User Agent: Mozilla/5.0 (Windows NT 5.1; rv:9.0) Gecko/20100101 Firefox/9.0 Build ID: 20111216140209 Steps to reproduce: Went to a news site (http://www.rtl.nl/financien/rtlz/vandaag/) that I visit frequently. Actual results: Most of the HTML is rendered as expected, however an iframe in the center of the page remains white. In addition, in the tab the word "Connecting..." continues to be shown as well as the turning green circle. The problem coincides with the release of FF version 9 and an upgrade to Java Version 6 Update 30. However, since this is the only site that I visit that has this problem I suspect FF Release 9. Expected results: The large white rectangle should have contained stock exchange data.

Confirmed on Firefox 9.0.1 to Nightly12.0a1. #1 For Firefox9.0.1 http://hg.mozilla.org/releases/mozilla-release/rev/c4405d7a95f6 Mozilla/5.0 (Windows NT 6.1; WOW64; rv:9.0.1) Gecko/20100101 Firefox/9.0.1 ID:20111220165912 Error: attempt to run compile-and-go script on a cleared scope Source file: http://www.quintrics.nl/rtlz/website/js/dbAandelen.js Line: 108 Error: LS_FqR is not defined Source file: http://www.quintrics.nl/rtlz/website/ls/lspushpage.js Line: 9 Error: attempt to run compile-and-go script on a cleared scope Source file: http://www.quintrics.nl/rtlz/website/ls/lspushpage.js Line: 9 Error: attempt to run compile-and-go script on a cleared scope Source file: http://www.quintrics.nl/rtlz/website/jq/jquery-1.4.2.min.js Line: 130 Regression window(m-c) Works: http://hg.mozilla.org/mozilla-central/rev/6dc468c41136 Mozilla/5.0 (Windows NT 6.1; WOW64; rv:9.0a1) Gecko/20110821 Firefox/9.0a1 ID:20110821030758 Fails: http://hg.mozilla.org/mozilla-central/rev/482742e6fff7 Mozilla/5.0 (Windows NT 6.1; WOW64; rv:9.0a1) Gecko/20110821 Firefox/9.0a1 ID:20110821111959 Pushlog: http://hg.mozilla.org/mozilla-central/pushloghtml?fromchange=6dc468c41136&tochange=482742e6fff7 Regression window(m-i) Works: http://hg.mozilla.org/integration/mozilla-inbound/rev/b727463566a1 Mozilla/5.0 (Windows NT 6.1; WOW64; rv:9.0a1) Gecko/20110820 Firefox/9.0a1 ID:20110820184005 Fails: http://hg.mozilla.org/integration/mozilla-inbound/rev/857f058efa56 Mozilla/5.0 (Windows NT 6.1; WOW64; rv:9.0a1) Gecko/20110820 Firefox/9.0a1 ID:20110821013315 Pushlog: http://hg.mozilla.org/integration/mozilla-inbound/pushloghtml?fromchange=b727463566a1&tochange=857f058efa56 Triggered by: 857f058efa56 Matheus Kerschbaum — Bug 679971 - Remove Navigator.taintEnabled(); r=gal #2 For Firefox10.0beta http://hg.mozilla.org/releases/mozilla-beta/rev/c130e43aa4b5 Mozilla/5.0 (Windows NT 6.1; WOW64; rv:10.0) Gecko/20100101 Firefox/10.0 ID:20111221135037 Error: $ is not defined Source file: http://www.quintrics.nl/rtlz/website/js/dbAandelen.js Line: 414 Error: LS_FqR is not defined Source file: http://www.quintrics.nl/rtlz/website/ls/lspushpage.js Line: 9 Error: attempt to run compile-and-go script on a cleared scope Source file: http://www.quintrics.nl/rtlz/website/jq/jquery-1.4.2.min.js Line: 128 Error: attempt to run compile-and-go script on a cleared scope Source file: http://www.quintrics.nl/rtlz/website/ls/lspushpage.js Line: 9 Error: attempt to run compile-and-go script on a cleared scope Source file: http://www.quintrics.nl/rtlz/website/jq/jquery-1.4.2.min.js Line: 130 #3 For Aurora11.0a2 and Nightly12.0a1 http://hg.mozilla.org/releases/mozilla-aurora/rev/e08be766bdba Mozilla/5.0 (Windows NT 6.1; WOW64; rv:11.0a2) Gecko/20111221 Firefox/11.0a2 ID:20111221172848 http://hg.mozilla.org/mozilla-central/rev/ed47a41ba26a Mozilla/5.0 (Windows NT 6.1; WOW64; rv:12.0a1) Gecko/20111222 Firefox/12.0a1 ID:20111222031055 Error: Component returned failure code: 0x8000ffff (NS_ERROR_UNEXPECTED) [nsIDOMHTMLDocument.writeln] Source file: http://www.quintrics.nl/rtlz/website/ls/lspushpage.js Line: 9

> 857f058efa56 Matheus Kerschbaum — Bug 679971 - Remove Navigator.taintEnabled(); r=gal Sounds like this site is doing broken browser-sniffing?

(In reply to Boris Zbarsky (:bz) from comment #2) > > 857f058efa56 Matheus Kerschbaum — Bug 679971 - Remove Navigator.taintEnabled(); r=gal > > Sounds like this site is doing broken browser-sniffing? It works on 9.0.1 to 12.0a1 if I set UA as follows. user_pref("general.useragent.override", "Mozilla/5.0 (Windows NT 6.1; WOW64) AppleWebKit/535.7 (KHTML, like Gecko) Chrome/16.0.912.63 Safari/535.7");

Yeah, definitely broken sniffing. John, can you please report the issue to the site, since you're actually one of their users?

http://www.quintrics.nl/rtlz/website/ls/lscommons.js ...if((document.childNodes)&&(!document.all)&&(!navigator.taintEnabled) ...

Yeah, and then inside that clause it UA-sniffs different webkit versions..... and fails to detect us as one and then goes off the rails.

> http://www.quintrics.nl/rtlz/website/ls/lscommons.js quintrics.nl redirects to http://www.sharecompany.nl So this code is probably not only used on rtl.nl

(In reply to Boris Zbarsky (:bz) from comment #4) > Yeah, definitely broken sniffing. > > John, can you please report the issue to the site, since you're actually one > of their users? Done. FYI: http://www.w3schools.com/jsref/met_nav_taintenabled.asp suggests pbm. is not limited to FF and found similar bug reported on jQuery site.

Hi, I'm from the company that wrote the incriminated code. That's indeed a bug in the old version of our libraries; an upgrade to the latest version would solve it. Sorry for the trouble.

anyway looks like http://www.rtl.nl/financien/rtlz/vandaag/ already did the upgrade so that the issue with that website should be resolved

(In reply to Mone from comment #9) > Hi, > > I'm from the company that wrote the incriminated code. That's indeed a bug > in the old version of our libraries; an upgrade to the latest version would > solve it. > > Sorry for the trouble. Confirmed that the iframe is now being rendered again. Appreciate it, it's a bit annoying to have to use two browsers. No need to apologize, bugs happen :) In addition based on the comments in this ticket I suggest you interpret this bug report as being helpful to get rid of deprecated code rather than incriminating.

> anyway looks like http://www.rtl.nl/financien/rtlz/vandaag/ already did the upgrade Is that based on source inspection? See below... If this is a general library, would it be possible for your company to contact past customers and get them to update their installations? It would be nice to be able to actually remove taintEnabled. > Confirmed that the iframe is now being rendered again. That may well be because we backed out bug 679971 because of all the pages that were sniffing for taintEnabled. :(

Reopening pending answer to first part of comment 12, btw.

>Is that based on source inspection? Yes, I took a look to the included file to verify the library version. I confirm that the current library works as expected on both cases (correct detection / wrong detection) >It would be nice to be able to actually remove taintEnabled. The detection via taintEnabled is still in place but with the result of that detection nothing "important" is executed (on previous versions we used that check to choose between appending an iframe and document.writing it). I'm going to check if I can remove it completely for the next version. >would it be possible for your company to contact past customers and get them to update their installations? We always notify our customers to upgrade but obviously we can't force them. We also distribute a free version, I can't contact users of that version. (anyway note that the version without the bug was released on 4 Feb 2011 in beta version and on 7 Jun 2011 as final)

Ah, OK. So pretty recent... Thanks for confirming that the site is actually fixed!