davatron5000 / decss3 Goto Github PK

deCSS3 won't work on Firefox due to the inability to access document.styleSheets[i].cssRules from a cross browser script. this is a bit of a bummer. Hopefully there's a workaround.

The reason I ultimately gave up on mine was that selectors are way harder to remove than properties and values. Ultimately it would be cool if this could negate selectors...

li:nth-last-child(2) {
     font-weight: bold;
}

so I guess it would find that, and either somehow remove it or go through each property and negate it:

li:nth-last-child(2) {
     font-weight: normal;
}

Totally not anywhere near as comprehensive, but just to give credit where it's due, this is oldest one I've seen: http://intridea.com/2010/4/12/toggle-css3-bookmarklet?blog=company

Hey ya'll -- I tried to do this a while back as well but mine is way ghetto: https://github.com/chriscoyier/CSS3-StripTease

I'm thinking multiple backgrounds should be pretty easy to test for, just look for a comma in the value for background-image and there is one set background-image: none;

Security error
[Break On This Error] if ( ! stylesheet.cssRules ) {

deCSS3...6324741 (line 105)

I dragged the bookmark to my toolbar in Firefox 9 and clicked it, result was the above error.

the site is behind a firewall if that makes any difference.

Most of the commits on this are markdown edits. And that's just tragic. If anyone wants to help prune/rebase/merge those edit, let me know.

If just injecting CSS isn't too lame, this should get the last four No's you have listed:

animation

* { 
   -webkit-animation: none !important;
   -moz-animation: none !important;
}

background-clip

* {
   -webkit-background-clip: border-box   !important; 
   -moz-background-clip:    border-box   !important; 
   background-clip:         border-box   !important; 
}

background-origin

* {
   -webkit-background-origin: 0 0        !important;
   -moz-background-origin:    0 0        !important;
   background-origin:        0 0        !important;
}

background-size

* {
   -webkit-background-size: auto         !important; 
   -moz-background-size:    auto         !important; 
   background-size:         auto         !important;
}

Looks like we lost media query support. I tried to get it back but failed. This may be a unique function since it's not a replace, just a wipeout, it's just unsetting the rule entirely.

when i try this out on a page with css3 by clicking on the bookmark, it does degrade gracefully, but i expected it to get back to css3 when i click it again,

things which happen,

• icon fonts disappear
• flex slider(which was disabled using display:none) appears again.
• it doesn't degrade radial gradients, but i'm sure this is an issue you are working on.

On the whole, a great job. very useful tool and encourages me to use modernizr even more.

Browser: Chrome 16
OS: Windows 7

Just another thing to remove.

-webkit-border-image: none            !important;
-moz-border-image:    none            !important;
border-image:         none            !important;

-webkit-gradient(linear,left top,left bottom,from(#F4F4F4),to(#ECECEC));

our current search for /linear-gradient(.*?)/g isn't cutting it here.

After hearing about deCSS3 on ShopTalk I decided to take a look.
It's working great, however I have run into a little but which I figured I'd report (sorry if this sounds a little bit ramble-y for a GitHub issue, but I don't submit many of these things).

Basically when using the bookmarklet on pages of my site, the background (which consists of multiple gradients) screws up and stretches. This is illustrated perfectly when going to http://www.dev-hq.net/ and activating then de-activating the bookmarklet - the background compresses together (I've tested in Chrome). It's interesting to note that when going on to different pages of my website, for example http://www.dev-hq.net/5, the gradient gets stretched the opposite way.

Sorry I couldn't be more helpful and give some sort of test case in which I've found it screws up or find the root of the issue, but I'm a little pushed for time.

Resetting column-count may have to be on a detection basis. Setting * { column-count: 1; } can cause layout shifts when child elements are involved.

In the current version, I have all of the following and neither work so well:

• column-count: 1!important // causes problems with nested stuff
• column-count: none!important // doesn't override declared column-count
• column-span: all!important // doesn't do anything?

Any solutions, let me know.

according to a 3rd party report.

Suggested by Paul Irish, the ability to toggle off/on the bookmarklet would be pimp.

Not sure how feasible this is give the fact we're using deleteRule() in the CSS, but worth thinking about.

When executing the bookmarklet it fails to execute the JavaScript, since it's served as MIME text/plain. Please use rawgit URL instead.