W3C home > Mailing lists > Public > www-style@w3.org > December 2010

RE: background-transform (Was: Re: [css3-images] Repeating oblique gradients)

From: Rik Cabanier <cabanier@adobe.com>
Date: Tue, 7 Dec 2010 21:08:24 -0800
To: Sylvain Galineau <sylvaing@microsoft.com>, Simon Fraser <smfr@me.com>
CC: "Tab Atkins Jr." <jackalmage@gmail.com>, Leif Arne Storset <lstorset@opera.com>, Brad Kemper <brad.kemper@gmail.com>, www-style list <www-style@w3.org>
Message-ID: <8A13F0222395BD428969E5BA529EFA747766F57EFC@NAMBX01.corp.adobe.com>
So, what you're proposing is something like:
@image{
	name: <string>;
	src: <url>;
	opacity: <float>;
	transform/transform-origin/width/height/etc
}

Background-image or border-image could then refer to the name specified in the @image instead of to a URL.
I'm unsure where to the line is drawn between the properties of @image and the ones of background/border-image.

> If the image definition says 'use the image at this URL, make it 100x100 then rotate 15 degrees' and the image bounding box adjusts for the transform, wouldn't it tile as you expect ?
I agree that it doesn't make much sense to do square tiling of the rotated image.

There was a question about the use-case for having a transform on a background image.
Adobe Illustrator and Flash Pro both support transforms on tiling images. Flash can only do it on images and gradients while PDF and postscript can do it on any type of content.
In the PDF world this feature is used very often but I'm unsure how common it is in Flash. This feature is also present in XAML and the UI frameworks of Windows and Mac.

Having the option of tiling and transforming a SVG or image file would give HTML the same feature and I'm sure designers will use it.

Rik
-----Original Message-----
From: Sylvain Galineau [mailto:sylvaing@microsoft.com] 
Sent: Tuesday, December 07, 2010 6:51 PM
To: Simon Fraser
Cc: Tab Atkins Jr.; Rik Cabanier; Leif Arne Storset; Brad Kemper; www-style list
Subject: RE: background-transform (Was: Re: [css3-images] Repeating oblique gradients)

> From: Simon Fraser [mailto:smfr@me.com]
> Sent: Tuesday, December 07, 2010 6:41 PM
> To: Sylvain Galineau


> If we decide that background-transform is valuable, then I think we'll 
> have a very hard time rejecting background-opacity (which I believe 
> we've done in the past).

Exactly. At which point I wonder what else authors might want; then what we do when authors really start playing with border-image and want the same capabilities in that context. 

> 
> However, it's still not obvious to me that transforms to image applied 
> via either @image rules or a functional syntax would affect the 
> orientation of the background tiling grid, whereas 
> background-transform would, I think.

Maybe I'm not following this one. What do you mean by orientation here ?

If the image definition says 'use the image at this URL, make it 100x100 then rotate 15 degrees' and the image bounding box adjusts for the transform, wouldn't it tile as you expect ? Or are you referring to something else entirely ?
Received on Wednesday, 8 December 2010 05:09:20 GMT

This archive was generated by hypermail 2.3.1 : Tuesday, 26 March 2013 17:20:35 GMT