Previously, when asked to render aliased text with distance fields,
we would get the aliased glyph from the cache and then try to
anti-alias the edge. This change instead grabs the anti-aliased glyph,
then deliberately aliases the edge.
Bug: chromium:707979
Change-Id: I05766af17d7ae58bca27aaffd9e08e5c586e789c
Reviewed-on: https://skia-review.googlesource.com/21728
Commit-Queue: Jim Van Verth <jvanverth@google.com>
Reviewed-by: Brian Salomon <bsalomon@google.com>
Instead of plugging private fields from the factory, pass to ctor.
Change-Id: Iea186e394182bddb6a569de1a05f2e40d6815c80
Reviewed-on: https://skia-review.googlesource.com/21734
Reviewed-by: Mike Reed <reed@google.com>
Commit-Queue: Florin Malita <fmalita@chromium.org>
A recent run of the experimental VC++ /analyze builder gave this
warning:
skia\src\gpu\gl\grglgpu.cpp(851) : warning C6287: Redundant code: the left and right sub-expressions are identical.
The code it complained about was this:
if (width < 0 || width < 0) {
which was introduced in https://skia-review.googlesource.com/c/20445/
This change makes the obvious fix.
Change-Id: I9d1743293fac9dd15ed82cf85efef907d727921e
Reviewed-on: https://skia-review.googlesource.com/21620
Reviewed-by: Jim Van Verth <jvanverth@google.com>
Reviewed-by: Bruce Dawson <brucedawson@google.com>
Commit-Queue: Jim Van Verth <jvanverth@google.com>
The cache is scoped with the SkColorSpaceXformer object.
This ensures we're not transforming nodes with a degree > 1 multiple
times, and preserves the DAG topology.
Change-Id: I0b072cdac95f9f1c34e0565ed4f258aba986e1ae
Reviewed-on: https://skia-review.googlesource.com/21726
Commit-Queue: Florin Malita <fmalita@chromium.org>
Reviewed-by: Mike Reed <reed@google.com>
just an experiment to address performance when imagefilters are cached.
Bug: skia:
Change-Id: Ic1033c897d0a569b46a339fb3ae7f8f961882953
Reviewed-on: https://skia-review.googlesource.com/21395
Commit-Queue: Mike Reed <reed@google.com>
Reviewed-by: Florin Malita <fmalita@chromium.org>
Previously, SkGifCodec treated an error in the LZW data as incomplete,
since we can still draw the partially decoded image. But a client doing
incremental decodes needs to distinguish this from truly incomplete
data. In the case of an error, the client should not attempt to provide
more data and decode again.
Add kErrorInInput, and return it when SkGifCodec sees a fatal error.
Treat it the same as kIncompleteInput when it comes to filling and DM.
Bug: skia:6825
Change-Id: Ic6ce3a62c0b065ed34dcd8006309e43272a3db9f
Reviewed-on: https://skia-review.googlesource.com/21530
Commit-Queue: Leon Scroggins <scroggo@google.com>
Reviewed-by: Mike Reed <reed@google.com>
Reviewed-by: Chris Blume <cblume@chromium.org>
numImages is read from untrusted data, and as an unsigned short could
be up to 65,536. Avoid calling new with this number, which could result
in a crash if it pushes the device over the memory limit.
Change-Id: Ifff9e0ac6ade2b3d8584af656ea7d2f9eb4998e2
Reviewed-on: https://skia-review.googlesource.com/21269
Reviewed-by: Derek Sollenberger <djsollen@google.com>
Commit-Queue: Leon Scroggins <scroggo@google.com>
Fixes build of Skia lib when GR_TEST_UTILS=0
Makes GR_TEST_UTILS=0 for official builds
Makes "Mini" builder bot exercise building GPU with is_official_build=true
Bug: skia:6786
Change-Id: I6186683a3a216d2e779645bd9e8276a66bcff4d5
Reviewed-on: https://skia-review.googlesource.com/21524
Commit-Queue: Brian Salomon <bsalomon@google.com>
Reviewed-by: Brian Osman <brianosman@google.com>
When asking for help on more than one item,
e.g. foo -h me -h oh -h my just print Flags:
once.
TBR=scroggo@google.com
Change-Id: I67080c7390e6bacf4c36096304dd959229fdd07b
Reviewed-on: https://skia-review.googlesource.com/21660
Reviewed-by: Cary Clark <caryclark@skia.org>
Commit-Queue: Cary Clark <caryclark@skia.org>
Fuzzer data may generate path without a computable
length.
R=reed@google.com
Bug: 739190
Change-Id: I052540932937f24951ce66699080b8b959fb1a46
Reviewed-on: https://skia-review.googlesource.com/21500
Reviewed-by: Mike Reed <reed@google.com>
Commit-Queue: Cary Clark <caryclark@google.com>
Calculate fStemV and fCapHeight using SkPaint in SkPDFFont if needed.
Change-Id: Iaf762e6c44178fac659dfe9fa312d0caed9138ae
Reviewed-on: https://skia-review.googlesource.com/21535
Commit-Queue: Hal Canary <halcanary@google.com>
Reviewed-by: Ben Wagner <bungeman@google.com>
This is to support a Mock backend that passes non-rendering context unit tests and has minimal config support.
Change-Id: I57c3ad2d347659d14382a8a7cf53424633bd86c6
Reviewed-on: https://skia-review.googlesource.com/21534
Reviewed-by: Brian Osman <brianosman@google.com>
Commit-Queue: Brian Salomon <bsalomon@google.com>
Bug: skia:
Change-Id: Iddeeb91b378bdb61d200070d8faa3610299ab733
Reviewed-on: https://skia-review.googlesource.com/21533
Reviewed-by: Brian Osman <brianosman@google.com>
Reviewed-by: Brian Salomon <bsalomon@google.com>
Commit-Queue: Greg Daniel <egdaniel@google.com>
Bug: skia:
Change-Id: I641b206fc3bc19ac190ad94ee755ab9e1caab9b3
Reviewed-on: https://skia-review.googlesource.com/21341
Reviewed-by: Jim Van Verth <jvanverth@google.com>
Commit-Queue: Ethan Nicholas <ethannicholas@google.com>
... faster and more accurate than using SkTableColorFilter
todo: update blink after this lands
Bug:737981
Change-Id: I55b5c60dd23b9d2cbe9d60f83c74be1a8f3dcfcf
Reviewed-on: https://skia-review.googlesource.com/21368
Commit-Queue: Mike Reed <reed@google.com>
Reviewed-by: Brian Osman <brianosman@google.com>
Reviewed-by: Mike Klein <mtklein@google.com>
This makes it trivial to copy/paste into spreadsheets for sorting/diffing
Bug: skia:
Change-Id: I02c920e2b8be8f59270da9fb9bb3e6763987e0bc
Reviewed-on: https://skia-review.googlesource.com/21378
Reviewed-by: Mike Klein <mtklein@google.com>
Reviewed-by: Florin Malita <fmalita@chromium.org>
Commit-Queue: Mike Reed <reed@google.com>
We have removed support for drawing Index8, so stop testing it in DM.
Bug: skia: 6828
Change-Id: Ib2c4d3ebd371be704151a9f956c0ca2aaf2926a6
Reviewed-on: https://skia-review.googlesource.com/21525
Reviewed-by: Mike Reed <reed@google.com>
Commit-Queue: Leon Scroggins <scroggo@google.com>
Bug: skia:
Change-Id: I25d4f4eeb3d06f3e2b64aad652ef91743494755a
Reviewed-on: https://skia-review.googlesource.com/21528
Reviewed-by: Eric Boren <borenet@google.com>
Commit-Queue: Mike Reed <reed@google.com>
my tests run ~14% faster.
- Split out gradient shaders from image shaders. new compilation
unit: SkPDFGradientShader
- Common functions InverseTransformBBox and PopulateTilingPatternDict
moved to SkPDFUtils
- Split SkPDFShader::State into image and gradient structures.
- SkPDFCanon is now a simpler structure, with no logic of its own.
I am considering just moving all of its fields into SkPDFDocument
- SkPDFShader::State (the image/fallback shader) now is POD, making
the use of a hashmap for canonicalization straightforward.
Formerly, we used a linear search.
- Do not bother trying to canonicalize the falback image shader.
- SkPDFGradientShader::Key is not POD; comparison of two objects
requires looking at the contents of two variable-sized arrays.
We now pre-calculate the hash of the arrays using SkOpts::hash and
store a hash for the object in the fHash field.
Using that hash, we can now canonicalize using a hashmap instead
of a linar search!
- several static functions renamed to follow style guidelines
- stop using codeFunction function pointer; I find that less
clear than it could be.
- operator==() for SkPDFShader::State and SkPDFGradientShader::Key is
now much simpler and can now be inlined.
- SkArrayEqual template in SkPDFUtils.h
No change to PDF output.
BUG=skia:3585
Change-Id: I354ad1b600be6d6749abccb58d13db257370bc0b
Reviewed-on: https://skia-review.googlesource.com/21376
Reviewed-by: Ben Wagner <bungeman@google.com>
Commit-Queue: Hal Canary <halcanary@google.com>
Similar to SkShaderBase::makeContext(), catch this condition and bail
early.
BUG=chromium:738682
Change-Id: I4c7a2036bed8ab8699023c4f8f3bc2161de0c41d
Reviewed-on: https://skia-review.googlesource.com/21521
Reviewed-by: Herb Derby <herb@google.com>
Commit-Queue: Florin Malita <fmalita@chromium.org>
This reverts commit fe11e458b0.
Reason for revert: android roll failed, even with newer images:
junit.framework.AssertionFailedError: 80 dpi: totalDiffPixelCount is 2
at junit.framework.Assert.fail(Assert.java:50)
at android.graphics.drawable.cts.DrawableTestUtils.compareImages(DrawableTestUtils.java:193)
at android.graphics.drawable.cts.BitmapDrawableTest.compareOrSave(BitmapDrawableTest.java:629)
at android.graphics.drawable.cts.BitmapDrawableTest.verifyPreloadDensityInner(BitmapDrawableTest.java:598)
at android.graphics.drawable.cts.BitmapDrawableTest.testPreloadDensity(BitmapDrawableTest.java:556)
Original change's description:
> enable new tiling for android
>
> android roll will need to be coordinated w/
> https://googleplex-android-review.git.corp.google.com/#/c/2483600/
>
> Bug: skia:
> Change-Id: Iee6d6cd246f2e5b64f24440a17791821c8fb2f94
> Reviewed-on: https://skia-review.googlesource.com/21369
> Commit-Queue: Mike Reed <reed@google.com>
> Reviewed-by: Derek Sollenberger <djsollen@google.com>
> Reviewed-by: Mike Reed <reed@google.com>
TBR=djsollen@google.com,reed@google.com
Change-Id: Ic8cd056d0737a817d558cc1de9d90d8d334d0ec1
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Bug: skia:
Reviewed-on: https://skia-review.googlesource.com/21522
Reviewed-by: Mike Reed <reed@google.com>
Commit-Queue: Mike Reed <reed@google.com>
SkMatrix::setPolyToPoly() may fail to map degenerate values. Handle
this case explicitly, instead of assuming it never fails.
BUG=chromium:738746
TBR=
Change-Id: Ie1049b98f7e07ae5d6bdb706ba7b4a399388e5d8
Reviewed-on: https://skia-review.googlesource.com/21375
Reviewed-by: Florin Malita <fmalita@chromium.org>
Commit-Queue: Florin Malita <fmalita@chromium.org>
