AuroraMiddleware/skia2
1
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity
23,501 Commits 1 Branch 0 Tags 544 MiB
3f0e6945f8
Commit Graph

661 Commits

Author SHA1 Message Date
halcanary
d0c38315e8 SkPDF: bikeshed: use auto less 
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2234623002

Review-Url: https://codereview.chromium.org/2234623002
 2016-08-09 20:04:01 -07:00
halcanary
8eccc308c8 SkPDF: SkPDFFont organization changes. 
SkPDFFont:
  - SkPDFType1Font::populate() encode advances correctly.
  - break out logically independent code into new files:
    * SkPDFConvertType1FontStream
    * SkPDFMakeToUnicodeCmap
    SkPDFFont.cpp is now 380 lines smaller.
    Expose `SkPDFAppendCmapSections()` for testing.

SkPDFFontImpl.h
  - Fold into SkPDFFont.

SkPDFConvertType1FontStream:
  - Now assume given a SkStreamAsset

SkPDFFont:
  - AdvanceMetric now hidden in a anonymous namespace.

No public API changes.
TBR=reed@google.com

GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2221163002

Review-Url: https://codereview.chromium.org/2221163002
 2016-08-09 13:04:34 -07:00
halcanary
8b1d32c8d0 SkPDF/SkAdvancedTypefaceMetrics: simplify ATM, PDF takes over 
No public API changes.
TBR=reed@google.com

GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2222523003

Review-Url: https://codereview.chromium.org/2222523003
 2016-08-08 09:09:59 -07:00
mtklein
4e97607d9a Use sse4.2 CRC32 instructions to hash when available. 
About 9x faster than Murmur3 for long inputs.

Most of this is a mechanical change from SkChecksum::Murmur3(...) to SkOpts::hash(...).

BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2208903002
CQ_INCLUDE_TRYBOTS=master.client.skia:Test-Ubuntu-GCC-GCE-CPU-AVX2-x86_64-Release-SKNX_NO_SIMD-Trybot;master.client.skia.compile:Build-Ubuntu-GCC-x86_64-Release-CMake-Trybot,Build-Mac-Clang-x86_64-Release-CMake-Trybot

Review-Url: https://codereview.chromium.org/2208903002
 2016-08-08 09:06:28 -07:00
halcanary
9be372041e std::move(SkTDArray) 
Since we don't support MSVC2013 anymore, we can be more
succinct when defining move constructors of compound types.

GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2227673002

Review-Url: https://codereview.chromium.org/2227673002
 2016-08-08 07:21:42 -07:00
halcanary
462d0148a5 SkPDF: Stop using kHAdvance_PerGlyphInfo 
SkPDFFont:
- never request kHAdvance_PerGlyphInfo from typeface.
- set_glyph_widths() fn uses a glyph cache to get advances.
- stop expecting vertical advances that are never requested.
- composeAdvanceData() now non-templated
- appendAdvance() one-line function removed

SkPDFDevice:
- use a glyph cache for getting repeated advances.
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2219733004

Review-Url: https://codereview.chromium.org/2219733004
 2016-08-05 13:51:46 -07:00
halcanary
dabd4f0b79 SkPDF: PDFShader code modernized. 
Motivation:  reduce code complexity.

SkCanon stores SkPDFShader::State next to SkDFObject, not inside.

many places use sk_sp<T> rather than T* to represent ownership.

SkPDFShader::State no longer holds bitmap.

SkPDFShader::State gets move constructor, no longer heap-allocated.

Classes removed:
  SkPDFFunctionShader
  SkPDFAlphaFunctionShader
  SkPDFImageShader

BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2193973002

Review-Url: https://codereview.chromium.org/2193973002
 2016-08-03 11:16:57 -07:00
bungeman
38d909ec28 Move off SK_SUPPORT_LEGACY_DATA_FACTORIES. 
This moves Skia code off of SK_SUPPORT_LEGACY_DATA_FACTORIES.
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2206633004

Review-Url: https://codereview.chromium.org/2206633004
 2016-08-02 14:40:46 -07:00
msarett
c573a40ed5 Add drawImageLattice() and drawBitmapLattice() APIs 
The specified image/bitmap is divided into rects, which
can be draw stretched, shrunk, or at a fixed size.  Will be
used by Android to draw 9patch (which are acutally N-patch)
images.

BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=1992283002

Review-Url: https://codereview.chromium.org/1992283002
 2016-08-02 08:05:56 -07:00
msarett
c213f0df22 Hide (mostly) unused SkColorSpace writeToICC() function 
This originally was intended to be used for serialization, but
we've chosen another approach.

Includes a revert of:
https://codereview.chromium.org/2012133003/

BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2193223002

Review-Url: https://codereview.chromium.org/2193223002
 2016-08-01 14:23:32 -07:00
reed
f1ac18229c implement isABitmap for imageshader, return localmatrix for bitmap's impl 
For imageshader, I only return true if the image is explicitly raster-backed. I do not return true for texture, nor for generator (i.e. lazy/picture) backed. Is that ok?

BUG=skia:5592
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2197323002

Review-Url: https://codereview.chromium.org/2197323002
 2016-08-01 11:24:14 -07:00
halcanary
fa25106f02 SkPDF: PDFStream has-a not is-a PDFDict 
Motivation:
SkPDFStream and SkPDFSharedStream now work the same.

Also:
- move SkPDFStream into SkPDFTypes (it's a fundamental PDF type).
- minor refactor of SkPDFSharedStream
- SkPDFSharedStream takes unique_ptr to represent ownership

BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2190883003

Review-Url: https://codereview.chromium.org/2190883003
 2016-07-29 10:13:18 -07:00
halcanary
5abbb44f68 SkPDF: flip saveLayer rightside up 
Broken in https://skia.googlesource.com/skia/+/4b1e17e

BUG=632574
TBR=tomhudson@google.com
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2197623002

Review-Url: https://codereview.chromium.org/2197623002
 2016-07-29 08:41:33 -07:00
halcanary
00450f1ae5 SkPDF: drawText makes use of SkPaint.measureText() for alignment. 
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2188303003

Review-Url: https://codereview.chromium.org/2188303003
 2016-07-28 14:33:13 -07:00
halcanary
4b1e17edc7 SkPdf: SkPDFFormXObject de-class-ified. 
We don't need an object, just a few standard fields on the
base class; the change lets us get rid of a bunch of
boilerplate code.

I think this also reduces the cognitive load of the SkPDF
internals.

BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2185803003

Review-Url: https://codereview.chromium.org/2185803003
 2016-07-27 14:49:46 -07:00
halcanary
fe8f0e0d31 SkPDF: refactor font subset: fewer copies 
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2190643002

Review-Url: https://codereview.chromium.org/2190643002
 2016-07-27 14:14:04 -07:00
halcanary
ac0e00dcec SkPDF: SkPDFStream takes a unique_ptr 
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2188623004

Review-Url: https://codereview.chromium.org/2188623004
 2016-07-27 11:12:23 -07:00
halcanary
56586b6f3d SkPDF: re-work SkPDFUtils::FloatToDecimal 
* do a lot less floating-point math by converting to
    an integer as early as possible [faster].
  * round rather than truncate.
  * use 8 significant digits rather than 9 when possible.
  * remove trailing zeros in fractions.

before:
      0.12 !    PDFScalar   nonrendering
after:
      0.07 !    PDFScalar   nonrendering

Accuracy guaranteed by existing unit test.

Example diffs:

    -/Shading <</Function <</C0 [.321568638 .333333343 .321568638]
    +/Shading <</Function <</C0 [.32156864 .33333334 .32156864]

    -/C1 [.258823543 .270588248 .258823543]
    +/C1 [.25882354 .27058825 .25882354]

    -1 0 0 -1 20 120.394500 Tm
    +1 0 0 -1 20 120.394501 Tm

    -1 0 0 -1 20 184.789001 Tm
    +1 0 0 -1 20 184.789 Tm

    -291.503997 0 l
    +291.504 0 l

BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2146103004

Review-Url: https://codereview.chromium.org/2146103004
 2016-07-25 13:59:30 -07:00
reed
2c9e20055b remove fClipStack and attach/deattach-from-canvas 
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2180793002

Review-Url: https://codereview.chromium.org/2180793002
 2016-07-25 08:05:23 -07:00
robertphillips
1f3923e4e5 Deprecate SkDevice::accessBitmap method 
Relies on https://codereview.chromium.org/2162423003/ (Add SK_SUPPORT_LEGACY_ACCESSBITMAP Skia guard) landing in Chromium first.

Calved off: https://codereview.chromium.org/2163323002/ (Add desired width & height to drawContext (as opposed to using the width & height of the RT))

GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2168483003

Review-Url: https://codereview.chromium.org/2168483003
 2016-07-21 07:17:54 -07:00
reed
cf5c846b41 remove dead methods now that we use specials exclusively for imagefilters 
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2164763003

Review-Url: https://codereview.chromium.org/2164763003
 2016-07-20 12:28:40 -07:00
reed
a2217ef965 use special-image for imagefilters and save/restore layer 
add special virtuals to device, in preparation for using them instead of bitmap for imagefilters

BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2155933002

patch from issue 2155933002 at patchset 20001 (http://crrev.com/2155933002#ps20001)

use specialimages instead of bitmaps for imagefiltering

Review-Url: https://codereview.chromium.org/2155063002
 2016-07-20 06:04:34 -07:00
reed
e51c356ae4 pre-land special methods on device 
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2161233002

Review-Url: https://codereview.chromium.org/2161233002
 2016-07-19 14:33:20 -07:00
robertphillips
e34f17d236 Make SkFont a bit more useable 
Split out of https://codereview.chromium.org/2163483002/ (Use SkFont in GrStencilAndCoverTextContext)

GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2163633002

Review-Url: https://codereview.chromium.org/2163633002
 2016-07-19 07:59:22 -07:00
reed
7503d60847 move responsibility for creating raster-device into pdf 
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2146333004

NOTRY=True
win failure seems unrelated

Review-Url: https://codereview.chromium.org/2146333004
 2016-07-15 14:23:30 -07:00
halcanary
eb92cb3e84 SkPdf: smaller color serialization 
SkPDFUtils now has a special function (SkPDFUtils::AppendColorComponent)
just for writing out (color/255) as a decimal with three digits of
precision.

SkPDFUnion now has a type to represent a color component.  It holds a
utint_8, but calls into AppendColorComponent to serialize.

Added a unit test that tests all possible input values.
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2151863003

Review-Url: https://codereview.chromium.org/2151863003
 2016-07-15 13:41:28 -07:00
halcanary
f0c30f5566 SkPDF: Join Positioned Text 
When N sequential positioned glyphs differ in positions by exactly the
advances of the first (N-1) glyphs, join the glyphs into a string
rather than changing the text matrix between each glyph draw.

Decreases PDF output size by about ~1.4%.  Potentially more on
text-heavy pages.

A single-typeface PDF of an 27kB ASCII document shaped with harfbuzz:
        before:  187743 Bytes
         after:  65513 Bytes
    difference:  -65.1%

Before:
    BT
    /F0 13 Tf
    1 0 0 -1 143.5 61 Tm
    <0029> Tj
    1 0 0 -1 150.634765 61 Tm
    <004C> Tj
    1 0 0 -1 154.602050 61 Tm
    <0055> Tj
    1 0 0 -1 160.245117 61 Tm
    <0048> Tj
    1 0 0 -1 167.925781 61 Tm
    <004B> Tj
    1 0 0 -1 176.469726 61 Tm
    <0052> Tj
    1 0 0 -1 184.518554 61 Tm
    <0056> Tj
    1 0 0 -1 190.980468 61 Tm
    <0048> Tj
    ET

After:
    BT
    /F0 13 Tf
    1 0 0 -1 0 0 Tm
    143.5 -61 Td <0029004C0055> Tj
    16.7451171 0 Td <0048004B005200560048> Tj
    ET

Also: update the Text matrix with the `Td` operator, instead of
overwriting it with the the `Tm` operator.  In the worst case, when
every glyph is positioned differently than it's advance, this still
makes the command stream smaller:

Before:
    ...
    1 0 0 -1 58.328125 660 Tm <0055> Tj
    1 0 0 -1 61.609375 660 Tm <004C> Tj
    1 0 0 -1 63.828125 660 Tm <0056> Tj
    ...

After:
    ...
    3.140625 0 Td <0055> Tj
    3.28125 0 Td <004C> Tj
    2.21875 0 Td <0056> Tj
    ...

GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2150393002

Review-Url: https://codereview.chromium.org/2150393002
 2016-07-15 13:35:45 -07:00
halcanary
39f988eb41 SkPDF: Always give a default advance for glyphs. 
Prior to this we assumed the default default was 0.  It is
actually 1000, according to the spec.

fixes BUG=skia:5321
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2146403005

Review-Url: https://codereview.chromium.org/2146403005
 2016-07-15 12:54:30 -07:00
reed
cd4051e1c4 remove unused CreateInfo flag 
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2153173002

NOTRY=True

Review-Url: https://codereview.chromium.org/2153173002
 2016-07-15 09:41:27 -07:00
halcanary
a76a10b730 SkPDF: Fix encoding of unichr outside of basic plane 
In ToUnicode table, write unicode codepoints as one or two UTF16BE
values, rather than a single hex, as the standard requires.

Factor out uint16 -> big-endian hex code.

SkUtils is now a namespace.

BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2120533002

Review-Url: https://codereview.chromium.org/2120533002
 2016-07-07 12:31:55 -07:00
halcanary
492d6b5b06 SkPDF: fix PDFA-only metadata error 
Review-Url: https://codereview.chromium.org/2119883002
 2016-07-07 12:28:31 -07:00
halcanary
3c35fb3310 SkPDF: Glyph Useage Map improvements 
Instead of having a fFontGlyphUsage on each device and one on each
document, just have the one on the document, and never merge.

Make fGlyphUsage accesible on SkPDFDocument.

Remove SkPDFGlyphSetMap::merge, ::reset, and SkPDFGlyphSet::merge.

SkPDFGlyphSetMap has an TArray of SkPDFGlyphSet, not TDArray of
SkPDFGlyphSet pointers.  SkPDFGlyphSet and SkPDFBitset get move
constructors.

All tests produce exactly identical output PDFs.

BUG=skia:5434
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2112943002

Review-Url: https://codereview.chromium.org/2112943002
 2016-06-30 11:55:07 -07:00
halcanary
9f4b332f59 SkPDF: allow overriding Producer metadata 
I recommend not using this functionality.

Also, some documentation.

BUG=skia:5436

GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2074583003

Review-Url: https://codereview.chromium.org/2074583003
 2016-06-30 08:22:04 -07:00
halcanary
2633e62f77 SkPDF: always assume SkStreamAsset behaves as specified 
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2104223003

Review-Url: https://codereview.chromium.org/2104223003
 2016-06-29 07:23:47 -07:00
halcanary
29ed2ae2da SkPDF: SkPDFStream takes only SkStreamAsset 
This is possible since https://crrev.com/869763002. Later, I'll clean
up the logic and class constructors.  For now, I assert everywhere I
expect a SkStreamAsset to be well-behaved (duplcate, hasLength).
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2098393002

Review-Url: https://codereview.chromium.org/2098393002
 2016-06-29 06:31:32 -07:00
martina.kollarova
b8d6af169b Remove unnecessary includes in src/pdf/ 
Use forward declaration more.

BUG=None
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2110033002

Review-Url: https://codereview.chromium.org/2110033002
 2016-06-29 05:12:32 -07:00
halcanary
ee41b7556d SkPDF: alloc less memory for strings 
before:
        micros   	bench
        250.98  	WritePDFText	nonrendering
    after:
        micros   	bench
        107.10  	WritePDFText	nonrendering

Also, be slightly more space-efficient in encoding strings.

Also, add a bench.

GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2099463002

Review-Url: https://codereview.chromium.org/2099463002
 2016-06-23 14:08:11 -07:00
cabanier
e75cdcb85b SkPDF: Use type 2/3 shading for gradient shaders 
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=1925233003

Review-Url: https://codereview.chromium.org/1925233003
 2016-06-17 12:38:54 -07:00
halcanary
7b9eabb392 SkPDF: handle unpremul right 
.
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2037083002

Review-Url: https://codereview.chromium.org/2037083002
 2016-06-03 08:57:03 -07:00
halcanary
99e22fbe56 SkPDF/SkColorSpace: use writeToICC 
.
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2012133003

Review-Url: https://codereview.chromium.org/2012133003
 2016-05-26 08:31:07 -07:00
bungeman
13b9c95295 Move SkTypeface to sk_sp. 
Committed: https://skia.googlesource.com/skia/+/6296da736fbf40aae881650c239420f64e576c3f
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=1933393002

Review-Url: https://codereview.chromium.org/1933393002
 2016-05-12 10:09:31 -07:00
scroggo
9a9a7b29e5 Revert of Move SkTypeface to sk_sp. (patchset #5 id:80001 of https://codereview.chromium.org/1933393002/ ) 
Reason for revert:
fontmgr_iterAndroid failing to draw emoji. E.g. https://gold.skia.org/search2?blame=6296da736fbf40aae881650c239420f64e576c3f&unt=true&head=true&query=source_type%3Dgm

Original issue's description:
> Move SkTypeface to sk_sp.
>
> Committed: https://skia.googlesource.com/skia/+/6296da736fbf40aae881650c239420f64e576c3f

TBR=reed@google.com,fmalita@chromium.org,tomhudson@google.com,bungeman@google.com
# Skipping CQ checks because original CL landed less than 1 days ago.
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true

Review-Url: https://codereview.chromium.org/1974783002
 2016-05-12 06:22:30 -07:00
bungeman
6296da736f Move SkTypeface to sk_sp. 
Review-Url: https://codereview.chromium.org/1933393002
 2016-05-11 12:38:18 -07:00
halcanary
e20a875170 SkAdvancedTypefaceMetrics: abstract out linked list 
+ use SkSinglyLinkedList<T>
+ move SkSinglyLinkedList.h to core
+ remove SkHackyAutoTDelete
+ getAdvanceData() -> setGlyphWidths()
+ finishRange no longer templated
+ remove unused templated functions

GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=1953153004

Review-Url: https://codereview.chromium.org/1953153004
 2016-05-08 18:47:16 -07:00
halcanary
ebbdfe63a9 SkAdvancedTypefaceMetrics: improve robustness 
Manually unwind linked list.

BUG=567031
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=1955033002

Review-Url: https://codereview.chromium.org/1955033002
 2016-05-06 07:30:13 -07:00
halcanary
f8e8d5ee43 SkPDF/None: fix compilation 
TBR=stephana@google.com

Review-Url: https://codereview.chromium.org/1929593002
 2016-04-27 10:57:52 -07:00
reed
1e7f5e708e remove 'deprecated' region from SkDraw 
Most call-sites that used it just took its bounds, so it was trivial to convert them
to get the bounds of the RasterClip. Two clients wanted the actual region:
1. layeriter for android
2. pdf

Android already only has BW clips, so should be safe.
PDF now overrides its clip methods to ensure that all clips are BW.

BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=1925693002

Review URL: https://codereview.chromium.org/1925693002
 2016-04-27 07:49:17 -07:00
halcanary
4b6566644f SkDocument/PDF: new API 
has a pdf/a switch.
sets metadata in factory.

BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=1916093002

Review URL: https://codereview.chromium.org/1916093002
 2016-04-27 07:45:18 -07:00
halcanary
d51bdae4e1 SkStringPrintf and SkString::printf now are no longer limted by a static buffer 
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=1403803002

Committed: https://skia.googlesource.com/skia/+/606cadd5aac62299ef2e277709b3684cae2bf96c

Review URL: https://codereview.chromium.org/1403803002
 2016-04-25 09:25:35 -07:00
halcanary
3361e6248b Revert of SkStringPrintf and SkString::printf now are no longer limted by a static buffer (patchset #5 id:80001 of https://codereview.chromium.org/1403803002/ ) 
Reason for revert:
breaking something

Original issue's description:
> SkStringPrintf and SkString::printf now are no longer limted by a static buffer
>
> GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=1403803002
>
> Committed: https://skia.googlesource.com/skia/+/606cadd5aac62299ef2e277709b3684cae2bf96c

TBR=tomhudson@google.com,reed@google.com
# Skipping CQ checks because original CL landed less than 1 days ago.
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true

Review URL: https://codereview.chromium.org/1908423002
 2016-04-22 08:19:04 -07:00