difference in the position where the top border meets the right border (as
also the right border X bottom border).
The workaround to have a test that will work on both modes is to cover the
affected area with a white div. This patch adds inset test.
What can this do? Reset `<input type=text>` fields back to their default
value through a call to a form's reset method. That's all for now!
Fixes compile error after rebase
This patch provides some of the groundwork for column spans greater than
1. It implements the column-span CSS property (prefixed so as not to be
exposed to content) as well as the corresponding colspan attribute;
although the former is not well-specified outside of CSS multi-column
layout, INTRINSIC refers to it. Although width is distributed to
spanning columns, they do not yet contribute minimum and preferred
widths; this will be implemented in a follow-up.
Additionally, this patch cleans up some miscellaneous formatting issues
and improves the handling of table rowgroups.
Additionally, this patch cleans up some miscellaneous formatting issues
and refactors files in `layout/css/` somewhat to eliminate needless
levels of indirection. It also fixes our handling of presentational
hints that only apply if border is nonzero.
The exact rendering is ill-spec'd. Some things are ugly (especially the
width and height of list style images) but they are infrequently used
and I believe this implementation matches the spec. Numeric lists are
not supported yet, since they will require a separate layout pass.
The implementation is a subclass of `BlockFlow`, on advice from Robert
O'Callahan.
This property is used by approximately 55% of page loads.
To implement the line breaking behavior, the "breaking strategy" has
been cleaned up and abstracted. This should allow us to easily support
other similar properties in the future, such as `text-overflow` and
`word-break`.
This assumes that there are no ligatures that span across multiple
words. Since we have a per-word shape cache, this is a safe assumption
as of now. I have left comments to ensure that, if and when this is
revisted, we make sure to handle it properly.
I had to use a somewhat unconventional method of computing text
indentation (propagating from blocks down to inlines) because of the way
containing blocks are handled in Servo.
(As a side note, neither Gecko nor WebKit correctly handles percentages
in `text-align`, at least incrementally -- i.e. when the percentages are
relative to the viewport and the viewport is resized.)
Groove and Ridge rendering shows a solid border when color is black, that is broken and the current patch will implement a similar behavior as Firefox.
a) If the groove border is different from a black border (the bug that this
patch set fixes);
b) If a ridge border is different from a black border (samething);
c) If groove is different from ridge (they should be inverse of each other).
All 3 new tests passes on OSX Yosemite (10.10).
The ligature disabling code has been manually verified, but I was unable
to reftest it. (The only way I could think of would be to create an
Ahem-like font with a ligature table, but that would be an awful lot of
work.)
Near as I can tell, the method used to apply the spacing (manually
inserting extra advance post-shaping) matches Gecko.
`invert` is not yet supported.
Objects that get layers will not yet display outlines properly. This is
because our overflow calculation doesn't take styles into account and
because layers are always anchored to the top left of the border box.
Since fixing this is work that is not related to outline *per se* I'm
leaving that to a followup and making a note in the code.
The Unicode awareness of `text-transform` is implemented as well as
possible given the Rust standard library's Unicode support. In
particular, the notion of an alphabetic character is used instead of a
letter.
Gecko has a subclass of text run to handle text transforms, but I
implemented this in a simpler way.
Let's build this incrementally. I implemented a `Blob` that can hold a `DOMString`, and has `size` attribute and `slice(...)` method. I'll finish the rest in later PRs.
This exposed some problems in our clipping logic, which was never
properly rewritten for the stacking context reform. The clipping code
worked in terms of a stack of clips, but the new stacking context code
has no concept of a stack of clip regions. Fixing that in turn exposed
some flaky/incorrect tests:
* `borders` had an incorrect reference image, as far as I can tell.
* `negative_margins` had some stray pixels, fixed by changing the text.
