| Commit message (Collapse) | Author | Age | Files | Lines |
|\
| |
| |
| | |
Change-Id: Idb141c715c7d06d2ecdce403d0f3740ee18b2b8f
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
Currently, elements created by the NodeInstantiator are children of the
instantiator itself. Change that to parent them in the instantiator
parent instead.
Rationale for that is that it will behave similarly to Repeater this
way, and gives a chance of using NodeInstantiator in frame graphs.
Indeed, anything which is not a FrameGraphNode gets pruned in that
context that would include branches starting from NodeInstantiator.
Change-Id: I651a51471d92fcf466c9abd5dbbbfa2960ef1247
Task-Id: QTBUG-55908
Reviewed-by: Sean Harmer <sean.harmer@kdab.com>
|
|/
|
|
|
| |
Change-Id: I8fd9fde776b81b09b3d107f2ad95f90672a354ab
Reviewed-by: Sean Harmer <sean.harmer@kdab.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Also fix up some more problems with the torus. We were drawing some
triangles more than once due to problems in the indexing. Fixed this
by factoring out the calculations of number of vertices and
triangles and takign care about needing more vertices to handle the
texture coordinate wrapping at the seams but not more indices.
Also renamed some members in the functors to be more consistent.
Now the torus geometry should be perfect.
Change-Id: I1198469531a5fd5e4ec8b86d47c9e66850fe50c7
Reviewed-by: Kevin Ottens <kevin.ottens@kdab.com>
|
|
|
|
|
|
|
| |
They will be needed to test other geometry convenience classes.
Change-Id: Ibece97c393e3b0239454f28e7862cfffd400a0e8
Reviewed-by: Kevin Ottens <kevin.ottens@kdab.com>
|
|
|
|
|
| |
Change-Id: I81634986b22871f2c0cb9175925ca6ac4714694f
Reviewed-by: Kevin Ottens <kevin.ottens@kdab.com>
|
|
|
|
|
| |
Change-Id: Id41b28c19cba2997547931e2cddfe394c82b3fe3
Reviewed-by: Sean Harmer <sean.harmer@kdab.com>
|
|
|
|
|
|
|
|
|
| |
It doesn't quite make sense to have a null scale by default. It basically
forces to always touch the property to get something moving. Use a scale of 1
instead by default.
Change-Id: I813997d23ffedf1a934f79868f1f71f886b0303f
Reviewed-by: Sean Harmer <sean.harmer@kdab.com>
|
|
|
|
|
|
| |
Change-Id: Ifa2c7f3c33cb410938e484f023e965d5ca5220cb
Task-number: QTBUG-55908
Reviewed-by: Sean Harmer <sean.harmer@kdab.com>
|
|
|
|
|
|
| |
Task-number: QTBUG-55836
Change-Id: Ic5ee72f5ffb236fe15e46fa667159a139ad05b03
Reviewed-by: Sean Harmer <sean.harmer@kdab.com>
|
|
|
|
|
|
|
|
| |
Replaces isPropertyToBeTransformed by propertyTransformType which returns the
TransformType (NoTransform) if the property isn't transformed.
Change-Id: Ib7d91bdcfe18409e6ad6d6f5bc119c46dd7fb143
Reviewed-by: Sean Harmer <sean.harmer@kdab.com>
|
|
|
|
|
| |
Change-Id: I6a570f16a303277b8305b36bef805991680ae1ea
Reviewed-by: Sean Harmer <sean.harmer@kdab.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Only store the original properties (does not store transformed values anymore
as these can change on a per thread basis and lead to races in the
ShaderData).
Instead introduce a method to check if a property needs to be transformed and
a method which given a viewMatrix returns the transformed value. This can
then be used concurrently by each UniformBlockValueBuilder without introducing
races.
Also since building UBO from a ShaderData has been disabled since 5.7, remove
all the nested ShaderData update logic.
Unit tests updated.
Change-Id: Id0e5956b9f7d93f8d019c3d8999268fc3ad68e79
Task-number: QTBUG-54818
Reviewed-by: Kevin Ottens <kevin.ottens@kdab.com>
|
|
|
|
|
| |
Change-Id: I64f815587947ad2f63462bb38027c533c7aa4bdc
Reviewed-by: Paul Lemire <paul.lemire@kdab.com>
|
|
|
|
|
|
|
|
|
|
| |
Add the new test, which renders to multiple windows simultaniously. Can be
used to test that the resume/suspend of render thread works, among other
things.
Change-Id: I9ace3e833a701674d6021498fca6d8e837e94adf
Reviewed-by: Tomi Korpipää <tomi.korpipaa@qt.io>
Reviewed-by: Paul Lemire <paul.lemire@kdab.com>
|
|
|
|
|
| |
Change-Id: Id57a69a80f9eddb28fd2e75ea55baf98dcac452c
Reviewed-by: Sean Harmer <sean.harmer@kdab.com>
|
|
|
|
|
|
| |
Change-Id: I79c159c70ed81345b58c6dabfd60437e20554e94
Task-Id: QTBUG-54585
Reviewed-by: Sean Harmer <sean.harmer@kdab.com>
|
|
|
|
|
|
| |
Change-Id: I2c8a34009d3bac400f5e156fe6321af744960de5
Task-Id: QTBUG-54585
Reviewed-by: Sean Harmer <sean.harmer@kdab.com>
|
|
|
|
|
| |
Change-Id: Ic4ab2d781403ada1206e7429ad1cd16ea2330aed
Reviewed-by: Sean Harmer <sean.harmer@kdab.com>
|
|
|
|
|
| |
Change-Id: If5a28e670e8eef0a9b5a75dc3451ee584f22e72c
Reviewed-by: Sean Harmer <sean.harmer@kdab.com>
|
|
|
|
|
|
|
|
|
|
| |
The bookkeeping and activation logic of ActionInput, InputChord and
InputSequence now move outside of the job, which make them testable
properly. The tests already show bugs which will be fixed in following
commits.
Change-Id: I4e434118a6ba285d08632c43d487b17ee5b4cb66
Reviewed-by: Sean Harmer <sean.harmer@kdab.com>
|
|
|
|
|
|
|
|
| |
Now it also comes with a backend node and a device integration
implementation so that one can simulate button presses for instance.
Change-Id: I87a3ecf9e68904fdd52f7c94f55abbfed28c6e9c
Reviewed-by: Sean Harmer <sean.harmer@kdab.com>
|
|
|
|
|
| |
Change-Id: I7e560a35f989890c2555808d0a86e0fa05b0e60b
Reviewed-by: Sean Harmer <sean.harmer@kdab.com>
|
|
|
|
|
| |
Change-Id: I6ffb9e0a743dfad78624b2edcfa2d93d69d685dd
Reviewed-by: Sean Harmer <sean.harmer@kdab.com>
|
|
|
|
|
| |
Change-Id: Ia600d227a2511b15f9305b69c135dc09188fbf68
Reviewed-by: Sean Harmer <sean.harmer@kdab.com>
|
|
|
|
|
| |
Change-Id: I2e580ea9cd382aff0224b38a9a1431c67661666b
Reviewed-by: Sean Harmer <sean.harmer@kdab.com>
|
|
|
|
|
| |
Change-Id: I8a598a0edad751d3e92c7e6893f5761d5b553070
Reviewed-by: Sean Harmer <sean.harmer@kdab.com>
|
|
|
|
|
|
| |
Change-Id: I52159b0fa78ab915e73c2dc84bda9bccc3dbd0ef
Task-number: QTBUG-54889
Reviewed-by: Sean Harmer <sean.harmer@kdab.com>
|
|
|
|
|
|
|
|
| |
Will allow to set the status in the backend and update the frontend with it.
Change-Id: Iac47e9960b8ef56c4e45ea31c616c4bc2a8f3444
Task-number: QTBUG-54889
Reviewed-by: Sean Harmer <sean.harmer@kdab.com>
|
|
|
|
|
| |
Change-Id: I6f92482750e5e013dc951a3028b5aab7c1c7de4c
Reviewed-by: Sean Harmer <sean.harmer@kdab.com>
|
|
|
|
|
| |
Change-Id: I9cbb718347605b5ea048e9f37251817339b42412
Reviewed-by: Sean Harmer <sean.harmer@kdab.com>
|
|
|
|
|
|
|
|
|
| |
Help reveal potential issues. Also adjusted blending arguments to work
properly.
Change-Id: I86687aefe3bbc24cd9de559fa813d7c815e321e0
Task-number: QTBUG-54591
Reviewed-by: Kevin Ottens <kevin.ottens@kdab.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
When cleanup is called that means an Entity needs to be recycled for reuse
later on. The children handles vector wasn't cleared which resulted in it
containing invalid handles when the Entity element was reused. In turn that
eventually resulted in asserts in the handle manager when that Entity was
released again (which explains crashes with the QSceneLoader changing
sources).
Change-Id: Ic2e69132e4a5186f92936593ddcf3da72a7e1446
Task-number: QTBUG-54889
Reviewed-by: Sean Harmer <sean.harmer@kdab.com>
|
|
|
|
|
| |
Change-Id: I5f8c7f0d1de5e54964d54711daa39703a003736d
Reviewed-by: Paul Lemire <paul.lemire@kdab.com>
|
|
|
|
|
|
|
|
|
|
| |
For coherency with other function names. This kind of API change is allowed
since Qt3DWindow is in the Extras module which is still in tech preview for
5.7
Change-Id: Ic67f96cfb2dfcb3b2d724c97f2cf472bfaf40a07
Task-number: QTBUG-54841
Reviewed-by: Paul Lemire <paul.lemire@kdab.com>
|
|
|
|
|
|
|
|
| |
This could otherwise have happened if an element was acquired and released
repeatedly.
Change-Id: I29a327f32fdd9d9a02460bff26c9ff56fdd3dd61
Reviewed-by: Paul Lemire <paul.lemire@kdab.com>
|
|
|
|
|
| |
Change-Id: I40bbe3cb61e6ee825401a1ae55a12f3362b1a95f
Reviewed-by: Sean Harmer <sean.harmer@kdab.com>
|
|
|
|
|
| |
Change-Id: I16c60a9cce1b003847a60b0f3e4a0f3f5127c70b
Reviewed-by: Sean Harmer <sean.harmer@kdab.com>
|
|
|
|
|
| |
Change-Id: I3602e2c837d351966d7b4216ecde458f80ba53a2
Reviewed-by: Sean Harmer <sean.harmer@kdab.com>
|
|
|
|
|
|
| |
Change-Id: If3f9e1e240f6323bf505f81424cba319c912c03a
Task-number: QTBUG-55163
Reviewed-by: Sean Harmer <sean.harmer@kdab.com>
|
|
|
|
|
|
| |
Change-Id: I8e89a6c0adc2b99e5fe2af61629df2f2a6ae3ccd
Task-number: QTBUG-55163
Reviewed-by: Sean Harmer <sean.harmer@kdab.com>
|
|
|
|
|
|
|
|
|
| |
The window size needs to be set as a QSizeF and not a QSize as the shader
expects a vec2. int to float conversion was failing and giving unexpected
values to the shader.
Change-Id: Ide4b4a884011e320bc909da27633e6adee1956c1
Reviewed-by: Sean Harmer <sean.harmer@kdab.com>
|
|
|
|
|
| |
Change-Id: Ieb2fe26d46279ef29a00194429889c0987f6ad40
Reviewed-by: Sean Harmer <sean.harmer@kdab.com>
|
|
|
|
|
|
|
|
|
|
| |
Add test for default construction. Making transforms non-
shareable will make it easier to implement feeding back the
world transform as a new property later without introducing
ambiguities.
Change-Id: I6d472b3da90ce1dacdfcc6f8366a23706044eb5c
Reviewed-by: Paul Lemire <paul.lemire@kdab.com>
|
|
|
|
|
|
| |
Change-Id: I776c5104579d500e557f4ca5b72271a5ff1b4a79
Task-number: QTBUG-55117
Reviewed-by: Sean Harmer <sean.harmer@kdab.com>
|
|
|
|
|
|
|
|
| |
Otherwise technique selection fails
Change-Id: I2ae2ca6fab83bcb3a7ce9cb17aac276d1d9890e3
Task-number: QTBUG-55117
Reviewed-by: Sean Harmer <sean.harmer@kdab.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
The normal lighting materials were behaving weirdly with cubes. Some
faces were behaving correctly whilst others were inverting the lit and
shadowed sides of normal mapped areas (the slopes). Tracked this down
to some inconsistencies in the generation of the cube geometry.
The cube geometry now generates data consistently such that:
* when you look at each face from the following locations and
orientations:
** Positive X face with positive y as up vector
** Negative X face with positive y as up vector
** Positive Z face with positive y as up vector
** Negative Z face with positive y as up vector
** Positive Y face with negative z as up vector
** Negative Y face with positive z as up vector
then:
* the u texture coordinate always increases to the right as you look at
any face of the cube
* the v texture coordinate always increases up the face
* the tangent vector is aligned with the u texture coordinate
and taking special care that the w component is correctly oriented.
This can be tested by using a normal mapped cube with a point light
orbiting around above it takign note of the regions that are lit and
shadowed.
Change-Id: I834749b08efb9e03548e8c749c6281958a4e6329
Reviewed-by: Kevin Ottens <kevin.ottens@kdab.com>
|
|
|
|
|
|
|
|
| |
Also comes with unit tests showing the bug and testing the basic
SortPolicy node behavior.
Change-Id: Ied522dcce4faaa7c6750c289f6aafb18f92ffc74
Reviewed-by: Sean Harmer <sean.harmer@kdab.com>
|
|
|
|
|
|
|
|
| |
This is in fact done at several places nowadays, sometimes with bugs, so
let's provide a single tested implementation for it.
Change-Id: I782ba18a15e5b4a3b320ac6b472173b1f8e417a3
Reviewed-by: Sean Harmer <sean.harmer@kdab.com>
|
|
|
|
|
|
|
|
|
| |
Also needed to properly specify namespace and class name for
signal arguments so they are picked up in old style connections.
Task-number: QTBUG-54711
Change-Id: Id752a3dd0703aa6bf02da4ec4252c6207fa60acb
Reviewed-by: Kevin Ottens <kevin.ottens@kdab.com>
|