When a width/height constraint is applied to GFC it should set its own
width/height to the sum of track sizes according to the spec.
Changes in layout tests are improvement over what we had before.
Previously, the width and height of grid items were set to match the
size of the grid area they belonged to. With this change, if a grid
item has preferred width or height specified to not "auto" value it
will be resolved using grid area as containing block and used instead.
While it's possible to getComputedStyle() on an unconnected element,
the resulting object is not supposed to have any values, since we can't
resolve style without a document root anyway.
This fixes a crash on https://bandcamp.com
If there are min or max size constraints in the cross axis for a flex
item that has a desired aspect ratio, we may need to adjust the main
size *after* applying the cross size constraints.
All the steps to achieving this aren't mentioned in the spec, but it
seems that all other browsers behave this way, so we should too.
Instead of bailing after resolving one violated constraint, we have to
continue down the list of remaining constraints.
We now also call the constraint solver for all replaced elements with
"auto" for both width and height.
Co-authored-by: 0GreenClover0 <clovers02123@gmail.com>
1. Stop using -1 to indicate infinity value of growth limit. Just use
INFINITY for that.
2. More complete implementation of "Expand Flexible Tracks" step.
3. Return AvailableSize from get_free_space: spec says that this
function can return indefinite size and it is ok.
The file gap.html, which previously had multiple grid tests, has now
been divided into smaller files, each containing only one grid test.
It is going to make it easier to identify what inputs have been
affected by changes in layout code.
The file template-areas.html, which previously had multiple grid tests,
has now been divided into smaller files, each containing only one grid
test. It is going to make it easier to identify what inputs have been
affected by changes in layout code.
Also this change removes parts of template-areas.html that we can't
layout correctly yet.
If the parent BFC can come up with a nice stretch-fit width for the flex
container, it will have already done so *before* even entering flex
layout. There's no need to do it again, midway through the flex layout
algorithm.
This wasn't just unnecessary, but we were also doing it incorrectly and
not taking margins into account when calculating the amount of available
space for stretch-fit. This led to oversized flex containers in the
presence of negative margins.
Fixes#18614
Width of table wrapper need to be set to to calculate width of table
box inside. Otherwise TFC will set wrong width assuming width of
containing block is 0.
"The official project language is American English […]."
5d2e915623/CONTRIBUTING.md (L30)
Here's a short statistic of the occurrences of the word "behavio(u)r":
$ git grep -IPioh 'behaviou?r' | sort | uniq -c | sort -n
2 BEHAVIOR
24 Behaviour
32 behaviour
407 Behavior
992 behavior
Therefore, it is clear that "behaviour" (56 occurrences) should be
regarded a typo, and "behavior" (1401 occurrences) should be preferred.
Note that The occurrences in LibJS are intentionally NOT changed,
because there are taken verbatim from the specification. Hence:
$ git grep -IPioh 'behaviou?r' | sort | uniq -c | sort -n
2 BEHAVIOR
10 behaviour
24 Behaviour
407 Behavior
1014 behavior
Although the algorithm for sizing tracks (rows or columns) is defined
once for both dimensions in the specification
(https://www.w3.org/TR/css-grid-2/#algo-track-sizing), we have
implemented it twice separately for sizing rows and columns.
In addition to code duplication, another issue is that these
implementations of the same algorithm have already diverged in some
places, and this divergence is likely to become even worse as our
implementation evolves.
This change unifies code for both dimension into one method that runs
track sizing.
While this change brings a bit of collateral damange (border.html and
minmax.html got changes in layout snaphots) it ultimately brings more
benefits because now we can evolve layout for both rows and colums
without duplicating the code :)
`Length::resolved(Node&)` transforms infinite values to "auto".
Following transformations:
Infinite (Length) -> "auto" -> 0 (px)
cause border-box width to be resolved in zero when it should be inf px.
Removing `Length::resolved(Node&)` makes it work right:
Infinite (Length) -> Infinite (px)
Fixes#18649
The actual cause for the "missing bits" is currently unknown, and this
test case doesn't actually start obviously breaking yet unless we start
reporting errors about missing bits. However, since we are touching the
BitStream implementation already, let's add the test early to make extra
sure that we aren't breaking anything.
https://www.w3.org/TR/CSS22/visuren.html#floats says that when a box
establishes BFC it should not overlap with floats. The way to avoid
overlaps is up to implementor. This change implements avoiding overlap
by narrowing width of a box because it seems like what other engines
do (in the scenarios I tested).
When there are floats present inside an IFC, we must coordinate with
the parent BFC to calculate the automatic width of the IFC's block box.
This is because the IFC is not directly aware of floats. Only the BFC
knows enough about them to account for them in automatic sizing.
In particular, we now blockify layout internal boxes (e.g table parts)
by turning them into `block flow`. This fixes a crash when viewing
our GitHub repo :^)
This patch does three things:
- Factors out the code that determines whether a box will create a new
formatting context for its children (and which type of context)
- Uses that code to mark all formatting context roots in layout tree
dumps. This makes it much easier to follow along with layout since
you can now see exactly where control is transferred to a new
formatting context.
- Rebaselines all existing layout tests, since the output format has
changed slightly.
Fixes the problem that width is incorrectly computed in intrinsic
sizing mode when there are blocks that have min-width or max-width
specified.
Actually that is just the fix of a symptom of the larger problem that
Length::to_px() returns 0 when value is auto regardless of available
size.
This implements conversion from profile connection space to the
device-dependent color for matrix-based profiles.
It only does the inverse color transform but does not yet do the
inverse tone reproduction curve transform -- i.e. it doesn't
implement many cases (LUT transforms), and it does the one thing
it does implement incorrectly. But to vindicate the commit a bit,
it also does the incorrect thing very inefficiently.
GCC 13 was released on 2023-04-26. This commit fixes Lagom build errors
when using an updated host toolchain:
- Adds a workaround for a bug in constraint handling, which made LibJS
fail to compile: https://gcc.gnu.org/bugzilla/show_bug.cgi?id=109683
- Silences the new `-Wdangling-reference` diagnostic globally. It
produces multiple false positives with no clear way to silence them
without `#pragmas`.
- Silences `-Wself-move` in `RefPtr` tests as GCC 13 adds this
previously Clang-exclusive warning.
The framebuffer size was reduced in f2c0cee, but this caused some niche
block layouts to write outside of the frame.
This could be fixed by adding checks to see if a block being predicted/
reconstructed is within the frame, but the branches introduced by that
reduce performance slightly. Therefore, it's better to keep the
framebuffer sized according to the decoded frame size in 8x8 blocks so
that any block can be decoded without bounds checking.
A test was added to ensure that this continues to work.
This fix resolves issue where calculating the min size of a block could
result in incorrect value if width of the block's children was
compensated by margins to fit into container width (which is equal to 0
during min size calculation).
When we determine that a size is definite because it can be resolved now
without performing layout, we also need to account for the box-sizing
property.
This lets us remove a hack from flex layout where box-sizing:border-box
was manually undone at one point in the layout algorithm.
This can be used to convert a profile-dependent color to the L*a*b*
color space.
(I'd like to use this to implement the DeltaE (CIE 2000) algorithm,
which is a metric for how similar two colors are perceived.
(And I'd like to use that to evaluate color conversion roundtrip
quality, once I've implemented full conversions.)
This commit implements following missing steps in table layout:
- Calculate final table height
- Resolve percentage height of cells and rows using final table height
- Distribute avilable height to table rows
If total max columns width (grid_max) is zero then available width
should be divided equally between columns. Previously there was
division by zero: `column.max_width / grid_max`.
Previously, the minimum height of a table row was calculated based
on the automatic height of the cells inner layout. This change makes
computed height of a cell boxes also be considered if it has definite
value.
When deciding on a box type transformation (blockify/inlinify) for a
pseudo element, we have to use the originating element as a reference
rather than the parent.
(The originating element *is* the parent for its pseudo elements.)
If there is a remaining margin-bottom in margin collapsing state
tracker after laying out all boxes in the current BFC, it must be
assigned to the last in-flow child since margin collapsing cannot
occur across a formatting context boundary.
The current issue where margin-bottom may be counted twice due to
"collapse through" margins in the last in-flow child box is addressed
with this fix by excluding such boxes during the search for a box to
assign the remaining margin.
Test case coming with this fix has a layout bug with incorrectly
computed line height.
Previously, statements containing malformed exists expressions such as:
`INSERT INTO t(a) VALUES (SELECT 1)`;
could cause the parser to crash. The parser will now return an error
message instead.
This fixes a bug that was seen when a combination of the grid having
been floated with `float: left` and a `minmax()` column size were used.
The issue was that a grid track size should be considered intrinsically
sized if both the min and max sizes are intrinsic, not just one of them.
Previously, `Heap` would store serialized data in blocks of 1024 bytes
regardless of the actual length. Data longer than 1024 bytes was
silently truncated causing database corruption.
This changes the heap storage to prefix every block with two new fields:
the total data size in bytes, and the next block to retrieve if the data
is longer than what can be stored inside a single block. By chaining
blocks together, we can store arbitrary amounts of data without needing
to change anything of the logic in the rest of LibSQL.
As part of these changes, the "free list" is also removed from the heap
awaiting an actual implementation: it was never used.
Note that this bumps the database version from 3 to 4, and as such
invalidates (deletes) any database opened with LibSQL that is not
version 4.
We now null out smart pointers *before* calling unref on the pointee.
This ensures that the same smart pointer can't be used to acquire a new
reference to the pointee after its destruction has begun.
I ran into this when destroying a non-empty IntrusiveList of RefPtrs,
but the problem was more general so this fixes it for all of RefPtr,
NonnullRefPtr, OwnPtr and NonnullOwnPtr.
This is a tiny bit messy because:
- The spec says we should expand this to `flex: <flex-grow> 1 0`
- All major engines expand it to `flex: <flex-grow> 1 0%`
Spec bug: https://github.com/w3c/csswg-drafts/issues/5742
This is far from perfect, but let's at least make an attempt at laying
out <svg> when encountering it inside another <svg>.
This makes https://awesomekling.substack.com actually load and render
instead of asserting. :^)
Although the spec doesn't mention it, if a flex item has box-sizing:
border-box, and the specified size suggestion is a definite size, we
have to subtract the borders and padding from the size before using it.
This fixes an issue seen in "This Week in Ladybird #4" where the
screenshots ended up in one long vertical stack instead of paired up
2 by 2.
Per SVG2, any coordinate pairs following a moveto command should be
treated as implicit lineto commands with the same absoluteness as the
moveto command.
This attribute is used to define how the viewBox should be scaled.
Previously the behaviour implemented was that of "xMidYMid meet", now
all of them work (expect none :P).
With this the Discord login backend is now correctly scaled/positioned.
This also brings our SVG code a little closer to the spec! With spec
comments and all :^)
(Minor non-visible update to layout tests)
When sizing a flex container with flex-direction:column under a
max-content height constraint, we were incorrectly truncating the
infinite available height to 0 when collecting flex items into lines.
This caused us to put every flex item in its own flex line, which is the
complete opposite of what we want during max-content intrinsic sizing,
as the layout would grow wide but not tall.
This allows accessing and looping over the path segments in a URL
without necessarily allocating a new vector if you want them percent
decoded too (which path_segment_at_index() has an option for).
This now defaults to serializing the path with percent decoded segments
(which is what all callers expect), but has an option not to. This fixes
`file://` URLs with spaces in their paths.
The name has been changed to serialize_path() path to make it more clear
that this method will generate a new string each call (except for the
cannot_be_a_base_url() case). A few callers have then been updated to
avoid repeatedly calling this function.
This isn't actually part of CSS-FLEXBOX-1, but all major engines honor
these properties in flex layout, and it's widely used on the web.
There's a bug open against the flexbox spec where fantasai says the
algorithm will be updated in CSS-FLEXBOX-2:
https://github.com/w3c/csswg-drafts/issues/2336
I've added comments to all the places where we adjust calculations for
gaps with "CSS-FLEXBOX-2" so we can find them easily. When that spec
becomes available, we can add proper spec links.
The consume(size_t) overload consumes "at most" as many bytes as
requested, but consume() consumes exactly one byte.
This commit makes sure to avoid consuming past EOF.
Fixes#18324.
Fixes#18325.
Previously, if we copied the last byte for a length of 100, we'd
recalculate the read span 100 times and memmove one byte 100 times,
which resulted in a lot of overhead.
Now, if we know that we have two consecutive copies of the data, we just
extend the distance to cover both copies, which halves the number of
times that we recalculate the span and actually call memmove.
This takes the running time of the attached benchmark case from 150ms
down to 15ms.
In CLDR 42 and earlier, we were able to assume all cldr-localename files
existed for every locale. They now do not exist for locales that don't
provide any localized data. Namely, this is the "und" locale (which is
an alias for the root locale, i.e. the locale we fall back to when a
user provides an unknown locale).
Further, we were previously able to assume that each currencies.json in
cldr-numbers contained all currencies. This file now excludes currencies
whose localized names are the same as the currency key. Therefore, we
now preprocess currencies.json to discover all currencies ahead of time,
much like we already do for languages.json.
Some of these are allocated upon initialization of the intrinsics, and
some lazily, but in neither case the getters actually return a nullptr.
This saves us a whole bunch of pointer dereferences (as NonnullGCPtr has
an `operator T&()`), and also has the interesting side effect of forcing
us to explicitly use the FunctionObject& overload of call(), as passing
a NonnullGCPtr is ambigous - it could implicitly be turned into a Value
_or_ a FunctionObject& (so we have to dereference manually).
VALUES-4 defines the internal representation of `calc()` as a tree of
calculation nodes. ( https://www.w3.org/TR/css-values-4/#calc-internal )
VALUES-3 lacked any definition here, so we had our own ad-hoc
implementation based around the spec grammar. This commit replaces that
with CalculationNodes representing each possible node in the tree.
There are no intended functional changes, though we do now support
nested calc() which previously did not work. For example:
`width: calc( 42 * calc(3 + 7) );`
I have added an example of this to our test page.
A couple of the layout tests that used `calc()` now return values that
are 0.5px different from before. There's no visual difference, so I
have updated the tests to use the new results.
Some refactoring of our root ca loading process:
- Remove duplicate code
- Remove duplicate calls to `parse_root_ca`
- Load user imported certificates in Browser/RequestServer
Now that these are kind of working, lets add a layout test to prevent
future regressions :^)
This test is the same as the previous example (it is copied, though
that seems to have been done for other tests, e.g. Acid 1).
As noted in serval comments doing this goes against the WC3 spec,
and breaks parsing then re-serializing URLs that contain percent
encoded data, that was not encoded using the same character set as
the serializer.
For example, previously if you had a URL like:
https:://foo.com/what%2F%2F (the path is what + '//' percent encoded)
Creating URL("https:://foo.com/what%2F%2F").serialize() would return:
https://foo.com/what//
Which is incorrect and not the same as the URL we passed. This is
because the re-serializing uses the PercentEncodeSet::Path which
does not include '/'.
Only doing the percent encoding in the setters fixes this, which
is required to navigate to Google Street View (which includes a
percent encoded URL in its URL).
Seems to fix#13477 too
`vformat()` can now accept format specifiers of the form
{:'[numeric-type]}. This will output a number with a comma separator
every 3 digits.
For example:
`dbgln("{:'d}", 9999999);` will output 9,999,999.
Binary, octal and hexadecimal numbers can also use this feature, for
example:
`dbgln("{:'x}", 0xffffffff);` will output ff,fff,fff.
While the tests for sed itself are simple to begin with, some
infrastructure was needed to make them simple.
Firstly, there was no home for tests for the applications under
Utilities, so I had to create a new subdirectory under Tests to host
them.
Secondly, and more importantly, there was previously no easy way to
launch an executable and easily feed it with data for its stdin, then
read its stdout/err and exit code. Looking around the repo I found that
the JS tests do a very similar thing though, so I decided to adapt that
solution for these tests, but with the higher purpose of someday moving
this new Process class to LibCore/Process, where the existing spawn
helpers are still very low level, and there is no representation of a
Process object that one can easily interact with.
Note that this Process implementation is very simple, offers limited
functionality, and it doesn't use the EventLoop, so it can break on long
inputs/outputs depending on the executable behavior.
When BufferedFile.can_read_line() was invoked on files with no newlines,
t incorrectly returned a false result for this single line that, even
though doesn't finish with a newline character, is still a line. Since
this method is usually used in tandem with read_line(), users would miss
reading this line (and hence all the file contents).
This commit fixes this corner case by adding another check after a
negative result from finding a newline character. This new check does
the same as the check that is done *before* looking for newlines, which
takes care of this problem, but only works for files that have at least
one newline (hence the buffer has already been filled).
A new unit test has been added that shows the use case. Without the
changes in this commit the test fails, which is a testament that this
commit really fixes the underlying issue.
WebP lossless files that use a color indexing transform with <= 16
colors use pixel bundling to pack 2, 4, or 8 pixels into a single pixel.
If the image's width doesn't happen to be an exact multiple of the
bundling factor, we need to:
1. Use ceil_div() instead of just dividing the width by the bundling
factor
2. Remember the original width and use it instead of computing
reduced width times bundling factor
This does these changes, and adds a simple test for it -- it at least
checks that the decoded images have the right size.
(I created these images myself in Photoshop, and used the same
technique as for Tests/LibGfx/test-inputs/catdog-alert-*.webp
to create images with a certain number of colors.)
For the test files, I opened Base/res/icons/catdog/alert.png in Adobe
Photoshop 2023, used Image->Mode->Index Color...->
Palette: Local (Perceptive) to reduce the number of colors to 13, 8, and
3 with transparency, and 2 without transparency, then converted it back
to Image->Mode->RGB Color (else it can't be saved as webp), then
File->Save a Copy... to save a WebP (mode lossless) for every palette
size.
The image is https://quakewiki.org/wiki/File:Qpalette.png in lossless
webp format with a color indexing transform.
I've created Qpalette.webp by running
examples/cwebp -z 0 ~/src/serenity/tmp.ppm -o Qpalette.webp
built at libwebp webmproject/libwebp@0825faa4c1 (without
png support, so I first ran
Build/lagom/image ~/Downloads/Qpalette.png -o tmp.ppm
to convert it from png to a format my cwebp binary could read).
This file also happens to explicitly set max_symbol, so it serves
as a test for that code path as well.
Grid and flex containers have their own rules for abspos items, so we
shouldn't try to be clever and put them in the "current" anonymous
wrapper block. That behavior is primarily for the benefit of block &
inline layout.
When calculating the intrinsic width of a block-level box, we now ignore
the preferred width entirely, and not just when the preferred width
should be treated as auto.
The condition for this was both confused and wrong, as it looked at the
available width around the box, but didn't check for a width constraint
on the box itself.
Just because the available width has an intrinsic sizing constraint
doesn't mean that the box is undergoing intrinsic sizing. It could also
be the box's containing block!
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 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.
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.
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.
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.
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.
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>
We were not testing this logic and I caused a regression while
modifying some of the hashing code, so let's add these. Note that I only
added two tests to test both 'families' of implementations for the SHA
hashes.
This builds on the existing ad-hoc ResourceLoader code for HTTP fetches
which works for files as well.
This also includes a test that checks that stylesheets loaded with the
"file" URL scheme actually work.
This would previously assert in InlineFormattingContext because we had
an outwardly inline box that wasn't inwardly flow.
Fix this by converting text-based input boxes to inline-blocks. This is
an ad-hoc solution, and there might be a much better way to solve it.
Before this change `apply_clip_overflow_rect` might crash trying to
access `clip_rect` that does not have value because we currently
support calculation of visible rectangle when `overflow: hidden`
is applied for both axis.
The deallocation of the test cases at the very end happens through a
NonnullRefPtr<TestCase>, meaning the deallocation will assume the wrong
object size and trip up ASAN. Therefore, we cannot use a TestCase
subclass.
I also took this opportunity and made use of the new LoaderError
formatter.
This fixes an issue where e.g `height: 100%` on a flex item whose
container has indefinite height was being resolved to 0. It now
correctly behaves the same as auto.
Previously, when having inline contexts consisting of just a `<br/>`
tag, we would not create a line box.
Ensure that there is always a line box when a line is explicitly being
broken and also ensure it won't be trimmed due to being empty.
This will a fix a number of sites that use `<br>` tags for layouts
between block elements (even though the spec says they shouldn't).
With Clang, the previous/next pointers in buckets of an
`OrderedHashTable` are not cleared when a bucket is being shifted up as
a result of a removed bucket. As a result, an unfortunate pointer mixup
could lead to an infinite loop in the `HashTable` iterator, which was
exposed in `HashMap::keys()`.
Co-authored-by: Luke Wilde <lukew@serenityos.org>
The image made the test flaky when running on my machine, so this
doesn't seem safe at the moment. We can just hardcode the dimensions.
Eventually we should make it possible to use external images in tests,
but for now let's not flake up the CI.
According to CSS Inline Layout Module Level 3 § 2.2 Step 1. atomic
inlines should be layed out in a line box based on their margin box.
However, up until this patch we were unconditionally considering only
the border box during line box height calculation. This made us
essentially drop all vertical margins for atomic inlines.
Similar to POSIX read, the basic read and write functions of AK::Stream
do not have a lower limit of how much data they read or write (apart
from "none at all").
Rename the functions to "read some [data]" and "write some [data]" (with
"data" being omitted, since everything here is reading and writing data)
to make them sufficiently distinct from the functions that ensure to
use the entire buffer (which should be the go-to function for most
usages).
No functional changes, just a lot of new FIXMEs.
Before, some loader plugins implemented their own buffering (FLAC&MP3),
some didn't require any (WAV), and some didn't buffer at all (QOA). This
meant that in practice, while you could load arbitrary amounts of
samples from some loader plugins, you couldn't do that with some others.
Also, it was ill-defined how many samples you would actually get back
from a get_more_samples call.
This commit fixes that by introducing a layer of abstraction between the
loader and its plugins (because that's the whole point of having the
extra class!). The plugins now only implement a load_chunks() function,
which is much simpler to implement and allows plugins to play fast and
loose with what they actually return. Basically, they can return many
chunks of samples, where one chunk is simply a convenient block of
samples to load. In fact, some loaders such as FLAC and QOA have
separate internal functions for loading exactly one chunk. The loaders
*should* load as many chunks as necessary for the sample count to be
reached or surpassed (the latter simplifies loading loops in the
implementations, since you don't need to know how large your next chunk
is going to be; a problem for e.g. FLAC). If a plugin has no problems
returning data of arbitrary size (currently WAV), it can return a single
chunk that exactly (or roughly) matches the requested sample count. If a
plugin is at the stream end, it can also return less samples than was
requested! The loader can handle all of these cases and may call into
load_chunk multiple times. If the plugin returns an empty chunk list (or
only empty chunks; again, they can play fast and loose), the loader
takes that as a stream end signal. Otherwise, the loader will always
return exactly as many samples as the user requested. Buffering is
handled by the loader, allowing any underlying plugin to deal with any
weird sample count requirement the user throws at it (looking at you,
SoundPlayer!).
This (not accidentally!) makes QOA work in SoundPlayer.
Percentage line-height values are relative to 1em (i.e the font-size
of the element). We have to resolve their computed values before
proceeding with inheritance.
If normal flow layout has caused us to progress past the current
innermost float in the block axis, we still need to consider the floats
stacked outside of it.
Fix this by always walking the currently stacked floats from innermost
to outermost when placing new floats.
When using the flex shrink factor, the flexible length resolution
algorithm was incorrectly ignoring the `frozen` flag on items and would
update the same items again, causing overconsumption of the remaining
free space on the flex line.
In case flex items had `margin: auto` on the primary flex axis, we were
still also distributing remaining space according to `justify-content`
rules. This lead to duplicated spacing in various places and overflows.
It looks like this issue was observed previously but missidentified
because there was logic to ignore margins at the start and end which
would partially paper over the root cause. However this created other
bugs (like for example not having a margin at beginning and end ;-)) and
I can find nothing in the spec or other browser behaviour that indicates
that this is something that should be done.
Now we skip justify-content space distribution alltogether if it has
already been distributed to auto margins.
The draft CSS-FLEXBOX-1 spec had a more detailed description of this
algorithm, so let's use that as our basis for the implementation.
Test by Aliaksandr. :^)
When resolving these constraints to CSS pixel sizes, we have to resolve
padding-top and padding-bottom against the flex container's *width*,
not its height.
The padding-top and padding-bottom properties are relative to the
*width* of the containing block, not the height.
It's funny how we keep making this same mistake again and again. :^)
We currently fully casefold the left- and right-hand sides to compare
two strings with case-insensitivity. Now, we casefold one code point at
a time, storing the result in a view for comparison, until we exhaust
both strings.
