This also moves the test for stocks into a new tests/examples/
directory, where we can put tests that test the examples.
TBR=abarth
Review URL: https://codereview.chromium.org/1182223004.
The underlying problem is that we lacked a RenderObjectWrapper for the
RenderView, which meant we couldn't handle changing the RenderObject that was
the root of the RenderView. This CL introduces a RenderViewWrapper and uses it
in a new AppContainer widget root. This change allows us to make App a
non-magical Component that is inserted into the AppContainer in the newly
introduced runApp function.
R=ianh@google.com
Review URL: https://codereview.chromium.org/1184823006.
This is a completion of Eric's WIP patch:
https://codereview.chromium.org/1179663005/
Low level support for creating a paragraph that contains
runs of styled text. The styles may be nested.
The Paragraph and RenderParagraph classes have been
replaced by Inline and RenderInline. Styled text is defined
with a tree of InlineText and InlineStyle objects.
R=abarth@chromium.org
Review URL: https://codereview.chromium.org/1177833012.
Make old users of RenderSizedBox use RenderConstrainedBox.
Change the semantics of BoxDecoration.apply* to actually apply the provided constraints to the current constraints, rather than the previous behaviour of merging them neutrally.
R=abarth@chromium.org
Review URL: https://codereview.chromium.org/1189603003.
- Fix the expectations file for stocks, since floating action button changed how it was being drawn.
- Fix edges theme data to not use a rounded rect path for canvas material.
- Add debugging hooks for TextStyle, and fix a warning.
TBR=eseidel
Review URL: https://codereview.chromium.org/1187483003.
Previously, the fn "Text" class rendered as a single block of text.
Now, there's a "Paragraph" class that contains "TextFragment"s, and "Text" is just a component that places a single TextFragment into a single Paragraph.
This allows other components, notably the Input component, to build more complicated paragraphs of styled text, without using 'display'.
R=eseidel@chromium.org
Review URL: https://codereview.chromium.org/1139123007
- add a checkbox to the stock app, so that we're testing the checkbox widget
(it's not currently wired up to anything, that can come later)
- make InkSplash use FlexContainer so that we can use flex in the popup menu items
- make effen's Text be more similar to Image and Container, so that it can be styled
- make layout.dart's RenderCSSText correctly support being styled
- also fixes a bug with the stock list where we were rendering one too few a row when scrolling
- check in the code to dump the DOM so I don't have to keep remembering how to do this
R=eseidel@chromium.org
Review URL: https://codereview.chromium.org/1134163003
This introduces a new kind of ContentNode similar to Style but which
instead of changing the styles that apply to the node, provides new
settings to apply to the "parentData" structure.
R=eseidel@chromium.org
Review URL: https://codereview.chromium.org/1122413006
We dropped Anchor support, and the markup changed a bit (mostly it got debug='' attributes to help debug output).
TBR=eseidel
Review URL: https://codereview.chromium.org/1129883004
Flights is not longer a good example of something you can build wity Sky.
Restore the flights-app-pixel test and move the assets needed for the test into
the tests directory.
R=ojan@chromium.org
Review URL: https://codereview.chromium.org/1034953003
Mostly just mechanical changes. The one surprising bit
is that RenderLayer no longer needs to explictly paint
outlines. I tested manually that before this patch,
the paintOutline call in RenderLayer was needed for
outlines on positioned elements and that after this
patch it's not.
R=esprehn@chromium.org
Review URL: https://codereview.chromium.org/878023002
This CL simplifies the sky_server to only map the build directory into /gen,
which will make the deploy script simpler. This CL updates all the imports to
use the /gen prefix when referring to generated files.
TBR=eseidel@chromium.org
Review URL: https://codereview.chromium.org/881093003
Commit efc3afd428bd85fd3d12e0dc941b5eb7248ca30b broke it because
we'd computed an empty foreground layer. Now that background
painting is part of the foreground phase, we need to use the
background rect for clipping. As best I can tell, the
background rect is always >= the size of the foreground
rect, so it should be safe to use.
R=abarth@chromium.org
Review URL: https://codereview.chromium.org/876243002
We now only preserve the whitespace inside a <t> element inside
the parser. This removes the known n^2 from reattaching whitespace
which should make parsing and appending nodes faster. I also
removed the dead WhitespaceMode code from the parser, and made
the dom-seralizer.sky auto indent the markup so the test output
would be readable.
R=abarth@chromium.org
Review URL: https://codereview.chromium.org/867963006
We were cloning elements with all the attributes that contained
expressions like attrName="{{ foo }}" which meant we'd go through the
reflection process converting that value and also call the
attrNameChanged() callback and the attributeChanged() with the braced
string which the element didn't really want to know about.
After this patch we create a "clone source node" which is a copy of the
original element without the attributes that have the expressions and
use that when cloning, then we assign the properties using data binding
later.
R=abarth@chromium.org, ojan@chromium.org
Review URL: https://codereview.chromium.org/868973002
Previously, if you had a generated background image and a translucent (or
rounded) border, we would draw the image nine times because we would inflate
the fill rect to draw behind the border. With no-repeat, we only draw the image
once.
This saves 4.5% of record time on flights-app.
R=esprehn@chromium.org
Review URL: https://codereview.chromium.org/844193009
Instead, always use InterpolationLow. In Skia, that boils down to bilinear,
which is fast on GPUs. We should eventually let authors control the
interpolation quality in case they want pixelation, etc.
This CL saves 6.3% on Layer::RecordPicture for flights-app.
R=esprehn@chromium.org
Review URL: https://codereview.chromium.org/860843006
All reflected attributes two way bind on SkyElement, so now doing
<sky-element name="sky-input" attributes="value:string"> is enough
to get two way binding on the value attribute so users doing
<sky-input value="{{ inputValue }}"> will get the inputValue property
updated as the user types.
R=abarth@chromium.org, ojan@chromium.org
Review URL: https://codereview.chromium.org/850383002
Now instead of calling addEventListener inside the created() callback you just
add on-{eventName} attributes to the <sky-element> around your element
definition. The attribute defines what method on your element to call when the
event with {eventName} is dispatched through your element.
Implementing this required refactoring how sky-element.sky stored information
from each registered element. Now instead of storing just a map of templates
we store a map of ElementRegistration objects which have metadata about the
element. Future patches will combine this system with the element registration
system in sky-binder.
R=ojan@chromium.org
Review URL: https://codereview.chromium.org/845283003
We were dumping pixels before the images loaded.
Instead, dig into the shadow root and only
dump the pixels once all the images have loaded.
This isn't an awesome solution for a real app,
but it's good enough to deflake this test since
all the images happen to be one shadow root deep.
R=esprehn@chromium.org
Review URL: https://codereview.chromium.org/792883008
Log an error whenever an element has an unknown attribute in a template.
This means you can't use generic attributes like Polymer, but we
probably want to discourage that anyway since attribute selectors should
be avoided for most things.
R=ojan@chromium.org
Review URL: https://codereview.chromium.org/845523004
This way we get some broad pixel test coverage.
Do this by dumping the png from the flights app,
uploading it to google storage and then having
the reference file load the uploaded file.
This isn't a good long-term solution. Long-term
we should dump paint commands (or some other
textual representation) and/or make more targetted
reftests that don't need pngs. As such, the process
for doing this is painful and manual. If we decide
we want to have pixel tests properly, we'll want
to do something more automated obviously.
Also, turn on the discard_transparency bool when
encoding the pngs. Otherwise, the png is lossy
and none of this works.
R=esprehn@chromium.org
Review URL: https://codereview.chromium.org/836363003
