- From: Andrew Fedoniouk <news@terrainformatica.com>
- Date: Fri, 17 Aug 2012 13:59:07 -0700
- To: Andy Davies <dajdavies@gmail.com>
- Cc: www-style list <www-style@w3.org>
CSS is a declaration of rules being applied in static fashion. This rule: div.hidden { visibility:hidden; } means "*at any time* when div has class 'hidden' its 'visibility' is 'hidden'" (if it's not overridden by other rule of course ) But this declaration: @media all and (min-width: 1024px) { html { class: +"big-screen"; } } is not clear at all. Is it like above: "*at any time* 'class' attribute of html element has 'big-screen' at the end" or rather this: "when this rule applied first time its 'class' gets appended by 'big-screen'" ? If later one then we are entering "transactional CSS" field where CSS is allowed to handle events (rule-assignment is an event). Just in case, I've implemented such 'transactional' extension for the CSS named CSSS! - CSS script: http://www.terrainformatica.com/htmlayout/csss!.whtm http://www.terrainformatica.com/htmlayout/csss!-dom-object.whtm By now we have 3 years experience of using it - feature is pretty convenient for some simple tasks. But can be over-used. As any other CSS feature though. -- Andrew Fedoniouk. http://terrainformatica.com On Thu, Aug 16, 2012 at 1:45 PM, Andy Davies <dajdavies@gmail.com> wrote: > Hi All, > > Reading François post of the syntax for CSS Variables and the idea of > a my- prefix for custom properties can I throw another idea in the > ring? > > What if we wanted to alter HTML attributes from CSS, so for example at > different viewports we wanted to change a add another CSS class to the > HTML element, set some data- attributes on a widget etc. > > For example we might want to control the appearance of the tweet > button based on viewport something like this > > @media all and (max-width: 480px) { > .twitter-share-button { > data-size: "large"; > data-count: "none"; > } > } > > (data-size and data-count would be an example of something that could > really do with a prefix to clarify they aren't CSS properties.) > > At the moment the only way to do this is include JS that establishes > the viewport and sets the attributes, leading to duplication of > viewport dimensions in JS and HTML attributes being buried in JS. > > I wrote up a bit more about the idea here: > http://andydavies.me/blog/2012/08/13/what-if-we-could-use-css-to-manipulate-html-attributes/ > > As and idea it's undoubtable got issues, I think the concept is useful > and can see scenarios where it would be of use, question is does > anyone else and is it worth considering further? > > Cheers > > Andy >
Received on Friday, 17 August 2012 20:59:35 UTC