Call my last post a case of irrational exuberance. I recall now that I had looked at BeginNavigate2 before, too. It only fires for documents, not for images and embeds.
So I can't do all the fancy prefiltering I described in the last post. Nevermind.
Wednesday, May 27, 2009
Tuesday, May 26, 2009
Upcoming release
"Release early; release often."
I know I just pushed up 1.2, but I just realized I could hook the BeforeNavigate2 event and pre-filter downloads from blocked URLs before the download even begins. This should presumably save bandwidth, and stops blocked images/scripts from even loading (avoiding that momentary flicker on the screen).
It also, in my highly unscientific tests, seems to be faster.
There is a downside. In this event, I don't know
a) the type of the thing being downloaded (image, flash, etc)
b) the hosting site
This breaks some of the specificity of our rules, though I don't think that specificity is really used. But it means that if you had:
http://bad/
http://bad/
then downloads to http://bad/, even from site "exception," would be blocked. This is counter to the defined semantics.
So I think I'm going to break the rule semantics a bit for the next release, which will be coming shortly. I have to think about this a bit more--and look at what other adblockers do--before I figure out the right approach.
I know I just pushed up 1.2, but I just realized I could hook the BeforeNavigate2 event and pre-filter downloads from blocked URLs before the download even begins. This should presumably save bandwidth, and stops blocked images/scripts from even loading (avoiding that momentary flicker on the screen).
It also, in my highly unscientific tests, seems to be faster.
There is a downside. In this event, I don't know
a) the type of the thing being downloaded (image, flash, etc)
b) the hosting site
This breaks some of the specificity of our rules, though I don't think that specificity is really used. But it means that if you had:
then downloads to http://bad/, even from site "exception," would be blocked. This is counter to the defined semantics.
So I think I'm going to break the rule semantics a bit for the next release, which will be coming shortly. I have to think about this a bit more--and look at what other adblockers do--before I figure out the right approach.
Monday, May 18, 2009
Monday, May 11, 2009
Bleeding Edge: Try a new, comprehensive ruleset
I haven't yet gotten around to writing a converter for Adblock Plus's filter rule lists (which is slightly complicated by the fact that we don't have a complete 1:1 mapping between our respective functionality, though it is close), but in the meantime, I spent a couple of minutes to create a host-based filter ruleset from the list at yoyo.org.
If you want to try this rather large rule file (nb: this many rules may degrade AdblockIE's performance), you can download it from . You should then copy this file to C:\Program Files\af0.net\AdblockIE\blacklist.xml (or wherever you installed AdblockIE to), overwriting the existing file. Reload IE and you're good to go.
Coming soon: An Adblock Plus-compatible importer.
If you want to try this rather large rule file (nb: this many rules may degrade AdblockIE's performance), you can download it from . You should then copy this file to C:\Program Files\af0.net\AdblockIE\blacklist.xml (or wherever you installed AdblockIE to), overwriting the existing file. Reload IE and you're good to go.
Coming soon: An Adblock Plus-compatible importer.
Subscribe to:
Posts (Atom)