This *also* got missed in the gcc-12 update, because we weren't
installing an explicit gcc version prior. Hopefully that's the last of
the long tail of issues from that migration!
The Demuxer class was changed to return errors for more functions so
that all of the underlying reading can be done lazily. Other than that,
the demuxer interface is unchanged, and only the underlying reader was
modified.
The MatroskaDocument class is no more, and MatroskaReader's getter
functions replace it. Every MatroskaReader getter beyond the Segment
element's position is parsed lazily from the file as needed. This means
that all getter functions can return DecoderErrors which must be
handled by callers.
As new demuxers are added, this will get quite full of files, so it'll
be good to have a separate folder for these.
To avoid too many chained namespaces, the Containers subdirectory is
not also a namespace, but the Matroska folder is for the sake of
separating the multiple classes for parsed information entering the
Video namespace.
Now we attempt to look for the path of e2fsck before checking if the
path can be found in any of the predefined routes. This fixes e2fsck
not being found on some "special" distros like NixOS.
Related #13754
This adds command line flags for WebDriver to pass its IPC socket path
(if running on Serenity) or its FD passing socket (if running elsewhere)
for the headless-browser to connect to.
Hand-picking the smallest index type that fits a particular generated
array started with commit 3ad159537e. This
was to reduce the size of the generated library.
Since then, the number of types using UniqueStorage has grown a ton,
creating a long list of types for which index types are manually picked.
When a new UCD/CLDR/TZDB is released, and the current index type no
longer fits the generated data, we fail to generate. Tracking down which
index caused the failure is a pretty annoying process.
Instead, we can just use size_t while in the generators themselves, then
automatically pick the size needed for the generated code.
When an IPC message returns a single value, we generate a class with a
constructor that is something like:
class MessageResponse {
MessageResponse(SingleReturnType value)
: m_value(move(value))
{
}
};
If that IPC message wants to return a value that SingleReturnType is
constructible from, you have to wrap that return call with braces:
return { value_that_could_construct_single_return_type };
That isn't really an issue except for when we want to mix TRY semantics
with the return type. If SingleReturnType is constructible from an Error
type (i.e. something similar to ErrorOr), the following doesn't work:
TRY(fallible_function());
Because MessageResponse would not be constructible from Error. Instead,
we must do some workaround with a custom TRY macro, as in 31bb792.
This patch generates a constructor that makes TRY usable as-is without
any custom macros. We perform a very similar trick in ThrowCompletionOr
inside LibJS. This constructor will allow you to create MessageResponse
from any type that SingleReturnType is constructible from.
Previously each emoji had its own symbol in the library which was then
referred to by another symbol. This caused thousands of avoidable data
relocations at load time.
This saves about 122kB RAM for each process which uses LibUnicode.
Previously the s_decomposition_mappings variable would refer to other
data in s_decomposition_mappings_data. This would cause thousands of
avoidable relocations at load time.
This saves about 128kB RAM for each process which uses LibUnicode.
Previously we'd fail to execute the resize2fs tool which then results
in us recreating the image from scratch:
resizing disk image...
Image resized.
line 132: /usr/sbin/resize2fs: No such file or directory
failed, not using existing image
done
Otherwise, we end up propagating those dependencies into targets that
link against that library, which creates unnecessary link-time
dependencies.
Also included are changes to readd now missing dependencies to tools
that actually need them.
The shared parts are now firmly compiled into LibC instead of being
defined as a static library and then being copied over manually.
The non-shared ("local") parts are kept as a static library that is
linked into each binary on demand.
This finally allows us to support linking with the -fstack-protector
flag, which now replaces the `ssp` target being linked into each binary
accidentally via CMake.
Even though the toolchain implicitly links against -lc, it does not know
where it should get LibC from except for the sysroot. In the case of
Clang this causes it to pick up the LibC stub instead, which might be
slightly outdated and feature missing symbols.
This is currently not an issue that manifests because we pass through
the dependency on LibC and other libraries by accident, which causes
CMake to link against the LibC target (instead of just the library),
and thus points the linker at the build output directory.
Since we are looking to fix that in the upcoming commits, let's make
sure that everything will still be able to find the proper LibC first.
Currently, if the script fails, it simply runs "exit 1". This exits the
script, but keeps the VM running, so CI hangs until it times out.
Instead of exiting, write a failure status to an error log and shutdown.
CI can then read that error log and fail the run if needed.
This file will be the basis for abstracting away the out-of-thread or
later out-of-process decoding from applications displaying videos. For
now, the demuxer is hardcoded to be MatroskaParser, since that is all
we support so far. The demuxer should later be selected based on the
file header.
The playback and decoding are currently all done on one thread using
timers. The design of the code is such that adding threading should
be trivial, at least based on an earlier version of the code. For now,
though, it's better that this runs in one thread, as the multithreaded
approach causes the Video Player to lock up permanently after a few
frames are decoded.
The class is virtual and has one subclass, SubsampledYUVFrame, which
is used by the VP9 decoder to return a single frame. The
output_to_bitmap(Bitmap&) function can be used to set pixels on an
existing bitmap of the correct size to the RGB values that
should be displayed. The to_bitmap() function will allocate a new bitmap
and fill it using output_to_bitmap.
This new class also implements bilinear scaling of the subsampled U and
V planes so that subsampled videos' colors will appear smoother.
We currently have two build-time parsers for the UCD's emoji-test.txt
file. To prepare for future changes, this removes the Bash parser and
moves its functionality to the newer C++ parser.
So far we've gotten away with using GCC 11 for Lagom and to compile the
toolchain, but via #15795 we discovered a compiler bug that has been
fixed in the latest version but would error the build with CI's GCC 11.
Time for an upgrade :^)
We already use ubuntu-22.04 images in most places, so this is pretty
straightforward. The only exception is Idan's self-hosted runner, which
uses Ubuntu Focal. LibJS should build fine with GCC 11, still.
There were some notable changes to the CLDR JSON format and data in this
release.
The patterns for a date at a specific time, i.e. "{date} at {time}", now
appear under the "atTime" attribute of the "dateTimeFormats" object.
Locale specific changes that affected test-js:
All locales:
* In many patterns, the code points U+00A0 (NO-BREAK SPACE) and U+202F
(NARROW NO-BREAK SPACE) are now used in place of an ASCII space. For
example, before the "dayPeriod" fields AM and PM.
* Separators such as U+2013 (EN DASH) are now surrounded by U+2009 (THIN
SPACE) in place of an ASCII space character.
Locale "en":
* Narrow localizations of time formats are even more narrow. For
example, the abbreviation "wk." for "week" is now just "wk".
Locale "ar":
* The code point U+060C (ARABIC COMMA) is now used in place of an ASCII
comma.
* The code point U+200F (RIGHT-TO-LEFT MARK) now appears at the
beginning of many localizations.
* When the "latn" numbering system is used for currency formatting, the
currency symbol more consistently is placed at the end of the pattern.
Locale "he":
* The "many" plural rules category has been removed.
Locales "zh" and "es-419":
* Several display-name localizations were changed.
This ensures that the toolchain building scripts will use the host
toolchain that has been found manually, and that they won't fall back to
`cc` (which in the worst case may not even be installed).
Homebrew does not add upstream LLVM's install location to $PATH so as
not to conflict with XCode tools, so we need to run `brew --prefix llvm`
to figure out its install path.
Due to the way we lazily construct prototypes and constructors for web
platform interfaces, it's possible for nested GC allocation to occur
while GC objects have been allocated but not fully constructed.
If the garbage collector ends up running in this state, it may attempt
to call JS::Cell::visit_edges() on an object whose vtable pointer hasn't
been set up yet.
This patch works around the issue by deferring GC while intrinsics are
being brought up. Furthermore, we also create a dummy global object for
the internal realm, and populate it with intrinsics. This works around
the same issue happening when allocating something (like the default UA
stylesheets) in the internal realm.
These solutions are pretty hacky and sad, so I've left FIXMEs about
finding a nicer way.
