[svgwg] Issue: Marker auto-start-reverse and closed sub-paths marked as SVG Markers

nikosandronikos has just labeled an issue for 
https://github.com/w3c/svgwg as "SVG Markers":

== Marker auto-start-reverse and closed sub-paths ==
I'm putting together some more marker tests for WPT and I came across 
this...

For a marker with 
**[orient](https://svgwg.org/svg2-draft/painting.html#OrientAttribute)**="auto",
 on a **[closed 
sub-path](https://svgwg.org/svg2-draft/paths.html#TermClosedSubpath)**,
 the marker start must be the **[direction of the path at that 
point](https://svgwg.org/svg2-draft/paths.html#PathDirectionality)**.
So for the following rect:
```
M 50,50 h50 v50 h-50 z
```
If marker-start, marker-mid, and marker-end were specified, four 
markers would be visible (though more would be drawn, they would just 
overlap) and each would be pointing at the mid point of the direction 
of the incoming segment and outgoing segment.
So at the top right of the rect, where the incoming segment travels 
east and the outgoing segment travels south, the marker would point 
south east).

For a marker with orient="auto-start-reverse", the orient for 
marker-start is 180 different than the orientation would be if "auto" 
were specified. This is because we want the marker at each end of an 
open path to point in opposite directions. We don't give any specific 
behaviour for closed paths, so currently this means the marker-start 
on a closed path would be oriented in the opposite direction as well.

Do we want to restrict auto-start-reverse behaviour to open sub-paths 
only? 

Here's a test to see what's going on:
https://jsfiddle.net/dodgeyhack/7h0csof5/

See https://github.com/w3c/svgwg/issues/300

Received on Tuesday, 13 December 2016 04:31:51 UTC