- From: Amelia Bellamy-Royds <amelia.bellamy.royds@gmail.com>
- Date: Wed, 17 Feb 2016 12:57:06 -0700
- To: Sebastian Zartner <sebastianzartner@gmail.com>
- Cc: www-svg <www-svg@w3.org>
- Message-ID: <CAFDDJ7wUH5-OJ7VXO7i_68juZK8V4Pt7GxPLBXLSvEPw+hivZg@mail.gmail.com>
[Replies to Sebastian Zartner's comments] I like the idea of the 'geometry' property, though I dislike the > overlapping of function and keyword names for circle, ellipse and rect. I > believe it would be better to reuse the functions defined in the CSS Shapes > module[1]. > > E.g. the UA stylesheet rule for <circle> elements may look like this then: > > circle { > geometry: circle(attr(r) at attr(cx) attr(cy)); > } > The `cx`, `cy`, `r` properties, etc., have already shipped in at least two browsers (WebKit & Chromium). So any new property would need to reference them. Regarding the statement above about CSS being universal, it needs to be > clarified what happens when an author defines this for example: > > textPath { > geometry: circle(10cm at 5cm 5cm); > } > The way I was thinking of it, it would actually turn into a circle. Except without the useful DOM object attached (as Tab mentioned in his comments). It would nonetheless be something to discourage as a poor authoring practice, similar to using nothing but <blockquote> in an HTML document and overriding styles to create the appearance of structure. > We then need one final master property to turn on all these SVG-specific >> rendering commands. Something I've talked with Tab about previously is to >> do this with the `display` property. >> > > I assume this discussion is related to the CSS Display module[2], right? > It sounds like this point is out of scope of this discussion to be talked > about separately. > ... > Can you elaborate a bit more on this? What is the problem of being able to > toggle the display off without changing the box model? > Yes, it is related, and definitely stretches outside of the scope of SVG. A different approach than the `box-suppress` method described in CSS Display, but it would address the same goal. But it would also address the issue that most of the `display` values are completely irrelevant and ignored for SVG, while providing a way for SVG-namespaced elements to "opt-in" to the CSS box layout model as part of an alternative textual display. > > [1] https://drafts.csswg.org/css-shapes/ > [2] https://drafts.csswg.org/css-display/ > >
Received on Wednesday, 17 February 2016 19:57:43 UTC