Web fonts are quite common in Lottie text assets. Expand the
ResourceProvider interface to allow loading of external (web) font data.
Fonts are passed back to Skottie as opaque SkData, and instantiated
using SkFontMgr::makeFromData().
Change-Id: Ia4c994bd60e6a3bd47c90172c267ed5f897a38e4
Reviewed-on: https://skia-review.googlesource.com/155601
Commit-Queue: Florin Malita <fmalita@chromium.org>
Reviewed-by: Ben Wagner <bungeman@google.com>
Use sk_ieee_float_divide to silence UBSAN.
(note: the result is already checked to be finite)
Bug: skia:10041
Change-Id: I97d1d6f84926b7162ed636be7afd052ca42d8e81
Reviewed-on: https://skia-review.googlesource.com/155165
Reviewed-by: Kevin Lubick <kjlubick@google.com>
Commit-Queue: Florin Malita <fmalita@chromium.org>
Replace sksg::Group::shrink_to_fit() with a factory accepting an
externally-built children array.
This provides more control for clients (e.g. reserve() instead of
shrink_to_fit()).
Change-Id: Iad587435e0e9da15251a9d3bc2510ca945950b5d
Reviewed-on: https://skia-review.googlesource.com/152861
Commit-Queue: Florin Malita <fmalita@chromium.org>
Reviewed-by: Mike Klein <mtklein@google.com>
Pulling this off requires deferring the group creation until after we've
resolved all the draws. Doable, but somewhat tricky due to the
interaction with the dangling/uncommitted animator logic.
TBR=
Bug: skia:8340
Change-Id: Id00c841152bd80330751db45f6b26462efc844a5
Reviewed-on: https://skia-review.googlesource.com/152860
Reviewed-by: Florin Malita <fmalita@chromium.org>
Commit-Queue: Florin Malita <fmalita@chromium.org>
Skia's containers grow too aggressively, stick to STL containers for
reduced head overhead.
Also introduce Group::shrink_to_fit() to be used when clients know they
are done adding children to a group.
TBR=
Bug: skia:8340
Change-Id: I842db2e80c7cfb65a462e4859af6e63496863a5c
Reviewed-on: https://skia-review.googlesource.com/151161
Reviewed-by: Florin Malita <fmalita@chromium.org>
Commit-Queue: Florin Malita <fmalita@chromium.org>
Only initialize a default font manager when needed.
Change-Id: I3473804d426f1f4981f3727d3539303fb670cdd1
Reviewed-on: https://skia-review.googlesource.com/150840
Reviewed-by: Ben Wagner <bungeman@google.com>
Commit-Queue: Florin Malita <fmalita@chromium.org>
SkTArray is not a good choice for tigh allocations:
- its minimum allocation count is 8
- reserve(sz) doesn't adjust capacity to the exact value, but padds
the same way as normal growth (max(sz, 8, 50%))
- no shrink_to_fit() function to trim unneeded capacity
Since keyframed properties with a small number of frames are quite
common in Lottie, this adds significant heap overhead.
Switch to std::vector(), reserve() to the estimated frame count and
shrink_to_fit() when done parsing.
Bug: skia:8340
Change-Id: Id575e2da2fd17537948c2b38485a8accdb9f7a8b
Reviewed-on: https://skia-review.googlesource.com/150905
Reviewed-by: Mike Klein <mtklein@google.com>
Commit-Queue: Florin Malita <fmalita@chromium.org>
The -Lottie bot runs with --nonativeFonts so, for some animations that
request a specific font (e.g. Helvetica), this would fail, but not
fallback to an actual font manager.
This makes sure we at least have some text, even if it's the "wrong"
font.
Bug: skia:8298
Change-Id: Ibce7fda5b0a82fc3a708ab1cf5a7a394484ca89e
Reviewed-on: https://skia-review.googlesource.com/150468
Reviewed-by: Florin Malita <fmalita@chromium.org>
Reviewed-by: Ben Wagner <bungeman@google.com>
Commit-Queue: Ben Wagner <bungeman@google.com>
In the presence of time remapping, time start/stretch values appear to
be ignored.
TBR=
Change-Id: I6a4dcc107f2d0fc6361f91ba4f4643e33da37d93
Reviewed-on: https://skia-review.googlesource.com/149701
Commit-Queue: Florin Malita <fmalita@chromium.org>
Reviewed-by: Florin Malita <fmalita@chromium.org>
-- introduce a new animatable value (TextValue)
-- introduce a new adapter (TextAdapter) to translate Lottie text props to SG text props
-- use existing animated property-bind machinery and the new constructs when parsing text layers
Change-Id: Ibbfb69daf5b0a3c9a5ce8d1ccdeedca5b5d0fa6f
Reviewed-on: https://skia-review.googlesource.com/149266
Reviewed-by: Ben Wagner <bungeman@google.com>
Commit-Queue: Florin Malita <fmalita@chromium.org>
Constant keyframes currently requiere an "h" property/marker.
In order to support keyframed text parsing, relax this to consider any
keyframe missing an end value as constant.
TBR=
Change-Id: Idf332dc0174ee0ee6773708518a0bbc284f8860f
Reviewed-on: https://skia-review.googlesource.com/149265
Reviewed-by: Florin Malita <fmalita@chromium.org>
Commit-Queue: Florin Malita <fmalita@chromium.org>
Plumb AnimationBuilder throught a bazillion layers to make it reachable
when parsing animatable values.
This is in preparation of keyframed text, which will require access to
the font set when parsing.
Refactor only, no functional changes.
TBR=
Change-Id: Ide2ef2ba66fbcc75fdcc785f987b364d45dff5b6
Reviewed-on: https://skia-review.googlesource.com/149264
Commit-Queue: Florin Malita <fmalita@chromium.org>
Reviewed-by: Florin Malita <fmalita@chromium.org>
Update the ResourceProvider interface to inherit from SkRefCnt, to
clarify sharing/ownership semantics in the Skottie Builder API. Now it
follows the same pattern as SkFontMgr.
Change-Id: I7ff8ad39023d9ecfe609e0180b5aabf776672d48
Reviewed-on: https://skia-review.googlesource.com/148991
Reviewed-by: Ben Wagner <bungeman@google.com>
Commit-Queue: Florin Malita <fmalita@chromium.org>
Allow embedders to pass a font manager.
In order to avoid excessive factory API clutter, introduce an
Animation::Builder helper to wrap factory options.
Also clean up various bits:
* hoist scene parsing out of the Animation ctor
* store the animation duration explicitly (instead of unused fps)
* plumb const SkFontMgr& internally (instead of a ref)
Change-Id: I3e180dfa85ba18c8462cfeb5a7385bef985ed6c4
Reviewed-on: https://skia-review.googlesource.com/148800
Reviewed-by: Ben Wagner <bungeman@google.com>
Commit-Queue: Florin Malita <fmalita@chromium.org>
* introduce AnimationBuilder to hold mostly immutable (modulo caching)
state
* split the scoped animator state into AnimatorScope
This will facilitate splitting the monolithic Skottie.cpp in follow-up
CLs.
Refactoring only, no functional changes.
TBR=
Change-Id: I0a8295e60be4559586fc4a9fea3dee4a7f5714d4
Reviewed-on: https://skia-review.googlesource.com/148390
Reviewed-by: Florin Malita <fmalita@chromium.org>
Commit-Queue: Florin Malita <fmalita@chromium.org>
The "closed" shape property cannot be interpolated -- so ensure we catch
different values in interpolation pre-checks.
Bug: skiai:8264
Change-Id: If2c7e09c1227b0013acba3833c314e0646715d52
Reviewed-on: https://skia-review.googlesource.com/147967
Reviewed-by: Kevin Lubick <kjlubick@google.com>
Commit-Queue: Florin Malita <fmalita@chromium.org>
Still loads to do. For now just draws trivial text nodes using a default fontmgr.
Change-Id: I7343b648726d2c4f376f43437f6ae1377ad8ba86
Reviewed-on: https://skia-review.googlesource.com/147465
Reviewed-by: Ben Wagner <bungeman@google.com>
Commit-Queue: Florin Malita <fmalita@chromium.org>
Use the SkJSON conversion idiom instead of ParseDefault<SkString>.
Change-Id: Ieeadb86891602eaef2ecf5b68a4eb17b4897e8f7
Reviewed-on: https://skia-review.googlesource.com/147103
Commit-Queue: Florin Malita <fmalita@chromium.org>
Reviewed-by: Mike Klein <mtklein@google.com>
Current impl can underflow int due to unchecked arithmetic.
Add a Parse<size_t> specialization and convert call sites which use
inline arithmetic. Underflowing unsigned types should be well defined
and caught in later tests.
Bug: oss-fuzz:9798
Change-Id: Iaebe8aad4009e2511fe1d8733d336f5f119bb384
Reviewed-on: https://skia-review.googlesource.com/146648
Reviewed-by: Kevin Lubick <kjlubick@google.com>
Commit-Queue: Florin Malita <fmalita@chromium.org>
- plumb a RenderContext RenderNode::render() argument to track pending
opacity
- defer opacity application until we can determine whether a layer is
required (group with multiple children) or the opacity can be pushed
onto the draw paint (for single/atomic draws)
Bug: skia:
Change-Id: I962ba08bad780395d5d738307bde986e9efa502b
Reviewed-on: https://skia-review.googlesource.com/146445
Reviewed-by: Mike Reed <reed@google.com>
Commit-Queue: Florin Malita <fmalita@chromium.org>
* reserve SkPath storage to avoid incremental allocations
* refactor eval() to avoid copying data when interpolation is not
triggered
Change-Id: I467affbbfd652e8fa2a486acab7d6c7b9165d49f
Reviewed-on: https://skia-review.googlesource.com/146166
Reviewed-by: Mike Reed <reed@google.com>
Commit-Queue: Florin Malita <fmalita@chromium.org>
Overwrite the layer content color with a color filter.
TBR=
Change-Id: I39f920225affb2641cc11ab1f0c1456d89b47cb7
Reviewed-on: https://skia-review.googlesource.com/145730
Reviewed-by: Florin Malita <fmalita@chromium.org>
Commit-Queue: Florin Malita <fmalita@chromium.org>
... and funnel all existing factories through the new one.
Change-Id: I01ffb95abf178eacc0ad430e730d680800a509c7
Reviewed-on: https://skia-review.googlesource.com/145428
Reviewed-by: Mike Reed <reed@google.com>
Commit-Queue: Florin Malita <fmalita@chromium.org>
Similar to saving PNG frames, now we can save SKPs.
TBR=
Change-Id: I5791b564a1d3e70424e45e62034e559e677795f1
Reviewed-on: https://skia-review.googlesource.com/141320
Reviewed-by: Joe Gregorio <jcgregorio@google.com>
Commit-Queue: Florin Malita <fmalita@chromium.org>
When all matrix properties are static/default, we simply discard the
matrix node. But we cannot do that in the presence of a parent matrix.
TBR=
Change-Id: I56c62216f18786249dea063690261bfae83fabec
Reviewed-on: https://skia-review.googlesource.com/141127
Reviewed-by: Florin Malita <fmalita@chromium.org>
Commit-Queue: Florin Malita <fmalita@chromium.org>
The adapter apply() method can get called before stops resolution, so
only log when some stops are present.
TBR=
Change-Id: I8e336e2cff781a0e64de31e6b63f3cf373b2daa0
Reviewed-on: https://skia-review.googlesource.com/140245
Reviewed-by: Florin Malita <fmalita@chromium.org>
Commit-Queue: Florin Malita <fmalita@chromium.org>
Some tranform properties are not optional, but many of them are static
identity. Detect these cases and don't attach SG nodes for them.
Reduces the number of sksg::Matrix nodes by ~18%.
TBR=
Change-Id: Ia51c865d6928e8c48c73b30f6d45541caf334880
Reviewed-on: https://skia-review.googlesource.com/140186
Reviewed-by: Florin Malita <fmalita@chromium.org>
Commit-Queue: Florin Malita <fmalita@chromium.org>
We already have a clip optimization when there is only one opaque mask.
Extend to cover multiple opaque masks, using Merge scene nodes.
TBR=
Change-Id: I24b61f0c0d080b13438c6777e98a8e2fefd09fdd
Reviewed-on: https://skia-review.googlesource.com/140002
Reviewed-by: Florin Malita <fmalita@chromium.org>
Commit-Queue: Florin Malita <fmalita@chromium.org>
The fill type is not reflected in shape keyframes, but it is tracked in
the scene graph. Make sure we don't overwrite when interpolating.
TBR=
Change-Id: I281a61c3f4181ce34f772b36b83d0c7a64e265f6
Reviewed-on: https://skia-review.googlesource.com/139176
Reviewed-by: Florin Malita <fmalita@chromium.org>
Commit-Queue: Florin Malita <fmalita@chromium.org>
- plumb the framerate in AttachContext (needed for remapping)
- expand CompTimeMapper to handle time remapping (in addition to
time start, time stretch)
TBR=
Change-Id: If33bae7b4fe224f45d5a094a47899b9025827991
Reviewed-on: https://skia-review.googlesource.com/138990
Reviewed-by: Florin Malita <fmalita@chromium.org>
Commit-Queue: Florin Malita <fmalita@chromium.org>
We currently drop static split-position adapters without commiting the
value to the scene graph.
TBR=
Change-Id: Id5852da35345fa58ecaf6c752456ecb32e5b405a
Reviewed-on: https://skia-review.googlesource.com/136782
Reviewed-by: Florin Malita <fmalita@chromium.org>
Commit-Queue: Florin Malita <fmalita@chromium.org>
Minor cleanup: BindProperty reports whether a non-default property was
applied or bound for animation -- use this mechanism to detect
fully-opaque masks.
TBR=
Change-Id: I3bd9429842621309d0c6bd966ef748917e498c66
Reviewed-on: https://skia-review.googlesource.com/136623
Reviewed-by: Florin Malita <fmalita@chromium.org>
Commit-Queue: Florin Malita <fmalita@chromium.org>
We currently apply start-time/stretch-time adjustments to the referenced
composition AND to the referencing layer local properties. That last
part is incorrect: the adjustment should only apply to the referenced
composition.
Introduce a specialized composition time mapper to handle t adjustments,
and push the logic down to AttachCompLayer (and out of the generic
AttachLayer).
TBR=
Change-Id: I0ddb86232010a8e7cdac6524aef2eea5823e306d
Reviewed-on: https://skia-review.googlesource.com/136166
Reviewed-by: Florin Malita <fmalita@chromium.org>
Commit-Queue: Florin Malita <fmalita@chromium.org>
Some BM versions wrap point values as single or even multi-element arrays.
Parse<SkScalar> already handles the former case - we can extend that
behavior to also cover the latter and simplify Parse<SkPoint>.
TBR=
Change-Id: I2152928944f43dc03a5d8f0d65865ac43974fd7a
Reviewed-on: https://skia-review.googlesource.com/135800
Reviewed-by: Florin Malita <fmalita@chromium.org>
Commit-Queue: Florin Malita <fmalita@chromium.org>
It's a tiny, core-ish component -- might as well treat as such to
simplify dependencies.
Change-Id: I6f31ce2d151f9a629d88bfc7f15d64891d5150c0
Reviewed-on: https://skia-review.googlesource.com/135780
Reviewed-by: Mike Klein <mtklein@google.com>
Reviewed-by: Kevin Lubick <kjlubick@google.com>
Commit-Queue: Florin Malita <fmalita@chromium.org>
Use std::swap instead. It does not appear that any external user
specializes SkTSwap, but some may still use it. This removes all use in
Skia so that SkTSwap can later be removed in a smaller CL. After that
the <utility> include can be removed from SkTypes.h.
Change-Id: If03d4ee07dbecda961aa9f0dc34d171ef5168753
Reviewed-on: https://skia-review.googlesource.com/135578
Reviewed-by: Hal Canary <halcanary@google.com>
Reviewed-by: Mike Klein <mtklein@google.com>
Commit-Queue: Ben Wagner <bungeman@google.com>
This reverts commit 2a2f675926.
Reason for revert: this appears to be what is holding up the Chrome roll.
Original change's description:
> SkTypes: extract SkTo
>
> Change-Id: I8de790d5013db2105ad885fa2683303d7c250b09
> Reviewed-on: https://skia-review.googlesource.com/133620
> Reviewed-by: Mike Klein <mtklein@google.com>
TBR=mtklein@google.com,halcanary@google.com
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Change-Id: Iafd738aedfb679a23c061a51afe4b98a8d4cdfae
Reviewed-on: https://skia-review.googlesource.com/134504
Reviewed-by: Hal Canary <halcanary@google.com>
Commit-Queue: Hal Canary <halcanary@google.com>
The constness on null_provider causes:
"default initialization of an object of const type
'const NullResourceProvider' without a user-provided default constructor"
See https://stackoverflow.com/a/47368753 for more.
TL;DR; clang 3.8 and older is not happy
Bug: skia:
Change-Id: Icfc38680163c3bd4952c0e35551706cad8dbfee6
Reviewed-on: https://skia-review.googlesource.com/131521
Commit-Queue: Kevin Lubick <kjlubick@google.com>
Commit-Queue: Florin Malita <fmalita@chromium.org>
Reviewed-by: Florin Malita <fmalita@chromium.org>
Most of the existing clients don't care about nested resources.
Change-Id: Ie7991dd25ebbd679b5b49e5624772c7e19e7ec79
Reviewed-on: https://skia-review.googlesource.com/131141
Reviewed-by: Mike Reed <reed@google.com>
Commit-Queue: Florin Malita <fmalita@chromium.org>
Update the script to search for headers in both 'sources' and 'public'.
Change-Id: I195c6e3720f3d3d99dea04628388821a58fa791b
Reviewed-on: https://skia-review.googlesource.com/130823
Reviewed-by: Mike Klein <mtklein@google.com>
Reviewed-by: Ben Wagner <bungeman@google.com>
Commit-Queue: Florin Malita <fmalita@chromium.org>
