Improve horizontal baseline detection.

Improve horizontal baseline detection.

The goal is to hint the baseline when hinting is possible, which is 
to say when the glyphs are be aligned with the pixel grid. The
current code has three shortcomings.

1. correctly snaps when the horizontal baseline is on the x-axis but
not when on the y-axis. Instead it is snapping the horizontal
baseline when there is y-skew.

2. tests against the full device matrix instead of the relaxed
matrix used by the scaler context.

3. has range issues when relaxing the matrix.

GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=1740163002

Committed: https://skia.googlesource.com/skia/+/27876bc1156bcdd15cac7f0fe6228eb7806b8e96

[bungeman-skia, 2016-02-26 16:04:53]

Description was changed from

==========
Don't snap baseline of unhinted text.

The current implementation of glyph placement snaps glyphs origins to
pixel aligned baselines when horizontal text would be axis aligned.
This is probably the right thing to do when hinting, but when the user
specifies not to hint the baselines should remain unhinted.
==========

to

==========
Don't snap baseline of unhinted text.

The current implementation of glyph placement snaps glyphs origins to
pixel aligned baselines when horizontal text would be axis aligned.
This is probably the right thing to do when hinting, but when the user
specifies not to hint the baselines should remain unhinted.
GOLD_TRYBOT_URL=
https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&is...
==========

[bungeman-skia, 2016-02-26 16:05:11]

The CQ bit was checked by bungeman@google.com to run a CQ dry run

[bungeman-skia, 2016-02-26 16:05:20]

Description was changed from

==========
Don't snap baseline of unhinted text.

The current implementation of glyph placement snaps glyphs origins to
pixel aligned baselines when horizontal text would be axis aligned.
This is probably the right thing to do when hinting, but when the user
specifies not to hint the baselines should remain unhinted.
GOLD_TRYBOT_URL=
https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&is...
==========

to

==========
Don't snap baseline of unhinted text.

The current implementation of glyph placement snaps glyphs origins to
pixel aligned baselines when horizontal text would be axis aligned.
This is probably the right thing to do when hinting, but when the user
specifies not to hint the baselines should remain unhinted.

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

[commit-bot: I haz the power, 2016-02-26 16:05:21]

Dry run: CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1740163002/1
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1740163002/1

[bungeman-skia, 2016-02-26 16:05:36]

Description was changed from

==========
Don't snap baseline of unhinted text.

The current implementation of glyph placement snaps glyphs origins to
pixel aligned baselines when horizontal text would be axis aligned.
This is probably the right thing to do when hinting, but when the user
specifies not to hint the baselines should remain unhinted.
GOLD_TRYBOT_URL=
https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&is...
==========

to

==========
Don't snap baseline of unhinted text.

The current implementation of glyph placement snaps glyphs origins to
pixel aligned baselines when horizontal text would be axis aligned.
This is probably the right thing to do when hinting, but when the user
specifies not to hint the baselines should remain unhinted.

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

[commit-bot: I haz the power, 2016-02-26 16:06:37]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2016-02-26 16:06:38]

Dry run: Try jobs failed on following builders:
  Build-Mac-Clang-Arm7-Debug-iOS-Trybot on client.skia.compile (JOB_FAILED,
http://build.chromium.org/p/client.skia.compile/builders/Build-Mac-Clang-Arm7...)
  Build-Mac-Clang-x86_64-Release-Trybot on client.skia.compile (JOB_FAILED,
http://build.chromium.org/p/client.skia.compile/builders/Build-Mac-Clang-x86_...)

[bungeman-skia, 2016-02-29 16:17:38]

Description was changed from

==========
Don't snap baseline of unhinted text.

The current implementation of glyph placement snaps glyphs origins to
pixel aligned baselines when horizontal text would be axis aligned.
This is probably the right thing to do when hinting, but when the user
specifies not to hint the baselines should remain unhinted.

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

to

==========
Improve horizontal baseline detection.

The goal is to hint the baseline when hinting is possible, which is 
to say when the glyphs are be aligned with the pixel grid. The
current code has three shortcomings.

1. correctly snaps when the horizontal baseline is on the x-axis but
not when on the y-axis. Instead it is snapping the horizontal
baseline when there is y-skew.

2. tests against the full device matrix instead of the relaxed
matrix used by the scaler context.

3. has range issues when relaxing the matrix.

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

[bungeman-skia, 2016-02-29 18:00:26]

bungeman@google.com changed reviewers:
+ herb@google.com, reed@google.com

[bungeman-skia, 2016-02-29 18:00:27]

https://codereview.chromium.org/1740163002/diff/1/src/core/SkScalerContext.cpp
File src/core/SkScalerContext.cpp (left):

https://codereview.chromium.org/1740163002/diff/1/src/core/SkScalerContext.cp...
src/core/SkScalerContext.cpp:836: if (0 == matrix[SkMatrix::kMSkewX]) {
This was incorrect, or at least doesn't do what the documentation says it does.
If checking for when the x axis is mapped to the y axis, the correct check would
have been '0 == matrix[SkMatrix::kMScaleX]'. This is actually checking if the y
axis stays the y axis (which would seem to be a job for a hypothetical
SkComputeAxisAlignmentForVText).

https://codereview.chromium.org/1740163002/diff/1/src/core/SkScalerContext.cpp
File src/core/SkScalerContext.cpp (right):

https://codereview.chromium.org/1740163002/diff/1/src/core/SkScalerContext.cp...
src/core/SkScalerContext.cpp:831: if (fRec.getHinting() != SkPaint::kNo_Hinting)
{
This is causing a lot of churn in gold because most of the gms end up using the
portable font manager, which always sets kNo_Hinting in onFilterRec. This may
argue that this test should be looking at the hinting request (from the paint)
as opposed to the hinting resolution (from the scaler context). It's hard to say
how to handle this at all when the primary font does not hint, but a fallback
font will, making things not necessarily line up. There will need to be some
more thinking on this, but for now let's fix the existing bugs.

https://codereview.chromium.org/1740163002/diff/1/src/core/SkScalerContext.cp...
src/core/SkScalerContext.cpp:832: if (0 == fRec.fPost2x2[1][0]) {
To answer the question, does this need to use getSingleMatrix? That multiplies
in fTextSize (a scale, which should have no effect on either test), fPreScaleX
(a scale in x which would have no effect), and fPreSkewX (which would also have
no effect). In other words, making the text bigger, stretching it along the
horizontal axis, or fake italicizing it does not move the baseline. This will be
made into a comment.

The one difference that does exist here (with or with out getSingleMatrix) is
that this is testing against fPost2x2 which is the sk_relax version of the
matrix. On the other hand, this is somewhat 'more correcter-er' for stating that
if hinting is happening then the baselines should be snapped.

[bungeman-skia, 2016-02-29 18:00:37]

The CQ bit was checked by bungeman@google.com to run a CQ dry run

[commit-bot: I haz the power, 2016-02-29 18:00:45]

Dry run: CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1740163002/40001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1740163002/40001

[commit-bot: I haz the power, 2016-02-29 18:16:03]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2016-02-29 18:16:04]

Dry run: This issue passed the CQ dry run.

[herb_g, 2016-02-29 18:37:10]

lgtm

https://codereview.chromium.org/1740163002/diff/40001/src/core/SkScalerContex...
File src/core/SkScalerContext.cpp (right):

https://codereview.chromium.org/1740163002/diff/40001/src/core/SkScalerContex...
src/core/SkScalerContext.cpp:837: // horizontal axis, or fake italicizing it
does not move the baseline.
italicizing; it does...

https://codereview.chromium.org/1740163002/diff/40001/src/core/SkScalerContext.h
File src/core/SkScalerContext.h (right):

https://codereview.chromium.org/1740163002/diff/40001/src/core/SkScalerContex...
src/core/SkScalerContext.h:254: /**
Do you want *'s aligned vertically?

[bungeman-skia, 2016-02-29 19:21:20]

https://codereview.chromium.org/1740163002/diff/40001/src/core/SkScalerContex...
File src/core/SkScalerContext.cpp (right):

https://codereview.chromium.org/1740163002/diff/40001/src/core/SkScalerContex...
src/core/SkScalerContext.cpp:837: // horizontal axis, or fake italicizing it
does not move the baseline.
On 2016/02/29 18:37:10, herb_g wrote:
> italicizing; it does...

it actually has no effect on the horizontal baseline (which is the implied
baseline, since this is For'H'Text.

https://codereview.chromium.org/1740163002/diff/40001/src/core/SkScalerContext.h
File src/core/SkScalerContext.h (right):

https://codereview.chromium.org/1740163002/diff/40001/src/core/SkScalerContex...
src/core/SkScalerContext.h:254: /**
On 2016/02/29 18:37:10, herb_g wrote:
> Do you want *'s aligned vertically?

So while we have a vertical bit on paint and we kinda need to support it for the
time being, we don't really support it everywhere. All the 'vertical' text in
Chrome and Android is actually positioned horizontal text.

[bungeman-skia, 2016-02-29 19:21:31]

The CQ bit was checked by bungeman@google.com

[commit-bot: I haz the power, 2016-02-29 19:21:43]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1740163002/40001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1740163002/40001

[commit-bot: I haz the power, 2016-02-29 19:22:58]

Description was changed from

==========
Improve horizontal baseline detection.

The goal is to hint the baseline when hinting is possible, which is 
to say when the glyphs are be aligned with the pixel grid. The
current code has three shortcomings.

1. correctly snaps when the horizontal baseline is on the x-axis but
not when on the y-axis. Instead it is snapping the horizontal
baseline when there is y-skew.

2. tests against the full device matrix instead of the relaxed
matrix used by the scaler context.

3. has range issues when relaxing the matrix.

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

to

==========
Improve horizontal baseline detection.

The goal is to hint the baseline when hinting is possible, which is 
to say when the glyphs are be aligned with the pixel grid. The
current code has three shortcomings.

1. correctly snaps when the horizontal baseline is on the x-axis but
not when on the y-axis. Instead it is snapping the horizontal
baseline when there is y-skew.

2. tests against the full device matrix instead of the relaxed
matrix used by the scaler context.

3. has range issues when relaxing the matrix.

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

Committed:
https://skia.googlesource.com/skia/+/27876bc1156bcdd15cac7f0fe6228eb7806b8e96
==========

[commit-bot: I haz the power, 2016-02-29 19:22:59]

Committed patchset #3 (id:40001) as
https://skia.googlesource.com/skia/+/27876bc1156bcdd15cac7f0fe6228eb7806b8e96