Related issue:
- https://github.com/uBlockOrigin/uBlock-issues/issues/780
- Handle DNS lookup failure
- Skip DNS lookup for non network-based URLs
- Benchmark code to be able to provide an estimate
based on objective measurements regarding added
overhead when DNS lookup is enabled
(quick answer: a complete non-issue)
Related issue:
- https://github.com/uBlockOrigin/uBlock-issues/issues/780
New webext permission added: `dns`, which purpose is
to allow an extension to fetch the DNS record of
specific hostnames, reference documentation:
https://developer.mozilla.org/en-US/docs/Mozilla/Add-ons/WebExtensions/API/dns
The webext API `dns` is available in Firefox 60+ only.
The new API will enable uBO to "uncloak" the actual
hostname used in network requests. The ability is
currently disabled by default for now -- this is only
a first commit related to the above issue to allow
advanced users to immediately use the new ability.
Four advanced settings have been created to control the
uncloaking of actual hostnames:
cnameAliasList: a space-separated list of hostnames.
Default value: unset => empty list.
Special value: * => all hostnames.
A space-separated list of hostnames => this tells uBO
to "uncloak" the hostnames in the list will.
cnameIgnoreList: a space-separated list of hostnames.
Default value: unset => empty list.
Special value: * => all hostnames.
A space-separated list of hostnames => this tells uBO
to NOT re-run the network request through uBO's
filtering engine with the CNAME hostname. This is
useful to exclude commonly used actual hostnames
from being re-run through uBO's filtering engine, so
as to avoid pointless overhead.
cnameIgnore1stParty: boolean.
Default value: true.
Whether uBO should ignore to re-run a network request
through the filtering engine when the CNAME hostname
is 1st-party to the alias hostname.
cnameMaxTTL: number of minutes.
Default value: 120.
This tells uBO to clear its CNAME cache after the
specified time. For efficiency purpose, uBO will
cache alias=>CNAME associations for reuse so as
to reduce calls to `browser.dns.resolve`. All the
associations will be cleared after the specified time
to ensure the map does not grow too large and too
ensure uBO uses up to date CNAME information.
... from platform-independent code.
This should complete the de-chromiumification of
uBO. Next step will be to swap the content of the
`platform/chromium` and `platform/firefox`
folders so as to reflect that Firefox is the
natural platform for uBO.
Related issue:
- https://github.com/uBlockOrigin/uBlock-issues/issues/710
Messages from unprivileged ports (i.e. from content scripts)
are no longer relayed to message handlers which are to be
strictly used to execute privileged code.
The last remaining case of unprivileged messages which
should be converted into a privileged ones will be taken
care of when the following issue is fixed:
- https://github.com/gorhill/uBlock/issues/3497
The motivation is to address the higher peak memory usage at launch
time with 3rd-gen HNTrie when a selfie was present.
The selfie generation prior to this change was to collect all
filtering data into a single data structure, and then to serialize
that whole structure at once into storage (using JSON.stringify).
However, HNTrie serialization requires that a large UintArray32 be
converted into a plain JS array, which itslef would be indirectly
converted into a JSON string. This was the main reason why peak
memory usage would be higher at launch from selfie, since the JSON
string would need to be wholly unserialized into JS objects, which
themselves would need to be converted into more specialized data
structures (like that Uint32Array one).
The solution to lower peak memory usage at launch is to refactor
selfie generation to allow a more piecemeal approach: each filtering
component is given the ability to serialize itself rather than to be
forced to be embedded in the master selfie. With this approach, the
HNTrie buffer can now serialize to its own storage by converting the
buffer data directly into a string which can be directly sent to
storage. This avoiding expensive intermediate steps such as
converting into a JS array and then to a JSON string.
As part of the refactoring, there was also opportunistic code
upgrade to ES6 and Promise (eventually all of uBO's code will be
proper ES6).
Additionally, the polyfill to bring getBytesInUse() to Firefox has
been revisited to replace the rather expensive previous
implementation with an implementation with virtually no overhead.
The environment flavor is used to by uBO to for self-configuration.
For users with spoofed UA at the `about:config` level, this might
cause uBO to misconfigure itself. Avoid UA and strictly rely on
browserInfo() for looking up environment parameters.
The purpose of this API is basically to satisfy AMO reviewers in the
future, since the use of innerHTML with variables (i.e., not plain text) will
be rejected without any questions.
Since this is not a problem for browsers other than Firefox, they will
use simple innerHTML assignment, however safe-parsing could be implemented
for them too.
