Rather than the very C-like API we currently have, accepting a void* and
a length, let's take a Bytes object instead. In almost all existing
cases, the compiler figures out the length.
This function has probably been added when we weren't as good with error
propagations as we are now. We can safely remove it and let future
calls to `read` fail if the file is corrupted.
This can be tested with the following bytes (already used in 9191829a):
ffd8ffc000000800080ef701101200ffda00030100
When trying to set the wallpaper from the menu, ImageViewer would
crash because setting the wallpaper requires the program to pledge
to the WindowManager domain. This patch adds that pledge.
This parses the new background-position-x/y longhands and properly
hooks up them up. This requires converting PositionStyleValue to
just contain two EdgeStyleValues so that it can be easily expanded
into the longhands.
This is very similar to the LittleEndianInputBitStream bit buffer change
from 8e834d4bb2.
We currently buffer one byte of data for the underlying stream. And when
we put bits onto that buffer, we do so 1 bit at a time.
This replaces the u8 buffer with a u64. And instead of looping at all,
we perform bitwise operations to write the desired number of bits.
Using the "enwik8" file as a test (100MB uncompressed, commonly used in
benchmarks: https://www.mattmahoney.net/dc/enwik8.zip), compression time
decreases from:
13.62s to 10.9s on Serenity (cold)
13.62s to 9.22s on Serenity (warm)
2.93s to 2.32s on Linux
One caveat is that this requires explicitly flushing any leftover bits
when the caller is done with the stream. The byte buffer implementation
implicitly flushed its data every time the buffer was byte-aligned, as
doing so would always fill the byte. This is no longer the case. But for
now, this should be fine as the one user of this class, DEFLATE, already
has a "flush everything now that we're done" finalizer.
This is something we're supposed to do according to the HTML spec.
Note that image loading is currently completely ad-hoc, and this just
adds a simple DocumentLoadEventDelayer to the existing implementation.
This will allow us to use images in layout tests, which rely on the
document load event firing at a predictable time.
Instead of passing the continuously merged initial forbidden token set
(with the new additional forbidden tokens from each parsed secondary
expression) to the next call of parse_secondary_expression(), keep a
copy of the original set and use it as the base for parsing the next
secondary expression.
This bug prevented us from properly parsing the following expression:
```js
0 ?? 0 ? 0 : 0 || 0
```
...due to LogicalExpression with LogicalOp::NullishCoalescing returning
both DoubleAmpersand and DoublePipe in its forbidden token set.
The following correct AST is now generated:
Program
(Children)
ExpressionStatement
ConditionalExpression
(Test)
LogicalExpression
NumericLiteral 0
??
NumericLiteral 0
(Consequent)
NumericLiteral 0
(Alternate)
LogicalExpression
NumericLiteral 0
||
NumericLiteral 0
An alternate solution I explored was only merging the original forbidden
token set with the one of the last parsed secondary expression which is
then passed to match_secondary_expression(); however that led to an
incorrect AST (note the alternate expression):
Program
(Children)
ExpressionStatement
LogicalExpression
ConditionalExpression
(Test)
LogicalExpression
NumericLiteral 0
??
NumericLiteral 0
(Consequent)
NumericLiteral 0
(Alternate)
NumericLiteral 0
||
NumericLiteral 0
Truth be told, I don't know enough about the inner workings of the
parser to fully explain the difference. AFAICT this patch has no
unintended side effects in its current form though.
Fixes#18087.
This now uses the current font (rather than the painter's default)
and scales it correctly. This is not perfect though as just naviely
doing .draw_text() here does not follow the proper text layout logic
so this is misaligned (by a pixel or two) with the text in the <li>.
At the end of HTML::EventLoop::process(), the loop reschedules itself if
there are more runnable tasks available.
However, the condition was flawed: we would reschedule if there were any
microtasks queued, but those tasks will not be processed if we're
currently within the scope of a microtask checkpoint.
To fix this, we now only reschedule the HTML event loop for microtask
processing *if* we're not already in a microtask checkpoint.
This fixes the 100% CPU churn seen when looking at PRs on GitHub. :^)
This is copy-pasted from the gzip utility, along with its existing TODO.
This is currently only needed by that utility, but this gives us API
symmetry with GzipDecompressor, and helps ensure we won't end up in a
situation where only one utility receives optimizations that should be
received by all interested parties.
This patch adds an additional control to KeyboardSettings allowing
the user to map Caps Lock to Ctrl. Previously, this was only possible
by writing to /sys/kernel/variables/caps_lock_to_ctrl.
Writing to /sys/kernel/variables/caps_lock_to_ctrl requires root
privileges, but KeyboardSettings will not attempt to elevate
the privilege of the user if they are not root. Instead, the
checkbox is rendered as un-editable.
As part of this, the CodeDocument now keeps track of the kind of
difference for each line. Previously, we iterated every hunk every time
the editor was painted, but now we do that once whenever the diff
changes, and then save the type of difference for each line.
HackStudio's Editor has displayed indicators in its gutter for a long
time, but each required manual code to paint them in the right place
and respond to click events. All indicators on a line would be painted
in the same location. If any other applications wanted to have gutter
indicators, they would also need to manually implement the same code.
This commit adds an API to GUI::TextEditor so it deals with these
indicators. It makes sure that multiple indicators on the same line
each have their own area to paint in, and provides a callback for when
one is clicked.
- `register_gutter_indicator()` should be called early on. It returns a
`GutterIndicatorID` that is then used by the other methods.
Indicators on a line are painted from right to left, in the order
they were registered.
- `add_gutter_indicator()` and `remove_gutter_indicator()` add the
indicator to the given line.
- `clear_gutter_indicators()` removes a given indicator from every line.
- The `on_gutter_click` callback is called whenever the user clicks on
the gutter, but *not* on an indicator.
Warn the user about seemingly known malloc() and free() patterns in the
fault address.
This brings back the functionality that was removed recently in the
5416a37fde commit, but this time we detect
these patterns in userspace code and not in kernel code.
Calling umask and open with same permission value will result in a file
with no permissions bits if the program is stopped while it is doing an
IO. This will result in an error with EACCES when we try to run the
benchmark with the same file name. The file needs to be manually
removed before continuing the benchmark.
There is no use in calling umask here, so just remove it so that
interrupting the program while it is doing an IO will not result int the
file with no permissions the next time we run the program.
Instead of reading bytes from the output stream into a buffer, just to
immediately write them back out, we can skip the middle-man and copy the
bytes directly into the output buffer.
This implements Intel's slicing-by-8 algorithm for CRC checksums (only
little endian CPUs for now, as I don't have a way to test big endian).
The original paper for this algorithm seems to have disappeared, but
Intel's source code is still available as a reference:
https://sourceforge.net/projects/slicing-by-8/
As well as other implementations for reference:
https://docs.rs/slice-by-8/latest/src/slice_by_8/algorithm.rs.html
Using the "enwik8" file as a test (100MB uncompressed, commonly used in
benchmarks: https://www.mattmahoney.net/dc/enwik8.zip), decompression
time decreases from:
4.89s to 3.52s on Serenity (cold)
1.72s to 1.32s on Serenity (warm)
1.06s to 0.92s on Linux
This was fine before as the last entry was a null string (which could be
printed), but we no longer use C-style sentinel-terminated arrays for
arguments.
This solves an awkward dependency cycle, where CalculatedStyleValue
needs the definition of Percentage, but including that would also pull
in PercentageOr, which in turn needs CalculatedStyleValue.
Many places that previously included StyleValue.h no longer need to. :^)
There were a mix of users between those who want to know if the Length
changed, and those that just want an absolute Length. So, we now have
two methods: Length::absolutize() returns an empty Optional if nothing
changed, and Length::absolutized() always returns a value.
...and replace it with AngleOrCalculated.
This has the nice bonus effect of actually handling `calc()` for angles
in a transform function. :^) (Previously we just would have asserted.)
This is intended as a replacement for Length and friends each holding a
`RefPtr<CalculatedStyleValue>`. Instead, let's make the types explicit
about whether they are calculated or not. This then means a Length is
always a Length, and won't require including `StyleValue.h`.
As noted, it's probably nicer for LengthOrCalculated to live in
`Length.h`, but we can't do that until Length stops including
`StyleValue.h`.
Due to CSSImportRule::has_import_result() being backwards, we never
actually entered imported style sheets when traversing style rules or
media queries.
With this fixed, we no longer need the "collect style sheets" step in
StyleComputer, as normal for_each_effective_style_rule() will now
actually find all the rules. :^)
The constructor is now only concerned with creating the required
streams, which means that it no longer fails for XZ streams with
invalid headers. Instead, everything is parsed and validated during the
first read, preparing us for files with multiple streams.
This was causing a huge slowdown when loading some pages with weirdly
huge number of style sheets. For example, amazon.com has over 200 style
elements, which meant we had to resort the StyleSheetList 200 times.
(And sorting itself was slow because it has to compare DOM positions.)
Instead of sorting, we now look for the correct insertion point when
adding new style sheets, and we start the search from the end, which is
where style sheets are typically added in the vast majority of cases.
This removes a 600ms time sink when loading Amazon on my machine! :^)
Setting the `data` of a text node already triggers `children changed`
per spec, so there's no need for an explicit call.
This avoids parsing every HTMLStyleElement sheet twice. :^)
We now select between nearest neighbor and bilinear filtering when
scaling images in CRC2D.drawImage().
This patch also adds CRC2D.imageSmoothingQuality but it's ignored for
now as we don't have a bunch of different quality levels to map it to.
Work towards #17993 (Ruffle Flash Player)
We currently decode back-references one byte at a time, while writing
that byte back out to the output buffer. This is only necessary when the
back-reference refers to itself, i.e. when the back-reference distance
is less than its length. In other cases, we can read the entire back-
reference block in one shot.
Using the "enwik8" file as a test (100MB uncompressed, commonly used in
benchmarks: https://www.mattmahoney.net/dc/enwik8.zip), decompression
time decreases from:
5.8s to 4.89s on Serenity (cold)
2.3s to 1.72s on Serenity (warm)
1.6s to 1.06s on Linux
Huffman codes have a useful property in that they are prefix codes. That
is, a set of bits representing a Huffman-coded symbol is never a prefix
of another symbol. This allows us to create a table, where each index in
the table are integers whose prefix is the entry's corresponding Huffman
code.
With Deflate, we can have codes up to 16 bits in length, thus creating a
prefix table with 2^16 entries. So instead of creating a table fit all
possible codes, we use a cutoff of 8-bit codes. Codes larger than 8 bits
fall back to the binary search method.
Using the "enwik8" file as a test (100MB uncompressed, commonly used in
benchmarks: https://www.mattmahoney.net/dc/enwik8.zip), decompression
time decreases from 3.527s to 2.585s on Linux.
We currently mix normal and bit streams during GZIP decompression, where
the latter is a wrapper around the former. This isn't causing issues now
as the underlying bit stream buffer is a byte, so the normal stream can
pick up where the bit stream left off.
In order to increase the size of that buffer though, the normal stream
will not be able to assume it can resume reading after the bit stream.
The buffer can easily contain more bits than it was meant to read, so
when the normal stream resumes, there may be N bits leftover in the bit
stream that the normal stream was meant to read.
To avoid weird behavior when mixing streams, this changes the GZIP
decompressor to always read from a bit stream.
Previously we used a parsing context with no access to the document, so
any URLs in url() functions would become invalid.
Fixes the images on Steam's store carousel, which sets
Element.style.backgroundImage to url() functions.
Some fonts (like the Bootstrap Icons webfont) have bogus glyph offsets
in the `loca` table that point past the end of the `glyf` table.
AFAICT other rasterizers simply ignore these glyphs and treat them as if
they were missing. So let's do the same.
This makes https://changelog.serenityos.org/ actually work! :^)
In situations where we need a width to calculate the intrinsic height of
a flex item, we use the fit-content width as a stand-in. However, we
also need to clamp it to any min-width and max-width properties present.
The spec tells us that for glyph offsets in the "loca" table, if an
offset appears twice in a row, the index of the first one refers to a
glyph without an outline (such as the space character). We didn't check
for this, which would cause us to render a glyph outline where there
should have been nothing.
Some versions of clang, such as Apple clang-1400.0.29.202 error out on
the previous out of line operators. Explicitly defaulting comparison
operators out of line is allowed per P2085R0, but was checked in clang
before version 15 in C++20 mode.
When debugging in Xcode, the waitpid() for the initial forked process
would always return EINTR or ECHILD. Work around this by blocking all
signals until we're ready to wait for the initial child.
In `flex-direction: column` layouts, a flex item's intrinsic height may
depend on its width, but the width is calculated *after* the intrinsic
height is required.
Unfortunately, the specification doesn't tell us exactly what to do here
(missing inputs to intrinsic sizing is a common problem) so we take the
solution that flexbox applies in 9.2.3.C and apply it to all intrinsic
height calculations within FlexFormattingContext: if the used width of
an item is not yet known when its intrinsic height is requested, we
substitute the fit-content width instead.
Note that while this is technically ad-hoc, it's basically extrapolating
the spec's suggestion in one specific case and using it in all cases.
When calculating the intrinsic width of a box, we now make its content
width & height indefinite before entering the intrinsic sizing layout.
This ensures that any geometry assigned to the box by its parent
formatting context is ignored.
For intrinsic heights, we only make the content height indefinite.
This is because used content width is a valid (but optional) input
to intrinsic height calculation.
This fixes an issue found on Linus's hosted WebServer. Now, if WebServer
is hosted at a non-root URL. (eg, `example.com/webserver` instead of
`example.com`) the links will correctly go to
`example.com/webserver/foo` instead of `example.com/foo`.
When scrolling the page, we may need to flush any pending layout
changes. This is required because otherwise, we don't know if the target
scroll position is valid.
However, we don't need to *force* a layout. If the layout tree is
already up-to-date, we can use it as-is.
Also, if we're scrolling to (0, 0), we don't need to update the layout
at all, since (0, 0) is always a guaranteed valid scroll position.
Currently significant portion of requests coming to WebDriver server
fails with error while parsing json body because requests are parsed
when they are not complete yet.
This change solves this by waiting for more data to arrive if HTTP
request parser found that there is not enough data yet to parse
the whole request.
In the future we would probably want to move this logic to LibHTTP
because this problem is relevant for any HTTP server.
Fixes the issue that XML parser fails when loader passes input that is
prefixed with byte order mark.
Also it generally makes sense to pass text source through encoding
decoder before parsing. Probably we would even want to introduce method
similar to `create_with_uncertain_encoding` in `HTMLParser` but for
`XMLParser` to be make harder unconsciously pass non-UTF8 input to XML
parser.
The "flex item automatic minimum size in the main axis is the
content-based minimum size" behavior should only apply to flex item
sizes in the main axis. There was one case where we incorrectly applied
this behavior in the cross axis
The "flex item automatic minimum size in the main axis is the
content-based minimum size" behavior should only apply to flex items
that aren't scroll containers. We were doing it for all flex items.
When calculating one of the intrinsic sizes for a flex container, we
already go through the flex layout algorithm.
There's no need to perform some of the algorithm steps a second time.
This is a relic from an earlier time when we tried to bail early from
the layout algorithm in the intrinsic sizing case. Now that we go
through the whole thing anyway, this is much simpler. :^)
This fixes a few issues I noticed when playing around with radio
buttons. Previously radio buttons would uncheck checkboxes with
the same "name" attribute, uncheck inputs across different forms,
and treated no name attribute as a group.
This now implements the radio button group check from the HTML spec.
In the case where an error is created from an errno, calling
string_literal() will print nothing. Using Error's formatter
instead gives a more descriptive error message.
Previously, the position of the mask used to calculate the new
selection did not match the position of the active layer. The program
would crash when trying to set a mask pixel outside the bounds of the
active layer.
If the font dictionary didn't specify custom glyph widths, we would fall
back to the specified "missing width" (or 0 in most cases!), which meant
that we would draw glyphs on top of each other in a lot of cases, namely
for TrueTypeFonts or standard Type1Fonts with an OpenType fallback.
What we actually want to do in this case is ask the OpenType font for
the correct width.
A limit of 1024 subroutines seemed like a sensible choice, but some
fonts actually do exceed it. We will now only assert that the specified
amount is positive.
Previously, get_inheritable_object would always try to find the object
and throw an error if it couldn't. The spec tells us that some page
attributes, like CropBox, are optional but also inheritable. Others,
like the media box and resources, are technically required by the spec,
but omitted by some documents.
In both cases, we are now able to search for inheritable objects and
find a suitable replacement if there wasn't one.
Since ef7d9c0, shortcut propagation was blocked for blocking modals and
popups. This however is an issue as some blocking modals (like
FilePicker) use shortcuts. This patch allows propagation of shortcuts
but only until the current window.
First, this patch renames the function
`propagate_shortcuts_up_to_application` to `propagate_shortcuts`.
Handling those levels, will allow us to differentiate shortcuts at
`Window` level and `Application` level. Which will be convenient to
handle dialog-specific shortcuts.
Introduced in 2c98eff, support for non-interleaved scans was not working
for frames with a number of MCU per line or column that is odd. Indeed,
the decoder assumed that they have scans that include a fabricated MCU
like scans with multiple components.
This patch makes the decoder handle images with a number of MCU per line
or column that is odd. To do so, as in the current decoder state we do
not know if components are interleaved at allocation time, we skip over
falsely-created macroblocks when filling them. As stated in 2c98eff,
this is probably not a good solution and a whole refactor will be
welcome.
It also comes with a test that open a square image with a side of 600px,
meaning 75 MCUs.
Negative width/height sizes are not allowed in CSS, so if our layout
algorithm resolves something to a negative size, we have boogs.
Instead of crashing or rendering something very wrong, we now clamp the
values to 0 and whine on the debug log so that someone can go and figure
out how we got the negative values in the first place. :^)
If the previous sibling of an out-of-flow box has been wrapped in an
anonymous block, we now stuff the out-of-flow box into the anonymous
block as well.
Co-authored-by: Aliaksandr Kalenik <kalenik.aliaksandr@gmail.com>
...simply by using LittleEndianInputBitStream::read_bit() instead of
read_bits(1). This puts us on the fast path for single-bit reads.
There's still lots of money on the table for bigger optimizations to
claim here, just picking an embarrassingly low-hanging fruit. :^)
Instead of going byte by byte, copy entire blocks at once and only check
if we need to update the state once per block. This pretty much
eliminates `::update()` from profiles and measurably improves
performance for utilities like `sha256sum`.
This will allow these to be more easily reused for other inputs.
Note that we need this as the base LibGfx pallet does not quite have
the required range for HTML inputs (though these shades are based off
the LibGfx palette).
The `read_image_data` function of each one of[PBM, PGM, PPM]Loader use
the same structure to read an image. This patch harmonizes the three
functions and use finite loops instead of reading until EOF. It allows
to quit early on bloated file, but it's mainly done for refactoring
purpose.
The function signature goes from:
`bool read_number(Streamer& streamer, TValue* value)`
to
`ErrorOr<u16> read_number(Streamer& streamer)`
It allows us to, on one hand use `ErrorOr` for error propagation,
removing an out parameter in the meantime, and on the other hand remove
the useless template.
