Adblock decoder ignore some portion when decoding #13
Comments
|
Caught another bug: After |
|
Okay you have to explain me AdBlock then @dnmTX 😸 I'm not a big fan of it as its syntax is confusing. So how do I differ legit from bad site in adblock ? I though that adblock was only about blocking not whitelisting 🤔 |
|
OK...... Much cleaner,no videos,no annoyances. |
|
Okay so what about this format ? Which of the following mark the domain as a bad or good boy ? I know you will not find them in real world but they are part of the tests for the decoder. |
|
The |
|
Okay I'm working on that implementation. So in this they are all legit right ? |
|
||google.com$script,image -this rule will not allow any scripts or images to be shown or executed on that domain |
|
Stay put,let me do some research on |
|
Ok... in the above example the rule |
|
Okay let me implement this issue first with the current format will then review with you for all tests as those need some hotfix. Never thought about whitelisting 😹 |
|
I know,it's Java,more complex.Took me a while to get around it but i'm getting there |
|
@funilrys make it simple.Everything that has |
|
Yeah but if I do that, I'll invalidate AdBlock/filter list like https://github.com/MajkiIT/polish-ads-filter 😸 |
|
Only need to take some time to understand how it works properly then will clean the mess I created! |
|
Look at this one for example.In it,all legit domains with rules to block certain elements only. |
|
Ok,i know it will take time but meanwhile,for everyone who uses the lists with dnsmasq etc etc and not adblockers. Can you PLEASE add https://raw.githubusercontent.com/Dawsey21/Lists/master/main-blacklist.txt to be filtered properly. |
|
Also you can start here,it's very well explained and will help you understand the basics: |
PLEASE |
This patch fix #13. Reverse: * Of the last patch for the way we check for URL. Introduction: * Of new test cases. * Of the force update for all version which are older than `0.94.3`. * Because of this patch. Review: * Of the way we extract domain and URL from the given adblock file. Deprecation: * Of all version which are equal or older than `0.109.0`. Thanks: * To @dnmTX * @adblockplus for their documentation * cf: https://adblockplus.org/filter-cheatsheet
|
@dnmTX , PyFunceble is fixed, please look at the tests for details. As you mentioned, there was really an issue with my way of handling adblock lists. Therefor here is the eratum: Please understand by As the tests were passed without any issue (cf.) I can attest that the next release and the current development version do not take any false positive anymore. Please let me know if there is something else. This issue will be closed on next release! Cheers, |
|
|
@funilrys from what i can tell and understand is What about anything with Another thing(just to make sure).Example: |
|
@dnmTX So from your point of view About
With my review, the domain which is in the href attribute is extracted and formatted (remove protocol and "decorators") 😸 |
|
Actually from my point of view the |
|
Wow you lost me 😹 For clarification, those are example of format do not consider those domains we are only talking about extracted domain from matched format 😸 Also if we match for example |
|
Maybe I misunderstood something 🤔 |
Ok,that's good,that's how it's suppose to be but...... |
|
Moved my answer to #227 as It's OT to OP's post and I hope there will be more activities in replies to this topic |
|
(my reply is also a reply to #227 (comment) at the same time):
|
This patch closes #227. This patch fixes #13. To quote @keczuppp (#13): > [.. ] but it seems you extract way too much in this mode on your own > and it might cause troubles... Therefore, I decided to rewrite the decoder completely. This patch introduces a real split between what is normally decoded and what is decoded within the aggressive mode. Within the "standard" mode, we only decode what is supposed to be blocked. On the other side, within the "aggressive" mode, we decode everything provided by the "standard" mode, plus everything behind a 'domain=' option or an 'href=' directive - if effective. Please report to the tests to understand the differences on a more deeper level and keep in mind that this new "direction" will evolve with the time. Decoding AdBlock or Filter lists is not an easy job and I hope to get much more feedback in the future. I didn't implement this because I have a use for it. But rather because it was asked by someone and I wanted to see if I was capable of implementing it. Now it's fully part of PyFunceble and people using it shouldn't be afraid to submit the "weird things" they find while using the decoder. Contributors: * @dnmTX * @jawz101 * @keczuppp * @kulfoon * @spirillen
|
Please take my commit and the underlying tests as the response. Is it still too much @keczuppp ? Let's discuss the future of that specific decoder. I'll inject any future report about missing decoding into the tests. So the more reports, the better that decoder will be 😄 As I wrote, I'm not one of those who write a filter list... So help or directions are welcome! |
|
Hello, I was already trying to test the new version of Adblock Decoder (
error
|
|
@keczuppp Thanks for the notice. I'll update the AdBlock decoder project as soon as possible. The simple way, is the Note to self: Cleanup documentation. |
|
So I've tried the newest version
Errors 1 log (EasyList)Errors 2 log (EasyList Polish)Errors 3 log (Official Polish Filters for AdBlock, uBlock Origin & AdGuard) |
|
@keczuppp, b37 is available and it should fix the error you reported. Thanks again for testing ! |
|
@keczuppp, the adblock-decoder is also upgraded to use the 4.0.0bX of PyFunceble. |
|
yep, good work:
more tests later |
|
And don't laught at me, fvcktard. |
spoiler errors log
|
Who is laughing at you? We are all here for some constructive work, enhancement, and discussion in our free time. If it is because of my emoji, sorry if it offended you. It wasn't meant to harm. Your last 3 cases are now into the source code that's going to be deployed next. I'm going to look into the issue of the standalone decoder later. |
This patch fixes funilrys/PyFunceble#13 (comment). Contributors: * @keczuppp
|
@keczuppp Please update and test the adblock-decoder |
Also:
================================================== OFF-TOPIC
Yes, it was about "simple way" + the emoji, your comment looks like you wanted to show how stupid I am just because I missed something which you describe as "simple" (a parameter + the fact it should be put in the conjunction with other parameters) in the (big) documentation, which might be not so obvious. Was it more funny to you, than your friend being unable to view history of my comment #13 (comment) ?
Really? Then why didn't you explain what was the purpose of the emoiji in your comment then. |
|
As for the
false hits |
This comment has been minimized.
This comment has been minimized.
This comment has been minimized.
This comment has been minimized.
This comment has been minimized.
This comment has been minimized.
This comment has been minimized.
This comment has been minimized.
This comment has been minimized.
This comment has been minimized.
This comment has been minimized.
This comment has been minimized.
|
funilrys, can we get some cleaning in this thread, could you put in the spoiler your OFF-TOPIC #13 (comment), just like I did with my OFF-TOPICS, thx |
|
OK, so I've just tested the newest PyFunceble dev right now and I've noticed that the reported issues mentioned in : The summarision:
Good improvement. |
As reported by @dnmTX at Ultimate-Hosts-Blacklist/dev-center#9:
are ignored.
The text was updated successfully, but these errors were encountered: