TL;DR: There are two available sets of coefficients for the conversion
matrices from XYZ to sRGB. We switched from one set to the other, which
is what the WPT tests are expecting.
All RGB color spaces, like display-p3 or rec2020, are defined by their
three color chromacities and a white point. This is also the case for
the video color space Rec. 709, from which the sRGB color space is
derived. The sRGB specification is however a bit different.
In 1996, when formalizing the sRGB spec the authors published a draft
that is still available here [1]. In this document, they also provide
the matrix to convert from the XYZ color space to sRGB. This matrix can
be verified quite easily by using the usual math equations. But hold on,
here come the plot twist: at the time of publication, the spec contained
a different matrix than the one in the draft (the spec is obviously
behind a pay wall, but the numbers are also reported in this official
document [2]). This official matrix, is at a first glance simply a
wrongly rounded version of the one in the draft publication. It however
has some interesting properties: it can be inverted twice (so a
roundtrip) in 8 bits and not suffer from any errors from the
calculations.
So, we are here with two versions of the XYZ -> sRGB matrix, the one
from the spec, which is:
- better for computations in 8 bits,
- and official. This is the one that, by authority, we should use.
And a second version, that can be found in the draft, which:
- makes sense, as directly derived from the chromacities,
- is publicly available,
- and (thus?) used in most places.
The old coefficients were the one from the spec, this commit change them
for the one derived from the mathematical formulae. The Python script to
compute these values is available at the end of the commit description.
More details about this subject can be found here [3].
[1] https://www.w3.org/Graphics/Color/sRGB.html
[2] https://color.org/chardata/rgb/sRGB.pdf
[3] https://photosauce.net/blog/post/making-a-minimal-srgb-icc-profile-part-3-choose-your-colors-carefully
The Python script:
```python
# http://www.brucelindbloom.com/index.html?Eqn_RGB_XYZ_Matrix.html
from numpy.typing import NDArray
import numpy as np
### sRGB
# https://www.w3.org/TR/css-color-4/#predefined-sRGB
srgb_r_chromacity = np.array([0.640, 0.330])
srgb_g_chromacity = np.array([0.300, 0.600])
srgb_b_chromacity = np.array([0.150, 0.060])
##
## White points
white_point_d50 = np.array([0.345700, 0.358500])
white_point_d65 = np.array([0.312700, 0.329000])
#
r_chromacity = srgb_r_chromacity
g_chromacity = srgb_g_chromacity
b_chromacity = srgb_b_chromacity
white_point = white_point_d65
def tristmimulus_vector(chromacity: NDArray) -> NDArray:
return np.array([
chromacity[0] /chromacity[1],
1,
(1 - chromacity[0] - chromacity[1]) / chromacity[1]
])
tristmimulus_matrix = np.hstack((
tristmimulus_vector(r_chromacity).reshape(3, 1),
tristmimulus_vector(g_chromacity).reshape(3, 1),
tristmimulus_vector(b_chromacity).reshape(3, 1),
))
scaling_factors = (np.linalg.inv(tristmimulus_matrix) @
tristmimulus_vector(white_point))
M = tristmimulus_matrix * scaling_factors
np.set_printoptions(formatter={'float_kind':'{:.6f}'.format})
xyz_65_to_srgb = np.linalg.inv(M)
# http://www.brucelindbloom.com/index.html?Eqn_ChromAdapt.html
# Let's convert from D50 to D65 using the Bradford method.
m_a = np.array([
[0.8951000, 0.2664000, -0.1614000],
[-0.7502000, 1.7135000, 0.0367000],
[0.0389000, -0.0685000, 1.0296000]
])
cone_response_source = m_a @ tristmimulus_vector(white_point_d50)
cone_response_destination = m_a @ tristmimulus_vector(white_point_d65)
cone_response_ratio = cone_response_destination / cone_response_source
m = np.linalg.inv(m_a) @ np.diagflat(cone_response_ratio) @ m_a
D50_to_D65 = m
xyz_50_to_srgb = xyz_65_to_srgb @ D50_to_D65
print(xyz_50_to_srgb)
print(xyz_65_to_srgb)
```
This color space is often used as a reference in WPT tests, having
support for it makes us pass 15 new tests:
- css/css-color/rec2020-001.html
- css/css-color/rec2020-002.html
- css/css-color/rec2020-003.html
- css/css-color/rec2020-004.html
- css/css-color/rec2020-005.html
- css/css-color/predefined-011.html
- css/css-color/predefined-012.html
That makes us pass the following WPT tests:
- css/css-color/prophoto-rgb-001.html
- css/css-color/prophoto-rgb-002.html
- css/css-color/prophoto-rgb-003.html
- css/css-color/prophoto-rgb-004.html
- css/css-color/prophoto-rgb-005.html
- css/css-color/predefined-009.html
- css/css-color/predefined-010.html
This color space is often used as a reference in WPT tests, having
support for it makes us pass 15 new tests:
- css/css-color/display-p3-001.html
- css/css-color/display-p3-002.html
- css/css-color/display-p3-003.html
- css/css-color/display-p3-004.html
- css/css-color/display-p3-005.html
- css/css-color/display-p3-006.html
- css/css-color/lab-008.html
- css/css-color/lch-008.html
- css/css-color/oklab-008.html
- css/css-color/oklch-008.html
- css/css-color/predefined-005.html
- css/css-color/predefined-006.html
- css/css-color/xyz-005.html
- css/css-color/xyz-d50-005.html
- css/css-color/xyz-d65-005.html
This makes us pass the following WPT tests:
- css/css-color/a98rgb-001.html
- css/css-color/a98rgb-002.html
- css/css-color/a98rgb-003.html
- css/css-color/a98rgb-004.html
- css/css-color/predefined-007.html
- css/css-color/predefined-008.html
Let's adapt this class a bit better to how it's actually being used.
Instead of having valid/invalid states and storing an error in case
it's invalid, a MappedFile is now always valid, and the factory
function that creates it will return an OSError if mapping fails.
These changes are arbitrarily divided into multiple commits to make it
easier to find potentially introduced bugs with git bisect.Everything:
The modifications in this commit were automatically made using the
following command:
find . -name '*.cpp' -exec sed -i -E 's/dbg\(\) << ("[^"{]*");/dbgln\(\1\);/' {} \;
We shouldn't reject indexed palette PNGs just because they have fewer
palette entries than the bit depth allows. Instead, we need to check
for OOB palette accesses and fail the decode *then*.
This brings mmap more in line with other operating systems. Prior to
this, it was impossible to request memory that was definitely committed,
instead MAP_PURGEABLE would provide a region that was not actually
purgeable, but also not fully committed, which meant that using such memory
still could cause crashes when the underlying pages could no longer be
allocated.
This fixes some random crashes in low-memory situations where non-volatile
memory is mapped (e.g. malloc, tls, Gfx::Bitmap, etc) but when a page in
these regions is first accessed, there is insufficient physical memory
available to commit a new page.
Old font functionality has been moved into BitmapFont
and an abstract Font interface has been introduced to
faciliate further development of TTF font integration.
Compared to version 10 this fixes a bunch of formatting issues, mostly
around structs/classes with attributes like [[gnu::packed]], and
incorrect insertion of spaces in parameter types ("T &"/"T &&").
I also removed a bunch of // clang-format off/on and FIXME comments that
are no longer relevant - on the other hand it tried to destroy a couple of
neatly formatted comments, so I had to add some as well.
We previously had a cached bold variant font in Gfx::Font that was very
haphazardly located by filename pattern. This patches replaces that
mechanism with a proper Gfx::FontDatabase lookup for the same font but
with bold weight (700).
The new algorithm is an iterative one with an arbitrary threshold for splitting
curves. It splits curves evenly. This should theoretically be less accurate
than the existing recursive approach, but seems to give subjectively better
results in practice.
I'm planning to make this a minimal-allocation TTF parser. This will
speed up start-up time for applications, but have some overhead for
rasterizing glyphs. Which should be okay, since rasterized glyph bitmaps
should be cached anyway.
This commit just adds the loading of the HEAD table.
This was causing WindowServer and Taskbar to crash sometimes when the
stars aligned and we tried cutting off a string ending with "..." right
on top of an emoji. :^)
