W3C home > Mailing lists > Public > www-style@w3.org > December 2013

[css-shapes] LC feedback

From: Sylvain Galineau <galineau@adobe.com>
Date: Thu, 5 Dec 2013 16:01:21 -0800
To: www-style list <www-style@w3.org>
CC: Alan Stearns <stearns@adobe.com>
Message-ID: <CEC653D1.1183D%galineau@adobe.com>
This post shares a few editorial items that came out of an internal thread with Alan. In order of sections:

3.1 Supported Shapes [1]

Very minor: for completeness I suggest being explicit about the resolution of percentage values for ellipse() with text such as:

# A percentage value here is resolved against the used width (for the rx
# value) and the used height (for the ry value) of the reference box.

4. Shapes from Image [2]

What happens when the image would require 2+ paths to enclose the specified pixels is currently undefined.

Imagine a simple image with, from top to bottom:
- A black fully opaque (alpha=1) rectangle
- A white fully transparent (alpha=0) rectangle
- Last, another black fully opaque rectangle.

What is the resulting float area for shape-image-threshold:0.5? Or, more specifically, how many float areas does this produce? Should:

A. The resulting path enclose both black rectangles and thus include the transparent white rectangle between them?
B. Or do we end up with one path for each of the black rectangles and adjoining inline elements are able to 'slide' into the area covered by the middle white transparent rectangle?

For Level 1 my initial inclination was to consider A optimal as covers the primary use-cases i.e. only one float area is produced per image. But it seems polygons may also produce such situations e.g. by positioning vertices outside the reference box?

6.1 The 'shape-outside' Property [3]

Bikeshed: I am unsure about 'auto' being the best name for 'using the margin-box as normal', in part because one can also specify shape-outside: margin-box and the difference is not very obvious at first (in the latter case, the outer margin box boundary is affected by border-radius). Would 'none' be better?

As for the explanation of what happens when only a <box> value is specified, I think it may help to be more specific e.g.:

# If a <box> value is specified by itself, it sets the float area to the
# outer edge of this box, including any border-radius curvature (see 5).


6.2 The 'shape-image-threshold' Property [4]

I think this section (or possibly section 4) should include a statement such as a the following:

# The absence of any pixels with an alpha value larger than the specified
# threshold results in an empty float area.


[1] http://dev.w3.org/csswg/css-shapes/#supported-basic-shapes

[2] http://dev.w3.org/csswg/css-shapes/#shapes-from-image

[3] http://dev.w3.org/csswg/css-shapes/#shape-outside-property

[4] http://dev.w3.org/csswg/css-shapes/#shape-image-threshold-property











Received on Friday, 6 December 2013 00:01:49 UTC

This archive was generated by hypermail 2.3.1 : Monday, 2 May 2016 14:39:17 UTC