W3C home > Mailing lists > Public > www-style@w3.org > September 2012

Re: Discussing possibilities of a 'CSS-ignore' rule.

From: Markus Ernst <derernst@gmx.ch>
Date: Mon, 24 Sep 2012 12:01:49 +0200
Message-ID: <50602F8D.6050303@gmx.ch>
To: "Tab Atkins Jr." <jackalmage@gmail.com>
CC: Ketan Singh <singh.ketan7@gmail.com>, www-style@w3.org
Am 21.09.2012 22:40 schrieb Tab Atkins Jr.:
> On Thu, Sep 20, 2012 at 4:53 PM, Ketan Singh <singh.ketan7@gmail.com> wrote:
>> The tricky part started when I decided to insert the project's theme
>> 'header' and 'footer' into that popup, which required me to include the css
>> file on it as well. Now that the external css file is hooked up with with
>> the popup, I have kind of, lost control on the styling that was originally
>> generated by Microsoft word. Though most of the template markup has inline
>> definitions, still some parts are not showing up as expected.
>>
>> This makes me feel a need for something that could be called 'css-ignore'.
>> E.g. If I had a div with id 'my_div', I could have used something like
>> #my_div{  css-ignore:external;  } which could have asked the browser not to
>> associate any external style sheets with 'my_div'. The css-ignore rule could
>> have possible values such as 'inline', 'internal', 'external', which could
>> make the browser ignore css of a certain 'kind' for a particular selection
>> of elements.
>> I guess, there are many possible ways this can be achieved, many of them
>> kind of 'hackish' like tweaking it using jQuery; but having a proper
>> legitimate rule for it would certainly be helpful for many of us.
>> In addition, it can also be a great tool for developers, who might need it
>> to test their codes, in case they don't have a proper debugger.
>>
>> I'd like to apologize in advance in case something similar to this already
>> exists. I couldn't spare much time to research on this topic.
>> In case there isn't anything similar to this, I'd love to discuss how this
>> can be taken forward.
>
> This is a good use-case, and we have several ways coming down the
> pipeline of helping to address it!
>
> First, we've discussed (but not yet put into a draft) an "all"
> property, which only takes the values "initial" and "inherit".  It
> acts as a shorthand for *all* properties.  This lets you shut down
> inheritance from the outside world, resetting you to initial values
> the same as if it was in an independent document.
>
> However, that's not always enough.  It prevents inheritance from
> leaking values into your element, but the outer document's CSS might
> still accidentally target elements *inside* your element.  The Web
> Components spec being developed in the WebApps working group can
> address this.  It will allow you to write your popup as a component,
> and block all outside CSS from targeting elements inside of it - only
> the styles you define or link *inside* of the popup will apply.
>
> So, nothing needs to be done here - your use-case is already being
> addressed by other technologies we're developing inside the W3C.

Would all:initial override inline styles, too? In some cases it would be 
handy to have all inline styles ignored, which would be possible with 
Ketan's proposal.

Use cases:
- Include big bunches of legacy content (which often has lots of style 
attributes), but apply new styles
- Include HTML fragments from external sources (which may have style 
attributes) via AJAX
Received on Monday, 24 September 2012 10:02:34 GMT

This archive was generated by hypermail 2.3.1 : Tuesday, 26 March 2013 17:21:00 GMT