How should degenerate closed paths render in SVG ? about svgwg

[nikosandronikos]

Does SVG currently define how degenerate open paths render?

PDF recently clarified that the rendering depends on cap.
Rounds caps render a circle, anything else doesn't render.

[nikosandronikos]

Test: https://jsfiddle.net/dodgeyhack/ze5v5ep0/

[heycam]

In Error handling in path data it says:

As mentioned in Stroke Properties, linecaps must be painted for zero length subpaths when ‘stroke-linecap’ has a value of round or square.

And the recently rewritten Path directionality section defines the orientation of the linecaps (and markers).

[nikosandronikos]

Reading the Error handling in path data join-linecap link above, I would say that the currently described result is that even for a closed sub-path, the linecap type defines the result.

The suggestion that the rendered result should be dependent on the linejoin style is nice for the example given. But I'm not sure it's good for all cases, i.e. choosing that option would preclude closed paths with round linejoins from ever disappearing - plus you would have to define additional results for each linejoin type.
I prefer keeping the behaviour of using the linecap style so that open and closed zero length subpaths are equivalent.
It also aligns with PDF for round and butt caps - I think we have different behaviour for square caps.
And I think there's some browser bugs regarding orientation of the caps.

[karip]

I’m ok to forget my suggestion to use line joins, consistency with pdf sounds good.

Perhaps you can make the spec clearer? See comment 3 for inconsistencies: https://bugs.chromium.org/p/skia/issues/detail?id=4784#c3

[nikosandronikos]

Some test results:

	Butt	Square	Round
Expected	nothing	square	round
FireFox (OS X)	nothing	star	round
Chrome (OS X)	nothing	nothing	nothing
Safari	nothing	square	round
Edge	nothing	square	round

Looks like we'll need to talk to the Chrome guys.

[nikosandronikos]

RESOLUTION: zero length closed subpaths should render the same as zero length open subpath using the stroke-linecap to determine the result
https://www.w3.org/2016/02/18-svg-minutes.html#item01

I'll file a bug on Chrome

[nikosandronikos]

https://bugs.chromium.org/p/chromium/issues/detail?id=588856