Re: Rendering of marker-start at the beginning of a subpath

Then the paragraph says "One way to prevent this is to set ‘marker-end’ to
none". Which a reader could easily interpret to mean it is talking about
start and end.

When you are totally familiar with what specification intends to say, the
meaning of a sentence can be clear.  On the other hand, when you are
reading it for the first time, and trying to parse intention from just a
few sentences, it can be easy to misinterpret the correct behaviour.  Hence
the different implementations that the OP pointed out.

I should not have replied to the email when I was tired because I ended up
making that mistake myself.  My apologies to you, Regina.

Paul



On 27 September 2013 03:13, Dr. Olaf Hoffmann <Dr.O.Hoffmann@gmx.de> wrote:

> Paul LeBeau:
> > That's the sentence I was referring to in my post.  That is later
> seemingly
> > contradicted (or clarified?) by the paragraph I posted.  It tends to
> > suggest that it may have been intended that start and end markers applied
> > to all subpaths.
>
> Do you mean sentences in the same paragraph:
> "Note that for a 'path' element which ends with a closed sub-path, the last
> vertex is the same as the initial vertex on the given sub-path. In this
> case,
> if 'marker-end' does not equal none, then it is possible that two markers
> will be rendered on the given vertex."
>
> I cannot see a contradiction (nevertheless, the usage of 'sub-path' can
> be a little bit confusing, but logically it does not cause a problem
> together
> with the initial definition), it only notes, that it is possible, that this
> happens.  This does not implicate a specific behaviour for sub-paths.
> I clarifies basically, that closed paths have an initial and final point as
> other paths and not just one for both.
> The Z-command implicates, that the initial point of the sub-path
> is the same as the final (and implicates something for some
> stroke-properties as well, but obviously not for markers - but
> currently one cannot have different linecaps at the begin or end of
> the path anyway and if closed the linejoin applies instead).
> Now there can be four different cases (assuming different markers
> for begin, end and mid):
> a) The initial point of the sub-path is the initial point of
> the complete path and the final point of the sub-path is
> the final point of the complete path - in this case, the markers for begin
> and end are drawn at the same vertex.
> b) The initial point of the sub-path is not the initial point of
> the complete path and the final point of the sub-path is
> the final point of the complete path - in this cae, a marker for mid and
> end are drawn at the same vertex.
> c) The initial point of the sub-path is the initial point of
> the complete path and the final point of the sub-path is
> not the final point of the complete path - in this case, a marker for
> begin and mid are drawn at the same vertex.
> d) The initial point of the sub-path is not the initial point of
> the complete path and the final point of the sub-path is
> not the final point of the complete path - in this case,
> two mid markers have a coincidence - because they are
> of the same type, nothing interesting will result from
> this.
>
> Olaf
>