Commit Graph

63 Commits

Author SHA1 Message Date
Matthias Clasen
7450427757 gsktransform: Convert docs 2021-03-11 16:37:32 +00:00
Matthias Clasen
ca3120919f docs: Improve gsk docs
Convert link formats, add summaries, and make
a few missing things show up in the docs.
2021-03-11 16:37:31 +00:00
Christian Hergert
6fab195643 gsk: move transform structure to private.h
This allows using the category without having to call a function from the
hot paths of the renderer.
2021-02-10 16:14:59 -08:00
Timm Bäder
fdda4e4e99 GskTransform: Add nullable annotations to _equals()
Both can be NULL here.
2020-12-21 10:26:03 +01:00
Timm Bäder
186287ecfb GskTransform: Fix identity transform class name 2020-12-21 10:26:02 +01:00
Timm Bäder
bd5d1615ac GskTransform: Avoid a ref+unref pair
If gsk_transform_is_identity() returned FALSE for the next transform,
the previous code did a ref + unref pair, even though it was unneeded.
2020-12-21 10:26:02 +01:00
Sophie Herold
cd0bef9390 Add some missing nullable return annotations 2020-10-17 15:32:36 +02:00
Matthias Clasen
431f144f37 gsk: Normalize when transforming bounds
Bounds are assumed to be normalized, and transforms
with negative scales or 3d rotations can make us get
negative sizes.

Fix by Benjamin Otte.
2020-09-28 11:34:16 -04:00
Matthias Clasen
41497aeb8b Add gsk_matrix_transform_rect
This is a projecting version of the corresponding
graphene api. While we are at it, rewrite
gsk_matrix_transform_bounds() to use
gsk_matrix_transform_rect().
2020-09-04 13:15:21 -04:00
Matthias Clasen
e36b594867 Replace uses of graphene_matrix_transform_
Replace our uses of graphene_matrix_transform_point,
_point3d and _bounds by our own versions that handle
projective transforms correctly.

This fixes render node bounds being incorrect for widgets
involving projective transforms (e.g. testrevealer swing
transformations), and also fixes picking on such widgets.
2020-09-01 09:18:38 -04:00
Matthias Clasen
99a7a2188a gsk: Add some private transform apis
Some of the graphene_matrix_transform apis don't work
correctly with projective matrices, so add our own
variants.
2020-09-01 09:18:38 -04:00
Björn Daase
6315cd977c *: Fix spelling mistakes found by codespell 2020-08-21 15:29:34 +02:00
Timm Bäder
fa85ec5cf5 GskTransform: Restructure _to_affine and _to_translation
Make the simple cases more explicit.
2020-07-28 05:34:12 +02:00
Timm Bäder
756b84033a GskTransform: Remove unused function from private header
We only use this in gsktransform.c
2020-07-28 05:34:12 +02:00
Matthias Clasen
13b37b0ba4 gsk: Fix a transform leak
This was found by running our testsuite under asan.
2020-07-08 17:44:08 -04:00
Timm Bäder
da85815399 transform: Fix docs to use "transform" instead of "matrix" 2020-06-19 05:26:25 +02:00
Timm Bäder
5dde4c0a39 transform: Fix equal() docs
This has nothing to do with matrices.
2020-06-19 05:26:25 +02:00
Timm Bäder
0c1cfbbad1 transform: Compare 3d points directly 2020-06-19 05:26:24 +02:00
Timm Bäder
a974c83423 gsktransform: Make next == NULL case explicit
This is handled with the is_identity() check as well but this way it's
much more obvious what's happening
2020-06-19 05:26:24 +02:00
Benjamin Otte
7597f6b594 transform: Don't crash for gsk_transform_transform (id, id)
See attached tests
2020-02-21 18:30:13 +01:00
Benjamin Otte
b50093d044 transform: Make sure the identity transform is equal to NULL 2020-02-21 18:25:05 +01:00
Benjamin Otte
41ef6e9fa5 transform: Add optimization for common case
Transforming identity by an other transform does not mean we need to
painstakingly apply the individual steps of other to construct a new
transform, it means we can just return other.

Or in math terms:
  I * B = B
so just return B.
2020-02-21 18:19:16 +01:00
Alexander Larsson
1f0438e7fe gsk: Fix angle normalization
I was getting assertions that normalize_angle() failed the
result < 260 check. Doing some research on this it turns out
to be a precision issue. If the incomming angle is very slightly
below zero, then adding 360 to it may end up with exactly 360.

I simplified the code a bit to avoid division and rounding, because in
practice most angles will be "just outside" the 0-360 degree anyway.
And i also added a workaround for the "result is 360" case by just
setting it to 0.
2020-02-11 13:51:06 +01:00
Timm Bäder
8388791f87 GskTransform: Remove useless fmod check
All angles are in the [0; 360[ range
2020-02-07 13:16:32 -05:00
Timm Bäder
cbb05a3eab GskTransform: add assertions to make normalize_angle() intentions clear
360deg should be expressed as 0deg and no angle should ever be
negative.
2020-02-07 13:16:32 -05:00
Timm Bäder
cff8304133 GskTransform: Add gsk_transform_transform_point()
Equivalent of gsk_transform_transform_bounds() and
graphene_matrix_transform_point() respectively.
2020-01-24 06:19:12 +01:00
Chun-wei Fan
a9b1d4a389 build: Check for sincosf()
sincosf() is really a GCC-specific function that may more may not be
supported on non-GCC compilers, so we want to check for it, otherwise we
use a fallback implementation, not unlike the one in
demos/gtk-demo/gtkgears.c.
2020-01-08 17:15:29 +08:00
Timm Bäder
7669e6e42e transform: Properly compare scale transforms
If the epsilon we use there is too much, we can run into rendering
errors because the GPU will use the actual scale value.
2020-01-07 17:27:19 +01:00
Timm Bäder
7a27122dab transform: Scale+translate bounds directyl
Make these two code paths look the same.
2020-01-07 17:27:18 +01:00
Timm Bäder
ca4ae81170 transform: Compare class before ->next 2020-01-07 17:27:18 +01:00
Timm Bäder
766d4dff76 GskTransform: Compute sin/deg of 90deg rotations manually
Otherwise we might end up with inaccuracies, e.g. when
gsk_transform_transform_bounds()'ing bounds.
2020-01-07 17:27:17 +01:00
Timm Bäder
e62f135c6a GskTransform: Normalize angles
Make sure all angles are in the [0..365) range
2020-01-07 17:27:17 +01:00
Timm Bäder
6c90d3a1b6 gsktransform: Fix documentation comment
There is no @m.
2019-09-09 17:36:23 +02:00
Rico Tzschichholz
5504c62af9 gsk: Add missing 'transfer full' annotations of instance parameters
Fixes https://gitlab.gnome.org/GNOME/gtk/issues/2107
2019-08-26 18:05:17 +02:00
Timm Bäder
dbc49e7742 transform: Fix identity transform fast paths
Returning an extra ref will cause leaks later.
2019-07-07 07:24:00 +02:00
Timm Bäder
3bc3e140dd transform: Add transform_bounds fast path for 2D_AFFINE transforms
E.g. anything involving a scale. This is important when e.g. scrolling
in the node list in the recorder, which scales every recorded node down
to fit in the list.
2019-06-29 09:49:38 +02:00
Timm Bäder
8dd74eac2e transform: Add skew parsing 2019-06-29 07:15:43 +02:00
Matthias Clasen
206970c1e8 Merge branch 'transform-tests' into 'master'
Transform tests

See merge request GNOME/gtk!910
2019-06-03 16:41:17 +00:00
Matthias Clasen
4ed57449ba gsk: Fix printing of identity transforms
These need to print as "none", otherwise
the parser does not accept them.
2019-06-03 16:20:38 +00:00
Matthias Clasen
d93fb5a3fa Cosmetics
Fix up a variable name mismatch in GskTransform docs.
2019-06-03 16:15:16 +00:00
Timm Bäder
a1161456ff GskTransform: NULL is a valid transform 2019-06-03 17:36:31 +02:00
Timm Bäder
05ca772631 GskTransform: Ignore identity transforms
We often end up with e.g. a scale of 1 or a translation by 0/0. Ignore
those transforms since they don't do anything.
2019-06-03 17:03:02 +02:00
Benjamin Otte
5a0c8805fc transform: Coalesce similar transforms
If somebody does a transform like
  scale(5) scale(10) translate(1,1) translate(5,0)
store it instead as
  scale(50) translate(6,1)
This way, less memory is consumed and transforms are easier to read.

In particular, this simplifies the typical transforms we do in GTK,
which are just one translation after another.
2019-06-02 21:18:56 +02:00
Benjamin Otte
69629ebb1a transform: Treat 0 perspective transforms as identity
This happens if you apply a perspective transform and its inverse, which
is the negative version of the perspective.
2019-06-02 15:12:37 +02:00
Matthias Clasen
3f940715e4 gsk: Documentation tweaks
Add some missing symbols.
2019-05-01 23:24:14 -04:00
Emmanuele Bassi
cf9deb7cf5 Use getters for translation and scaling components
Instead of accessing the matrix cells directly.
2019-04-29 02:26:31 +02:00
Emmanuele Bassi
40beb69487 Never compare floats for equality
Floating point values cannot ever be compared for equality. GLib has a
G_APPROX_VALUE macro that lets us compare two value within a provided
precision, so we should use that instead.
2019-04-29 02:26:31 +02:00
Emmanuele Bassi
f1cadee196 Use matrix equality in GskTransform
Graphene has new API to check for equality, so we should use it instead
of doing a byte-by-byte comparison.
2019-04-29 02:26:31 +02:00
Emmanuele Bassi
478fdaa632 Use atomic boxing instead of manual refcounting
Artisanal, homegrown, locally sourced, vegan reference counting has been
replaced by the appropriate API in GLib, which does small things like
saturation and type checking.
2019-04-29 02:25:18 +02:00
Benjamin Otte
8001c7d972 transform: Add gsk_transform_parse()
It uses the new CSS parser.
2019-04-12 19:34:29 +02:00