Related issue:
https://github.com/uBlockOrigin/uBlock-issues/issues/2773
The `randomize` paramater introduced in https://github.com/gorhill/uBlock/commit/418087de9c
is now named `directive`, and beside the `true` value which is meant
to respond with a random 10-character string, it can now take the
following value:
war:[web_accessible_resource name]
In order to mock the XHR response with a web accessible resource. For
example:
piquark6046.github.io##+js(no-xhr-if, adsbygoogle.js, war:googlesyndication_adsbygoogle.js)
Will cause the XHR performed by the webpage to resolve to the content
of `/web_accessible_resources/googlesyndication_adsbygoogle.js`.
Should the resource not exist, the empty string will be returned.
Now support AdGuard's `randomize` parameter. If `true`,
the scriplet will generate a random 10-character string
to be returned as the response.
Reference:
https://github.com/AdguardTeam/Scriptlets/blob/master/wiki/about-scriptlets.md#prevent-xhr
AdGuard's `prevent-xhr` also support `length:n-m` form,
but since I do not see it being used, for now it's not
supported in uBO's `no-xhr-if`.
Additionally, the scriptlet will now honor `responseType` and
return the proper response type accordingly.
`fetchPropsToMatch` is an optional variable argument. If provided,
the scriplet will take effect only when the JSON data is obtained
through `Response.json()` and if there is a match with the value of
`fetchPropsToMatch` and the properties of the `Response` instance.
Examples of usage:
...##+js(json-prune, ads, , , fetchPropsToMatch, ?param=)
...##+js(json-prune, ads, , , fetchPropsToMatch, url:?param= method:get)
The optional variable argument `fetchPropsToMatch` acts as an additional
narrowing condition to fulfill before the JSON data is pruned.
Specifically, the variable argument `log` can take one of three
values to enable logging mechanism:
..., log, match => log only when there is a match
..., log, nomatch => log only when there is no match
..., log, all => log unconditionally
Prepend pattern with `!` to test for unmatched patterns in
stack trace. This applies to sctiplet parameters which purpose
is to test against the stack, i.e. `aost` and `json-prune`.
Additionally, dropped support for JSON notation in favor of
optional variable arguments notation.
Related discussion:
- https://github.com/uBlockOrigin/uBlock-discussions/discussions/789#discussioncomment-6520330
This extends logging capabilities of `json-prune` scriptlet as
follow:
...##+js(json-prune, a, b, stackNeedle, log, [logneedle], logstack, 1)
Whereas before, the only way to log `json-prune` usage was to skip
providing the property chain:
...##+js(json-prune, , b)
Where `b` was the expression to filter out logging output.
With the extended logging capabilities, the logging output can
be filtered out with `logneedle`, which can be a regex literal.
Additionally, to log the stack trace the `stackNeedle` argument
must be set to non-empty string. You can use `/.^/` to log the
stack trace without matching it.
Related issue:
- https://github.com/AdguardTeam/Scriptlets/issues/332
Additionally, uBO's own scriplet syntax now also accept quoting
the parameters with either `'` or `"`. This can be used to avoid
having to escape commas when they are present in a parameter.
Related issue:
- https://github.com/uBlockOrigin/uAssets/issues/18725
Testing the context was causing the deprecated static property
RegExp.$1 to be clobbered, causing webpage breakage because this
property was subsequently used used by the caller.
New official name: `no-window-open-if`.
The pattern will now be matched against all arguments passed
to `window.open()`: all the arguments are joined as a single
space-spearated string, and the result is used as the target
for matching the pattern.
To enable logging, used the extra parameters approach, i.e.
`log, 1`, which should come after the positional arguments
`pattern`, `delay`, and `decoy`.
Added support for extra parameter `reload, 1`. If present,
the scriplet will force a reload of the webpage if the cookie
being set was not already set.
As per feedback from filter list maintainers.
The scriptlet remove the *whole* text of a DOM node. Usage:
example.com##+js(remote-node-text, nodeName, condition, ...)
Where `condition` is a pattern to find in the target node for the
removal to occur. Since the text of the node is wholly removed,
this is not a scriplet which requires a trusted-source, as it is
virtually a similar capability as removing DOM nodes through
procedural cosmetic operator `:remove()` (which can't target
`#text` nodes), which does not require trusted-source.
Alias: `rmnt`
The extra parameters are the same as those documented for
`replace-node-text`.
For consistency, the alias for `replace-node-text` has been renamed
`rpnt`.
When scriptlets can receive extra optional paramaters, these will
now be passed as pair of extra paramaters in the filter declaration,
whereas each pair is a `name, value` instance.
As a result, the optional paramaters that can be passed to the
`aeld` scriptlet can be passed this way, i.e. no longer need
a JSON approach, example:
github.com##+js(aeld, click, , log, 1)
github.com##+js(aeld, , , runAt, idle, log, 1)
The non-optional paramaters are always positional, after which
the optional paramaters are non-positional pairs of values.
Alias: `trusted-set`
Behaves exactly like set-constant, except that any arbitrary JSON-
compatible value can be set.
By default the value is treated as a string, which can be anything.
If the value starts with `{` and ends with `}`, the value will be
JSON-parsed, and the `value` property of the resulting object will
be used.
As with any scriptlet requiring trust, filters using
`trusted-set-constant` can only come from trusted filter lists,
otherwise they are discarded.
Related discussion:
- https://github.com/uBlockOrigin/uAssets/discussions/18185#discussioncomment-5977456
This reflects the _world_ of the MV3 scripting API:
https://developer.mozilla.org/en-US/docs/Mozilla/Add-ons/WebExtensions/API/scripting/ExecutionWorld
MAIN: page's world
ISOLATED: extension's content script world
Some scriptlets are best executed in either world, so this
commit allows to pick in which world a scriptlet should execute
(default to MAIN).
For instance, the new sed.js scriptlet will now execute in
the ISOLATED world.
Related commit:
- 41876336db
The `tryCount` parameter has been removed.
The new default behavior of the sed.js scriptlet is to bail out
when the document itself has been fully loaded, i.e. when
DOMContentLoaded event is fired.
Two new parameters have been added to override the default quit out
behavior:
`stay, 1`
Use to force the scriptlet to stay at work forever.
`quitAfter, ms`
This tells the scriptlet to quit `ms` milliseconds after the
page has been loaded, i.e. after the DOMContentLoaded event has
been fired.
The mutation observer of the sed.js scriptlet can be a significant
overhead for pages with dynamically updated DOM, and in most cases
the scriptlet is useful only for DOM changes occurring before the
DOMContentLoaded event, so the default is to quit out when that
event is received ("quit out" means discarding the mutation observer
and having the scriptlet garbage-collected by the JS engine).
To ensure smooth transition in next stable release: people on
an older version of uBO could end up updating assets.json,
hence we needs to keep a local URL to the non-minified packaged
version of the lists.
At the moment, the only filter lists deemed from a "trusted source"
are uBO-specific filter lists (i.e. "uBlock filters -- ..."), and
the user's own filters from "My filters".
A new scriptlet which can only be used by filter lists from trusted
sources has been introduced: `sed.js`.
The new `sed.js` scriptlet provides the ability to perform
text-level substitutions. Usage:
example.org##+js(sed, nodeName, pattern, replacement, ...)
`nodeName`
The name of the node for which the text content must be substituted.
Valid node names can be found at:
https://developer.mozilla.org/en-US/docs/Web/API/Node/nodeName
`pattern`
A string or regex to find in the text content of the node as the target of
substitution.
`replacement`
The replacement text. Can be omitted if the goal is to delete the text which
matches the pattern. Cannot be omitted if extra pairs of parameters have to be
used (see below).
Optionally, extra pairs of parameters to modify the behavior of the scriptlet:
`condition, pattern`
A string or regex which must be found in the text content of the node
in order for the substitution to occur.
`sedCount, n`
This will cause the scriptlet to stop after n instances of substitution. Since
a mutation oberver is used by the scriptlet, it's advised to stop it whenever
it becomes pointless. Default to zero, which means the scriptlet never stops.
`tryCount, n`
This will cause the scriptlet to stop after n instances of mutation observer
run (regardless of whether a substitution occurred). Default to zero, which
means the scriptlet never stops.
`log, 1`
This will cause the scriptlet to output information at the console, useful as
a debugging tool for filter authors. The logging ability is supported only
in the dev build of uBO.
Examples of usage:
example.com##+js(sed, script, /devtoolsDetector\.launch\(\)\;/, , sedCount, 1)
example.com##+js(sed, #text, /^Advertisement$/)
Related issue:
- https://github.com/uBlockOrigin/uBlock-issues/issues/2618
Usage:
example.com##+js(spoof-css, selector, property-name, property-value, ...)
- selector: a valid CSS selector which matches the elements for which
the spoofing must apply.
- property-name: a CSS property name (can be dashed- or camel-cased)
- property-value: the value to return regardless of the currently
computed value.
There can be any number of property-name/property-value pairs, all
separated by commas.
A special property-name/property-value pair `debug/1` can be used
to force the browser to break when `getComputedStyle()` or
`getBoundingClientrect()` is called, useful to help pinpoint usage
of those calls in the page's source code:
example.com##+js(spoof-css, .ad, debug, 1)
Related issue:
- https://github.com/uBlockOrigin/uBlock-issues/issues/2615
Expand `set-constant`: 3rd parameter and beyond are tokens which
modify the behavior of `set-contant`. Valid tokens:
- `interactive`, `end`, `2`: set the constant when the event
`DOMContentInteractive` is fired.
- `complete`, `idle`, `3`: set the constant when the event
`load` is fired.
- `asFunction`: the constant will be a function returning the
specified value.
- `asCallback`: the constant will be a function returning a
function returning the specified value.
- `asResolved`: the constant will be a promise resolving to
the specified value.
- `asRejected`: the constant will be a promise failing with
the specified value.
Harden `no-setimeout-if` and `no-setinterval-if` as per feedback
from filter list maintainers.
As discussed internally with filter list maintainers.
Additionally, added a search field to filter out lists. This
is still a work in progress, no need to open issues about this,
I am aware of what is missing (i18n, more tags, etc.)
To be replaced with five sublists.
Users who are already subscribed to the list will have to manually
unsubscribe from it, which will be moved to _Custom_ section, and
subscribe to the sublists which have narrower purpose.
As per discussion with filter list maintainers.
THis requires to use JSON notation for parameter passing:
- "runAt": "end" = execute scriptlet at `DOMContentLoaded` event
- "runAt": "idle" = execute scriptlet at `load` event
A new optional parameter has been added to `set-constant`
scriptlet: `runAt`, default to `0`.
..##+js(set, document.body.oncontextmenu, null, 2)
When the `runAt` parameter is present, uBO will take it into
account to possibly defer execution of `set-constant`:
- `runAt` not present: execute immediately
- `runAt` = 1: execute immediately
- `runAt` = 2: execute when document state is "interactive"
- `runAt` = 3: execute when document state is `"complete"
maintainers.
To enable logging, use the JSON approach to pass parameters to
`acs` scriptlet. Example:
..##+js(acs, { "target": "document.oncontextmenu", "log": true })
Whereas "target", "needle", and "context" correspond to their
respective positional argument. Using JSON form to pass parameters
allows to specify extra paramters to facilitate debugging of that
scriptlet:
- `"log": true` => output useful information at the dev console.
- `"debug": true` => break at key locations in the scriptlet.
The added logging/debugging capabilities work only in the dev build
of uBO or if the advanced setting `filterAuthorMode` is set to
`true`.
The scriptlet addEventListenerLogger has been removed.
The logging of addEventListener() calls can now be done with the
addEventListenerDefuser scriptlet, which now supports the
following named arguments:
"type": the event type to match. Default to '', i.e. match-all.
"pattern": the pattern to match against the handler argument
Default to '', i.e. match-all.
"log": an integer value telling when to log:
- 1: log only when both type and pattern matches, i.e. when a
call to addEventListener() is defused
- 2: log when either the type or pattern matches
- 3: log all calls to addEventListener()
"debug": an integer value telling when to break into the
debugger, useful to inspect the debugger's call stack.
- 1: break into the debugger when both type and pattern match,
so effectively when defusing is taking place.
- 2: break into the debugger when either type or pattern matches.
The usage of named arguments is optional, positional arguments
are still supported as documented. Named arguments is required
to use "log" and/or "debug" arguments.
Obviously, do not use "log" or "debug" in any filter list, these
are investigative tools for filter list authors.
Examples of usage using named arguments:
wikipedia.org##+js(aeld, { "type": "/mouse/", "pattern": "/.^/", "log": 2 })
Above filter will log calls to addEventListener() which have the
pattern "mouse" in the event type (so "mouseover", "mouseout",
etc.) without defusing any of them (because pattern can't match
anything).
wikipedia.org##+js(aeld, { "type": "/.^/", "log": 2 })
Above filter will log all calls without defusing any of them
(because type can't match anything)
wikipedia.org##+js(aeld, { "log": 1 })
Above filter will log and defuse all calls to addEventListener().
Builtin scriptlets are no longer parsed as text-based resources,
they are imported as JS functions, and `toString()` is used to
obtain text-based representation of a scriptlet.
Scriptlet parameters are now passed as function call arguments
rather than by replacing text-based occurrences of `{{i}}`. The
arguments are always string values (see below for exception).
Support for argument as Object has been added. This opens the
door to have scriptlets using named arguments rather than
positional arguments, and hence easier to extend functionality
of existing scriptlets. Example:
example.com##+js(scriplet, { "prop": "adblock", "value": false, "log": true })
Compatibility with user-provided scriptlets has been preserved.
User-provided scriptlets can benefit some of the changes:
Use the form `function(..){..}` instead of `(function(..){..})();`
in order to received scriptlet arguments as part of function call
-- instead of using `{{i}}`.
If using the form `function(..){..}`, you can choose to receive
an Object as argument -- just be sure that your scriptlet's
parameter is valid JSON notation.
The purpose is to prevent a call to an existing function from
throwing an exception. The exception will be caught by the
scriptlet and neutralized.
The first argument must be a reference to a function call. At
the moment, the function call must exist at the time the
scriptlet is called.
Related issue:
- https://github.com/uBlockOrigin/uBlock-issues/issues/2508
If the first argument is a regex with multine flag set, the
scriptlet will execute the regex against the whole text, and
remove matching text from the whole text.
If the matching text does not contains whole lines, the text
won't be removed, i.e. it is not allowed to remove only part
of a line.
Related issue:
- https://github.com/uBlockOrigin/uBlock-issues/issues/2531#issuecomment-1462389539
Usage:
- example.com##+js(href-sanitizer, a[href^="/go?to="]:not([title]))
- example.com##+js(href-sanitizer, a[href^="/go?to="][title], [title])
The second argument is the attribute from which to extract the text
to be used for the `href` attribute of the link. If the second
attribute is absent, the text content of the element will be used.
Related issue:
- https://github.com/uBlockOrigin/uBlock-issues/issues/2531
Usage:
example.com##+js(href-from-text, a[href^="/tracker-link?to="]
The above scriptlet will find all elements matching the selector
passed as 1st argument, and replace the `href` attribute with the
text content of the element, if all the following conditions are
met:
- The element is a link (`a`) element
- The link element has an existing `href` attribute
- The text content of the element is a valid `https`-based URL
Related issue:
- https://github.com/uBlockOrigin/uBlock-issues/issues/2358
We need two different branches of uAssets to properly add the
default filter lists into the packages. I couldn't figure out
how to do this with submodules, using simple `git clone` as a
dependency just work.
Related issue:
- https://github.com/uBlockOrigin/uAssets/issues/14849
Arguments:
1. Required. The selector of elements which are to be removed.
Example: Period[id*="-roll-"][id*="-ad-"]
2. An optional selector that must have a match in the document
for the pruning to occur. No selector means the pruning can
be performed regardless.
3. An optional URL which must be a match for the pruning to
occur. If left blank, the pruning can be performed regardless.
Related discussion:
- https://github.com/uBlockOrigin/uBlock-issues/discussions/2270
If the argument to the window-close-if scriptlet is a regex, the
match will be against the whole location URL, otherwise the
match will be against the part+query part of the location URL.
If you click the Home button for one of the EasyList filter lists in uBlock settings, now you will go to a page that succinctly describes what the filter does. The old links took you to a forum with no clear way to find info on a given filter, especially for a new user.
This scriplet supersedes abort-current-inline-script (acis),
and accepts an optional third argument which is matched
against the `src` property of script resources.
When the third argument is not provided, the scriptlet
behaves essentially the same as `acis`, and because of
this `acis` is now aliased to `abort-current-script`, and
all existing `acis` filters will execute with no change
in behavior.
In the long run, usage of `abort-current-inline-script` or
its alias `acis` should go away and be replaced with
`abort-current-script` or its alias `acs`.
Some sublists fail to load with this CDN, and
uBO fails to properly error out on the whole
list in such case, causing a truncated whole
list to be used in the end.
Related issue:
- https://github.com/uBlockOrigin/uBlock-issues/issues/1553
This commit ensures FLoC is opt-in. The generic filter
`*##+js(no-floc)` in "uBlock filters -- Privacy" ensures
the feature is disabled when using default settings/lists.
Users can opt-in to FLoC by adding a generic exception
filter to their custom filters, `#@#+js(no-floc)`; or they
can opt-in only for a specific set of websites through a
more specific exception filter:
example.com,shopping.example#@#+js(no-floc)
Too likely to cause breakage. It was originally added without
having been really evaluated. As per feedback, it's too likely
to cause breakage.
The list is seemingly used to evaluate filters, which are moved
to EasyPrivacy once they are deemed valid and useful.
As discussed internally with list maintainers:
- EasyList China doesn't leverage uBO's extended
filter syntax
- EasyList China's home page is no longer available
to the public
Additionally, "CJX's EasyList Lite" has been removed
from stock lists, as the list hosted on GitHub is no
longer updated, and "AdGuard Chinese" is the official
list to enable to address all filter issues for
Chinese sites.
Related issue:
- https://github.com/uBlockOrigin/uBlock-issues/issues/1445
A third (optional) argument has been added to `remove-attr`
scriptlet, which can be one or more space-separated tokens
dictating the behavior of the scriptlet:
`stay`: This tells the scriplet to stay and act on DOM
changes, whiĺe the default behavior is to act only once
when the document becomes interactive.
`complete`: This tells the scriplet to start acting only
when the document is complete, i.e. once all secondary
resources have been loaded, while the default is to start
acting when the document is interactive -- which is earlier
than when the document is complete.
Example:
...##+js(remove-attr, class, .j-mini-player, stay)
When no-fetch-if scriptlet is used without argument, the
parameters passed to no-fetch-if will be output to the
console, as `uBO: fetch([...list of arguments...])`.
The new scriptlet allows to defuse calls to fetch() by returning
a promise which always resolve to an empty response.
There is only one argument, which is a space-separated list
of conditions which must be ALL fulfilled in order for the
defusing to take place.
Each condition is a pair of property name and property value
separated by a column. Valid property names are those
documented as valid `init` options:
https://developer.mozilla.org/en-US/docs/Web/API/WindowOrWorkerGlobalScope/fetch
The URL of the fetch() is a special case and does not have to
be associated with a property name. Example of usage:
...##+js(no-fetch-if, method:HEAD)
Which means: defuse the call to fetch() if there is an
explicit option which contains `HEAD`. Another example:
...##+js(no-fetch-if, adsbygoogle.js)
Which means: defuse the call to fetch() if the URL contains
`adsbygoogle.js`. Multiple conditions can be provided:
...##+js(no-fetch-if, adsbygoogle.js method:HEAD)
If at least one condition does not match, the defusing will
not take place.
The string against which to match can be a literal regular
expression:
...##+js(no-fetch-if, /adsbygoogle.js$/ method:/HEAD|POST/)
Additonally, the following deprecated scriplets have been
removed:
- requestAnimationFrame-if.js
- setInterval-defuser.js
- setTimeout-logger.js
