On 9/29/13 11:02 AM, "Dirk Schulze" <dschulze@adobe.com> wrote: >Hi, > >The specification allows to use images to calculate the shape outside. >Some images do not have intrinsic dimensions. It is currently not >specified which sizing algorithm should be used. CSS Images does define a >default sizing algorithm [1] which should be sufficient enough to use >IMO. I assume that the relevant object is the float. Yes, thanks! I think this helps out in quite a few cases where we otherwise would have a mismatch between the displayed image's size and the shape drawn from the image data. I've added in a sentence to section 3.3 pointing to the whole Concrete Object Size Resolution section [1] as I think we should be allowing for the cover and contain sizing steps as well. > >However, for <basic-shape>s the spec allows to use the 'box-sizing' >property to define the rectangle that is used for percentage resolving. >Do we want that for CSS Images too (which takes the padding box right now >IIRC)? Maybe it is not worth it to discuss that for the first level and >just referencing the default sizing algorithm is ok. I would expect box-sizing to contribute to the specified size of the element, so it would already be accounted for in the default sizing algorithm. Thanks, Alan [1] http://www.w3.org/TR/2012/CR-css3-images-20120417/#concrete-size-resolutionReceived on Monday, 30 September 2013 22:30:39 UTC
This archive was generated by hypermail 2.4.0 : Monday, 23 January 2023 02:14:32 UTC