Also fixed missing entry in CHANGELOG.md

TBR=ianh@google.com

Review URL: https://codereview.chromium.org/1108393003

TBR=ianh@google.com

Review URL: https://codereview.chromium.org/1104093004

This is just copy-pasted from shelldb but should work.

TBR=ianh@google.com

Review URL: https://codereview.chromium.org/1112433003

This is a fork of sky/tools/download_material_design_assets
which does not depend on depot_tools and assumes that its
right next to the assets directory instead of finding it
relative to the mojo root.

R=ianh@google.com

Review URL: https://codereview.chromium.org/1106373002

R=ianh@google.com

Review URL: https://codereview.chromium.org/1111573002

TBR=mpcomplete@chromium.org

Review URL: https://codereview.chromium.org/1106343002

R=ianh@google.com

Review URL: https://codereview.chromium.org/1107203003

R=eseidel@chromium.org

Review URL: https://codereview.chromium.org/1104233003

TBR=hansmuller

Review URL: https://codereview.chromium.org/1102393002

shelldb: Make shelldb analyze also analyze the packages, since that's where most development is actually happening at the moment.

R=eseidel@chromium.org

Review URL: https://codereview.chromium.org/1101893002

Add support for extra mojom dirs to deploy_sdk.py

TBR=blundell

Review URL: https://codereview.chromium.org/1102883003

This exposes the last hooks needed to implement flexbox layout. For now,
I didn't worry too much about the exact API we're exposing since this will
all change with the upcoming redesign (e.g. https://codereview.chromium.org/1093633002).

minContentWidth == the width if the element were to wrap at every wrapping point (not including border/padding)

maxContentWidth == the width if the element were to only wrap at hard wrapping points (e.g. \n inside a whitespace: pre).

R=eseidel@chromium.org

Review URL: https://codereview.chromium.org/1101793003

BUG=
R=eseidel@chromium.org

Review URL: https://codereview.chromium.org/1092793005

This tries to make the menu controller more careful about the state
transitions it goes through, in particular, not firing off animations
while we're already closing (since that used to trigger the "we're
done animating" callback which then closes the menu forcibly).

R=eseidel@chromium.org

Review URL: https://codereview.chromium.org/1074883005

R=ppi@chromium.org

Review URL: https://codereview.chromium.org/1098243003

R=eseidel@chromium.org

Review URL: https://codereview.chromium.org/1097343002

Sky doesn't need the git repo of the mojo repo.

R=eseidel@chromium.org

Review URL: https://codereview.chromium.org/1059113004

In MojoShell we never hit this path since the NetworkFetcher
aborts earlier on, not knowing what content_handler to use.

R=ojan@chromium.org, ianh@google.com

Review URL: https://codereview.chromium.org/1080993005

TBR=eseidel

Review URL: https://codereview.chromium.org/1065653009

remove members that are never read
remove imports that are never used

R=eseidel@chromium.org

Review URL: https://codereview.chromium.org/1099203002

- make the ScrollBehavior instance long-lived, rather than recreating
  it each time we update the list contents.
- have OverscrollBehavior track the total height of the contents and
  the height of the scrollable region, so that it can determine when
  to stop scrolling down.
- teach OverscrollBehavior about how to determine when to stop
  scrolling down, and how to bounce when it's too far down.
- replace the 'energy' concept in Particles with a method that sets
  the energy and direction at the same time, instead of assuming that
  the direction is always positive when setting energy.
- make FixedHeightScrollable lists track the number of items in the
  list and have them update their ScrollBehavior regarding this
  information as it changes.
- track how many items are currently showing in the list stock list.

R=eseidel@chromium.org

Review URL: https://codereview.chromium.org/1097373002

R=jamesr@chromium.org

Review URL: https://codereview.chromium.org/1100833002

R=qsr@chromium.org

Review URL: https://codereview.chromium.org/1093033002

[Effen] Only skip rows we're showing, when skipping past the rows that we've scrolled beyond in the stock list.

Currently, if you then scroll down N items with a filter set, we select which
stock to show by taking the entire list of stocks, skipping the first N, then
filtering the list, then selecting as many stocks as needed to fill the list.

So suppose the list is A, B, Cx, Dx, Ex, F, and the filter is "x", and you've
scrolled down 1 item.

Currently we'd show Cx, Dx, Ex.

Scroll down 1 again.

We still show Cx, Dx, Ex.

What we _should_ show is Dx, Ex if you've scrolled down 1, and just Ex if you've
scrolled two.

This patch fixes this. We now skip rows _after_ filtering. This fixes
the bug whereby if you set a filter then fling the list, we show the
same item over and over as if in a loop.

R=eseidel@chromium.org

Review URL: https://codereview.chromium.org/1057603006

TBR=eseidel

Review URL: https://codereview.chromium.org/1065643003

TBR=eseidel

Review URL: https://codereview.chromium.org/1066513006

Pipe the remaining time for an animation all the way out to the AnimatedValue.
Make splashes abortable, which causes them to continue to fade but no longer grow, by having them fade in 100ms or however long the animation was still going to go for, whichever is quickest.
Make Scrollables support objects registering with them to be told when scrolling happens.
Make UINode support subclasses being informed when _remove() was called.
Hook all that together to make splashes go away when scrolling by having them register with any ancestor Scrollables such that when those scroll, all the splashes get aborted. Unregister when removed.

R=eseidel@chromium.org

Review URL: https://codereview.chromium.org/1092423003

- the keyboard service now has a showByRequest() method that invokes Android's InputMethodManager's imm.showSoftInput with no flags set.
- the Input component calls through to the keyboard service's showByRequest() when it receives a pointer down.

R=eseidel@chromium.org

Review URL: https://codereview.chromium.org/1089163005

Turns out this is
actually three simulations in a row, so we now have a system that can
transition from one to another.

R=eseidel@chromium.org

Review URL: https://codereview.chromium.org/1100583002

TBR=eseidel

Review URL: https://codereview.chromium.org/1091163003

TBR=eseidel

Review URL: https://codereview.chromium.org/1070313005

R=eseidel@chromium.org, qsr@chromium.org

Review URL: https://codereview.chromium.org/1088793003

https://github.com/domokit/mojo/issues/98

R=eseidel@chromium.org

Review URL: https://codereview.chromium.org/1088263004

At present there is a special case in Dart bindings generation which
makes the Mojo SDK and things under mojo/services portable. However,
we must not require things under mojo/services to be next to the
Mojo SDK under mojo/public/... in client repos. Further, other
collections of mojoms should also be portable.

To fix this, instead of using the path in the host repo as the
canonical package: import URI for the generated Dart code, this CL
uses the mojom module name.

BUG=https://github.com/domokit/mojo/issues/73
R=blundell@chromium.org, johnmccutchan@google.com, tonyg@chromium.org

Review URL: https://codereview.chromium.org/1071693003

This reverts commit 9b03c575a4e12f80739d1de7d2c6d66a7cdb214a.
https://codereview.chromium.org/999253005

I've not seen the stale caching behavior.  I think we need
to debug that and fix it.  Living with out the cache
causes very long load times in SkyShell.

R=ianh@google.com

Review URL: https://codereview.chromium.org/1086233002

Hopefully this will make it slightly more clear that the sky_sdk
repository is not actually a stand-alone repository, but rather
generated by a script.

R=ianh@google.com

Review URL: https://codereview.chromium.org/1088223003

This will require me to roll the sky pub package, sigh.

R=ianh@google.com

Review URL: https://codereview.chromium.org/1090713002

https://github.com/domokit/sky_sdk/pull/2
https://github.com/domokit/sky_sdk/pull/3

R=sethladd@google.com, ianh@google.com

Review URL: https://codereview.chromium.org/1091613002

If your layout manager changes, then you need to setNeedsLayout,
otherwise the next frame doesn't even get scheduled. We already
correctly handle going from not having layout to having one or
vice versa by doing the more drastic reattaching since that
requires changing your actual render class.

R=eseidel@chromium.org

Review URL: https://codereview.chromium.org/1060013005

It was returing all element descendants instead of all element
immediate children.

R=eseidel@chromium.org

Review URL: https://codereview.chromium.org/1048333006