You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
For example, I noticed that in the PR to push $removeparam to stable brave/brave-browser#23927 that they used ||uni-verse.github.io^$removeparam=test to test it and get approved... I don't know why the $removeparam wasn't tested with more rules and not just the ONLY one that works.
Nothing else will work from the $removeparam list, unless the list is modified, and I am talking that the feature is not compatible with rules like:
Or something like ||amazon.*$removeparam=at_link_id or ||bbc. that are made that way so it covers all domains from whatever company, just like happens with the ones for Yandex and Aliexpress as well.
And also it is not compatible with rules like ||mkaff.com/landings/$removeparam or ||alza.de/*.htm$removeparam=kampan or just anything in between.
The removeparam feature works great for the most part, but not when loading those lists because the majority of rules for some reason are expecting to have a $document to make them work (so it looks like example$document,removeparam), and without $document the rule is not going to work, unless it is just a normal simple one like ||example^
This is a big incompatibility for lists, and while I took the Adguard list and added $document to all the rules and they work fine, updating the lists and making sure to remove the ones already included in Query String Filter so it doesn't cause conflict (crashes) takes time.
Hope this helps to improve it so lists can work without problems or modifying anything.
The text was updated successfully, but these errors were encountered:
@Emi-TheDhamphirInLoveUnderTheFrozenStar thanks for reporting.
uBlock Origin only implicitly adds $document to network rules with exact hostname patterns; rules with patterns that aren't just something like ||example.com^$ wouldn't apply to documents by default. Apparently it also adds $document to all removeparam rules too (as does AdGuard).
This should be fixed by 8a755bd, which I'll get released soon.
Hi @antonok-edm
I finally made an account to report these incompatibilities in the syntax, which make it impossible to load a list like the one @fmarier mentioned https://github.com/AdguardTeam/FiltersRegistry/blob/master/filters/filter_17_TrackParam/filter.txt, and don't have to modify it in order to use $removeparam.
For example, I noticed that in the PR to push $removeparam to stable brave/brave-browser#23927 that they used
||uni-verse.github.io^$removeparam=test
to test it and get approved... I don't know why the $removeparam wasn't tested with more rules and not just the ONLY one that works.Nothing else will work from the $removeparam list, unless the list is modified, and I am talking that the feature is not compatible with rules like:
generic remove param rules like:
$removeparam=famad_xuid
Or something like
||amazon.*$removeparam=at_link_id
or||bbc.
that are made that way so it covers all domains from whatever company, just like happens with the ones for Yandex and Aliexpress as well.And also it is not compatible with rules like
||mkaff.com/landings/$removeparam
or||alza.de/*.htm$removeparam=kampan
or just anything in between.The removeparam feature works great for the most part, but not when loading those lists because the majority of rules for some reason are expecting to have a
$document
to make them work (so it looks likeexample$document,removeparam
), and without$document
the rule is not going to work, unless it is just a normal simple one like||example^
This is a big incompatibility for lists, and while I took the Adguard list and added $document to all the rules and they work fine, updating the lists and making sure to remove the ones already included in Query String Filter so it doesn't cause conflict (crashes) takes time.
Hope this helps to improve it so lists can work without problems or modifying anything.
The text was updated successfully, but these errors were encountered: