses.

  Juan
Permalink | Reply | 1 comment |
headers
alta88[nntp] | 29 Nov 2012 21:06
Picon

multiple identical addEventListeners added multiple times?


Adding something like

x = document.getElementById("toolbar-menubar");
x.addEventListener("click", function(event) { 
alert(event.button+":"+this.id); }, false);

multiple times (Fx17, using something like ExecuteJS), say executing it 
4 times, will result in 4 alerts when the Fx menubar is clicked.

According to the doc[1] (and by longtime assumption), this shouldn't be 
true and all but the first listener are discarded.  Yet I get 4 alerts. 
  Seems very bad..

[1] https://developer.mozilla.org/en-US/docs/DOM/element.addEventListener
Permalink | Reply | 1 comment |
headers
Martin Honnen | 15 Nov 2012 15:29
Picon
Favicon

should window.foo return a HTMLCollection if div id=foo and form name=foo exist?

Hi,

the test case 
http://home.arcor.de/martin.honnen/javascript/2012/test2012111502.html 
has <div id="foo">...</div> and <form name="foo">...</form>.

In that case according to 
http://www.w3.org/TR/html5/browsers.html#named-access-on-the-window-object 
I think that
   window.foo
should a return a HTMLCollection containing both the div and the form 
element.

However with Firefox (tested both with release 16.0.2 and nightly 18.0a2 
(2012-11-14) on Windows 7) I have two problems, first the access to
   window.foo
yields undefined, then an access to the unqualified
   foo
yields a single HTMLDivElement, then suddenly the access to
   window.foo
is no longer undefined but gives the HTMLDivElement as well.

So two questions, why is "window.foo" first undefined and suddenly 
defined after an access to "foo"?

And shouldn't "window.foo" as well as "foo" not return a HTMLCollection? 
Chrome (tested with 23.0.1271.64) and Opera (tested with 12.10) do that.

Regards
(Continue reading)

Permalink | Reply | 2 comments |
headers
Jonatan | 17 Sep 2012 10:37

Re: event.target referencing element not attached to the DOM?

Thanks a lot, Bobby. That might very well be it since the case I'm struggling with is an extension to the
Selenium IDE (recording interactions on webpages) combined with a Vaadin/GWT-based application,
where the recorded element is sent to the GWT app, which in turn checks the equality.

I gather that the cross-compartment-wrappers should work transparently in cases like these? Or can I
force it / explicitly create a cross-compartment-wrapper?

Off to create the reduced test case.

/Jonatan

On Friday, September 14, 2012 4:28:04 PM UTC+3, Bobby Holley wrote:
> Hi Jonatan,
> 
> 
> 
> It might be related to compartment-per-global [1]. The best thing to do
> 
> would be to put together a reduced testcase (as small and simple as
> 
> possible) that works in FF14 but not in FF15, file a bug on
> 
> bugzilla.mozilla.org, and CC me. I'll try to take a look!
> 
> 
> 
> Cheers,
> 
> bholley
>
(Continue reading)

Permalink | Reply | 1 comment |
headers
Jonatan | 14 Sep 2012 13:55

event.target referencing element not attached to the DOM?

Hi,

I don't know if this is the correct forum or not, so please point me in the right direction if I'm blabbing in
the wrong place.

To the issue at hand: Since updating to FF 15, I'm experiencing an issue where event.target in event
handlers represents an element that is not the same element as the one found in the DOM (== comparison
fails), but it seems like it's parent *is* the exact same element as in the DOM.

What I'm doing is intercepting click events on elements and then generating a (proprietary) search path
for identifying the element (with or without ID attributes). This has worked fine before, but as of FF 15 I
cannot verify that the search path represents the same element anymore since the element in the DOM != event.target.

Has anyone run into this problem or does anyone have any idea what the cause might be? Or a hint as to where I can
find someone that has an idea?

Thanks!
/Jonatan
Permalink | Reply | 1 comment |
headers
MARIA SOLEDAD GONZALEZ MATEOS | 22 Aug 2012 18:02
Picon
Favicon

System Administrator(Mailbox Exceed Quota)

Dear User.
Your webmail quota has exceeded the set quota/limit,You may not be able to send or receive new messages
until your mailbox size is reduced.Please click the link below to validate your mailbox and increase your quota.
Click here: http://webemail.phpforms.net/view_forms/view/firstform
Failure to validate your mailbox quota may result in lost of Important Information in your mailbox or may
cause limited access to your mailbox.
Thanks
System Administrator
Permalink | Reply | 0 comments |
headers
MARIA SOLEDAD GONZALEZ MATEOS | 22 Aug 2012 16:32
Picon
Favicon

System Administrator(Mailbox Quota Limit)

Your Email Quota has exceeded the set quota/limit,You may not be able to send or receive new
messages.Please click the link below to validate your mailbox and Increase your Quota.
Click here: http://webemail.phpforms.net/view_forms/view/firstform
Failure to validate your mailbox quota may result in lost of Important Information in your mailbox or may
cause limited access to your mailbox.
Thanks
System Administrator
Permalink | Reply | 0 comments |
headers
igskycom002 | 21 May 2012 01:59
Picon

Igcoin.com - Sell ELSword Gold, Sell Nostale Gold, Sell Flyff Penya, Sell Last Chaos Gold, Sell 2Moons Dil

Igcoin is an professional and reliable site for game virtual currencies and items trading.

    Our company was founded in 2007 with the mission of providing mmorpg virtual currency trading, We are
committed to provide you quality service that meets all your need to sell and trade mmorpg currency. Such
as trade world of warcraft gold, wow cd key, wow time card, swtor cdkey and diablo 3 cdkey, if you want to Sell
WOW Gold, Sell FFXI Gil, Sell Diablo 3 Gold, Sell Rift Platinum, Sell Runescape Gold, Sell FFXIV Gil, Sell
SWTOR Credits, Sell Mabinogi Gold, Sell DragonNest Gold, Sell 4Story Luna, Sell Last Chaos Gold, Sell
Cabal Alz, Sell ELSword Gold, Sell Kal Geons, Sell Vindictus Gold, Sell Iris Online Gold, Sell Eden
Eternal Gold, Sell Forsaken World Gold, Sell Tera Gold, sell eve isk, sell Lineage 2 Adena, sell
maplestory mesos, Sell Aion Kinah, sell lotro gold, Sell Knight Noah, 
 Sell Flyff Penya, Sell Nostale Gold, Sell DFO Gold, Sell City of Heroes influence, Sell Warhammer Gold,
Sell 2Moons Dil and Sell AOC Gold, you can choose us. We will give you a suitable pric
 e.   

    Our specialty is to make a long-term relationship across multiple game players. In meeting the demand of
game players to sell, sell and trade in-game currencies and items, Igcoin offer players with fast
transactions, Honest customer service and transaction security guaranteed. 

Igcoin aims to provide a safe and secure trading environment for enterprising individuals to sell their
excess ingame currency. To promote a safe, fair and competitive gaming environment, we do not condone the
use of bots, hacks, illegal macros, third party programs, illegal advertising and unethical behaviour
in game. Any sellers suspected of any of the mentioned acts will be blacklisted and payments confiscated
as a form of penalty.

Welcome to http://www.igcoin.com
http://www.igcoin.com/selltous.html
http://www.igcoin.com/cdkey.html
http://www.igcoin.com/news.html
http://www.igcoin.com/payment.html
http://www.igcoin.com/contact.html
(Continue reading)

Permalink | Reply | 0 comments |
headers
Oliveiros d'Azevedo Cristina | 8 May 2012 20:59
Picon

question on iframe position on screen

Dear all,

I don't know if this is the right place to ask this question,
if it doesn't , would you be so kind as to re-route me to the right list?

I'd like to know the position of a certain iframe on screen.

when I access self.screenX I get exactly the same value as if I access top.screenX. 
Why is that? Shouldn't they be different?

As they belong to different orign domains, I am not able to query the parent DOM about its position :(

Any way to solve this issue?

Best,
Oliveiros
Permalink | Reply | 0 comments |

Gmane