Re: marker improvements

On 11-06-12 22:02, Stephen Chenney wrote:
> Regarding the zero-length line and markers issue. I asked about this
> some time back, and understood the target behavior to be as follows ...
> 
> 1. Isolated moveTo segments generate both a marker-begin and marker-end
> at each move location. e.g. M 10 10 m 10 10 m 10 10 would produce both
> marker-begin and marker-end at (10,10), (20,20) and (30,30). The use
> case for this is scatter plot drawing, but it seems the positioned
> marker spec handles that case now, so I suggest we draw no markers for
> isolated moves.

As long as it is valid to create isolated moveTo's, then please don't
special-case them more than necessary. The current spec has a pretty
decent description of the rules for determining the orientation of
nodes. I imagine this will carry over to any enhancements, and will take
care of what to do with zero-length subpaths (including, but not limited
to, isolated moveTo's).

Received on Tuesday, 12 June 2012 07:30:37 UTC