Re: [css-round-display] Percentages of 'polar-distance' when origin is not the center of the containing block

> On Jan 28, 2016, at 1:02 AM, Florian Rivoal <> wrote:
>>> On Jan 28, 2016, at 16:39, Jihye Hong <> wrote:
>>>> On Jan 28, 2016, at 4:06 PM, Florian Rivoal < > wrote:
>>>> On Jan 27, 2016, at 20:43, Jihye Hong <> wrote:
>>>> Sorry, I used confusing expression, 'circular' dependency.
>>>> What I want to explain was, using the #2, the calculated value of the
>>>> percentage polar-distance changes depending on the polar-angle value.
>>>> There are some usecases when all the elements in a containing block
>>>> have '50%' for polar-distance properties and the origin of polar
>>>> coordinates isn't center.
>>>> For #1 [1], all the calculated value of the percentage polar-distance
>>>> of elements are same because they have same percentage values.
>>>> But for #2 [2], the calculated distances between the each element and
>>>> the origin point are different.
>>>> I couldn't find any usage of percentage for the property's value
>>> which
>>>> has dependency on another properties.
>>>> If there exists cases, then #2 seems to be appropriate, but if not,
>>> #1
>>>> or another way can solve this problem.
>>> I see what you meant. Ultimately, we will need to deal with this
>>> dependency to take care of "polar-distance: ***% contain", so I don't
>>> think it makes a huge difference.
>>> Speaking of which what use cases do we have for using polar-distance
>>> with a percentage and not using contain? I'm wondering if we should
>>> make "contain" the default, or possibly the only behavior for
>>> percentages.
>> There is an usecase about using "polar-distance: 100%".
>> As you can see in the main screen UI of a page[1], there are buttons for
>> calls, messages, and mail.
>> And there are rounded text boxes to notify the number of calls, messages,
>> and mail at upper-right part of the buttons.
>> Those text boxes can be positioned with "polar-distance: 100%; polar-angle:
>> 30deg" when their containing blocks are the buttons.
>> In this case, the polar-distance with a percentage without contain seems
>> useful.
>> [1]
> Thanks. I see, it's a good example. That leaves the question of which one
> should be the default behavior open, but being able to support both use cases
> does make sense.

That example could also be accomplished by using a containing circle that is large enough to enclose the notification indicator, but would itself have no background or border. Then, for the gray circle, you could use an centered image background or centered ::before  pseudo, smaller than that container. 

Received on Sunday, 31 January 2016 18:10:01 UTC