- From: Andrew Fedoniouk <news@terrainformatica.com>
- Date: Tue, 14 Jun 2005 09:35:40 -0700
- To: "Ian Hickson" <ian@hixie.ch>
- Cc: "David Woolley" <david@djwhome.demon.co.uk>, <www-style@w3.org>
| On Mon, 13 Jun 2005, Andrew Fedoniouk wrote: | > | | > | What Andrew meant was "would allow". He's referring to | > | his own, currently rejected, proposal for a new type of unit. | > | > What David meant was "not accepted" rather | > than "currently rejected" :-P. | | Actually until a formal, clear, well-defined spec for what "%%" means is | sent to this list, it's not even a proposal. | | So far it has -- at best -- been described anecdotally. | I've already published this in the list not once: ------------------------------------------------ %% units are "percentage from free space". %% units are applicable to border, padding, margin, width and height CSS attributes only. UA will effectively compute them for all blocks in 'normal' flow. Thus width:100%% of floated block will not be interpreted at all (will be treated as undefined). Formal definition of %% length units: When allotting space among element attributes competing for space along axis, UA allot all values for attributes having other than %% lengths first, then divide up remaining available space among %% lengths. Each %% length receives a portion of the available free space that computed as a percentage from total sum of all %% units along axis. If total sum of all %% lengths along axis is less then 100 then 100%% value is used as a total sum of all %% units. Thus, if 100 pixels of space are available after the user agent allots pixel and percentage space, and the competing relative lengths are 10%%, 20%%, and 70%%, the 10%% will be alloted to 10 pixels, the 20%% will be alloted 20 pixels, and the 70%% will be alloted 70 pixels. If 100 pixels of space are available, and the competing relative lengths are 10%%, 20%%, and 30%%, the 10%% will be alloted to 10 pixels, the 20%% will be alloted 20 pixels, the 30%% will be alloted 30 pixels, and remaining 40 pixels will be left undistributed. ---------------------------------------------------------------- >From formal point of view "formal" definition at least is not worse than definition of multi-length units in current HTML spec[1]. Please consider this as a formal proposal. If needed I can provide more definition-by-examples. Andrew Fedoniouk. http://terrainformatica.com [1] http://www.w3.org/TR/html4/types.html#type-multi-length [2] http://lists.w3.org/Archives/Public/www-style/2004May/0084.html [3] http://lists.w3.org/Archives/Public/www-style/2004May/0037.html | -- | Ian Hickson U+1047E )\._.,--....,'``. fL | http://ln.hixie.ch/ U+263A /, _.. \ _\ ;`._ ,. | Things that are impossible just take longer. `._.-(,_..'--(,_..'`-.;.' |
Received on Tuesday, 14 June 2005 16:34:01 UTC