502536 - Omnibus ad-blocking bug for July 2009

Status: RESOLVED FIXED

(Camino Graveyard :: Annoyance Blocking, defect)

Description

[Smokey Ardisson (offline for a while; not following bugs - do not email)]

Trouble on Madison Avenue starts right here, or something like that :P

Comment 1

[Chris Lawson (gone)]

There are some ads here that sneak through:

http://www.mosnews.com/weird/2009/07/07/strongvagina/

(probably not safe for most workplaces, as it makes liberal use of the V-word).

cl

Comment 2

[Christopher Henderson]

Entire articles aren't available on the news.cnet.com subdomain. Eg:

http://news.cnet.com/8301-17938_105-10280682-1.html

The div[id^="ads_"] is to blame.

@-moz-document domain(news.cnet.com) {
div[id^="ads_"] { display: block !important }
}

Fixes it.

Comment 3

[philippe (part-time)]

fwiw,
1. the offending selector was added for some ads on spiegel.de (in bug 330801). As far as I can tell, it is not used anymore.
2. I commented out the offending selector and surfed quite a bit around on mainstream sites without side effects.

Do we want to fix that CNET issue quickly (using the fix in comment 2 as a temporary solution at least) ?

Comment 4

[Smokey Ardisson (offline for a while; not following bugs - do not email)]

That's the only cnet page I've found where that happens (well, the "Crave" section).  Cnet's probably a big enough site that we should push a fix now, though, along the lines of comment 2, but I'd prefer to whitelist just the div in question (ads_backsplashSkin).

Comment 5

[Smokey Ardisson (offline for a while; not following bugs - do not email)]

(In reply to comment #1)
> There are some ads here that sneak through:
> 
> http://www.mosnews.com/weird/2009/07/07/strongvagina/

Doesn't look like there's anything there we can easily block there; sorry.

---

Also on cnet, there are iframe[src="http://bwp.news.com/…"], which is basically a new domain for their ad-server (see http://bonsai.mozilla.org/cvsblame.cgi?file=mozilla/camino/resources/application/ad_blocking.css&rev=1.47&mark=425-427#425).  

We might consider replacing all of those with iframe[src^="http://bwp."]

Comment 6

[Chris Lawson (gone)]

(In reply to comment #5)
> We might consider replacing all of those with iframe[src^="http://bwp."]

But that'll block everything from my new domain bwp.com :-p

Comment 7

[Smokey Ardisson (offline for a while; not following bugs - do not email)]

On http://www.slate.com/id/2220430/

div[id="right_ad_wrapper"] or perhaps div[id^="p360"]

Comment 8

[Chris Lawson (gone)]

On 

http://letters.salon.com/tech/col/smith/2009/07/17/askthepilot327/view/?show=all

http://media.adfrontiers.com/cmimgs/5//146/without_300x250.jpg

showed up as part of

<div id="top" class="ad_content">
	<script type="text/javascript">
	<!--
		OAS_AD('TopLeft');
		OAS_AD('Top');
	// -->
	</script>
	<noscript><a href="http://judo.salon.com/RealMedia/ads/click_nx.cgi/www.salonmagazine.com/home.html@Top"><img src="http://judo.salon.com/RealMedia/ads/adstream_nx.cgi/www.salonmagazine.com/home.html@Top" border="0" alt="Advertisement" /></a><br /><br /></noscript>
</div>

cl

Comment 9

[Smokey Ardisson (offline for a while; not following bugs - do not email)]

(In reply to comment #7)
> On http://www.slate.com/id/2220430/
> 
> div[id="right_ad_wrapper"] or perhaps div[id^="p360"]

Additional p360 ads on http://www.newsweek.com/id/207837

# div id="p360-autotag-AFA28B5A-4F8E-11DE-882C-6423EDADD848"
# div id="p360-format-box"
# div id="p360-ad-list"
# a
# span class="p360-ad"

Comment 10

[Smokey Ardisson (offline for a while; not following bugs - do not email)]

On http://www.businessweek.com/investor/content/jul2009/pi20090723_059987.htm

<div class="module textAd">

Comment 11

[Chris Lawson (gone)]

On Cnet, we block the entire content here:

http://news.cnet.com/8301-17938_105-10296034-1.html

Probably a false-positive on div id="overviewHead".

Comment 12

[Smokey Ardisson (offline for a while; not following bugs - do not email)]

(In reply to comment #11)
> On Cnet, we block the entire content here:
> 
> http://news.cnet.com/8301-17938_105-10296034-1.html
> 
> Probably a false-positive on div id="overviewHead".

No, that's comment 2, as we've already established in comment 4.

Comment 13

[Chris Lawson (gone)]

Oh, heh, I had forgotten about that. Oops. Sorry for the bugspam :-p

Comment 14

[Smokey Ardisson (offline for a while; not following bugs - do not email)]

Attachment: July 09 Omnibus Ad-Blocking

This fixes everything here I've not already mentioned as unfixable, as well as the dependent bugs (within reason on the German ones).

I'm a little bit wary of div[class^="ad_"]; see bug 503745 comment 3.

(In reply to comment #9)
> > div[id="right_ad_wrapper"] or perhaps div[id^="p360"]

My notes remind me we can't use div[id^="p360"] (it breaks pbpBB), so I used one of the less specific rules from comment 9.

Comment 15

[Stuart Morgan]

Comment on attachment 391002 [details] [diff] [review]
July 09 Omnibus Ad-Blocking

sr=smorgan

I guess we'll just have to wait and see what happens with ad_

Comment 16

[Smokey Ardisson (offline for a while; not following bugs - do not email)]

Checked in on cvs trunk and CAMINO_2_0_BRANCH.

We'll want to take at least the CNet fix (and bug 503786) for 1.6.9 even if we can't yet take the whole patch.

Comment 17

[Smokey Ardisson (offline for a while; not following bugs - do not email)]

We'll need to take the alternate fix for bug 503745 comment 3, or leave out div[class^="ad_"] entirely, in 1.6.9; see bug 506849 comment 2.

Comment 18

[Smokey Ardisson (offline for a while; not following bugs - do not email)]

Attachment: Rollup patch for 1.6.9

This is a rollup patch for 1.6.9; it's bug 485865 (Apr/May/Jun), this bug, bug 507539 (whitespace/style cleanup), bug 499829 (unbreak lastfm.de), and the alternate fix for bug 503745 comment 3.

The lastfm.de and alternate fix are the only things that haven't been on trunk for at least 2 weeks.  The lastfm.de fix is strictly unbreaking, so I think it's safe (and we really need to ship it); I think the alternate fix for bug 503745 comment 3 is safe, but we can omit that rule entirely for 1.6.9 if philippe or Stuart feel strongly.

Comment 19

[Smokey Ardisson (offline for a while; not following bugs - do not email)]

…And Stuart's not on the cc list to see that comment :P

Comment 20

[philippe (part-time)]

(In reply to comment #18)
> I think the alternate fix for bug
> 503745 comment 3 is safe, but we can omit that rule entirely for 1.6.9 if
> philippe or Stuart feel strongly.

div[class^="ad_IDS"] feels much safer to me.

Comment 21

[Smokey Ardisson (offline for a while; not following bugs - do not email)]

Checked in the rollup patch for 1.6.9 on MOZILLA_1_8_BRANCH.