This is a manual but clean revert of all commits from #12595.
Adding a partial implementation of the resizable ArrayBuffer proposal
without implementing all the updates to TypedArray infrastructure that
is also covered by the spec introduced a bunch of crashes, so we
decided to revert it for now until a full implementation is completed.
This commit has no behavior changes.
In particular, this does not fix any of the wrong uses of the previous
default parameter (which used to be 'false', meaning "only replace the
first occurence in the string"). It simply replaces the default uses by
String::replace(..., ReplaceMode::FirstOnly), leaving them incorrect.
Not being able to map the file chunk-by-chunk feels like a deficit of
MappedFile and/or LibArchive, so it's weird that `unzip` is enforcing
this size limit, and an especially arbitary one at that.
Since "replace one error message with another error message" is the best
possible outcome here, and making the user pass a useless flag in cases
where it may not even be needed is the worst, let's just remove that
file size limit.
However, the `FIXME` about mapping files partially is left in because
this is something that we definitely want to take a look at in the
future.
This allows us to skip storing the actual backing data in a separate
`Vector`, as we know that we are working with `environ`-backed storage
here.
Also, while the logic is currently very similar to what `getenv` does
internally, this allows us to eventually implement custom environment
variable filters while remaining linear in run time.
To achieve this goal:
- The Browser unveils "/tmp/portal/filesystemaccess"
- Pass the page through LoadRequest => ResourceLoader
- ResourceLoader requests a file to the FileSystemAccessServer via IPC
- OutOfProcessWebView handles it and sends a file descriptor back to
the Page.
Previously keymap did switching of the selected keymap twice when the
command was executed. First set it to the first keymap on the list and
later, if present, to the chosen one.
Currently the switching to the first keymap on the list is done only
when the selection is not present or it's not on the list of keymaps.
This ensures that js's error printing logic is used instead of the
generic value printing logic, which then lets eshost correctly parse
thrown SyntaxErrors using the normal LibJS exception format.
With this, the headless browser can now connect to the web.
Thanks a lot to Ali and Sin-ack for their help with this!
Co-authored-by: Ali Mohammad Pur <mpfard@serenityos.org>
Co-authored-by: sin-ack <sin-ack@users.noreply.github.com>
Previously the netstat utility crashed when LookupServer wasn't running
because it tried to unveil nonexistent /tmp/portal/lookup socket. This
commit fixes that.
Previously, `touch` remained limited to creating files and updating
their current access and modifications time to the current time. It's
now capable of accepting two different timestamp formats with flags `-d`
and `-t` and referencing timestamps of other files with flag `-r`.
`touch` can also update only the last access time with `-a`, only the
last access time with `-m`, or update both as usual. `touch` updates
both left unspecified.
With `-c`, `touch` does not create a file if it doesn't already exist.
Using a Vector<Value> is unsafe as GC cannot see the stored values.
This is then vended to outside users of ConsoleClient, e.g. LibWeb and
WebContent, which is then outside of LibJS's control.
An example issue is if the client stores it for later use and forgets
to visit the stored values, meaning they can be destroyed at any time.
We can save the client from this by vending a MarkedVector<Value> to
them.
Instead of just printing 'ECMAScriptFunctionObject' (and leaking an
implementation detail in the process - this is not a public facing name)
let's instead print a different type string for each function kind, and
only keep the old class_name() printing for other JS::FunctionObject
subclasses.
The file is now renamed to Queue.h, and the Resampler APIs with
LegacyBuffer are also removed. These changes look large because nobody
actually needs Buffer.h (or Queue.h). It was mostly transitive
dependencies on the massive list of includes in that header, which are
now almost all gone. Instead, we include common things like Sample.h
directly, which should give faster compile times as very few files
actually need Queue.h.
