<html>
  <head>
    <meta content="text/html; charset=ISO-8859-1"
      http-equiv="Content-Type">
  </head>
  <body text="#000000" bgcolor="#FFFFFF">
    <br>
    <div class="moz-cite-prefix">On 25.04.2014 12:26, Gerd Petermann
      wrote:<br>
    </div>
    <blockquote cite="mid:DUB112-W128AB73EDB353D97940F9BA9E5A0@phx.gbl"
      type="cite">
      <style><!--
.hmmessage P
{
margin:0px;
padding:0px
}
body.hmmessage
{
font-size: 12pt;
font-family:Calibri
}
--></style>
      <div dir="ltr">Hi Felix,<br>
        <br>
        okay, I think that this explains your problems. The current code
        in mkgmap simply doesn't <br>
        expect lines based on the same OSM way to go in different
        directions.<br>
        I am pretty sure that you get unpredictable results when doing
        that.<br>
        <br>
        In most places I don't see a problem to support that, but I fear
        that the <br>
        code for restriction relations will be a problem.<br>
        Thinking about it...<br>
      </div>
    </blockquote>
    Well I'm still using --no-turn-restrictions anyhow....<br>
    <br>
    However I would see a second possibility. Create another file in the
    style-file called direction-dependent-lines<br>
    where one lists all types <br>
    (0x?????<br>
    0x??<br>
    0x????<br>
    ....<br>
    ) <br>
    that mkmgap may not reverse because the outcome is dependent on the
    direction of the way due to the layout chosen in the typfile. <br>
    <br>
    Or add all tags - but that would be less reliable.<br>
    <br>
    <br>
    Anyhow - the turn-restrictions are still a mystery to me - both in
    mkgmap as well as in OSM. I'm pretty sure 50% of the turn
    restrictions in OSM will not be followed by cyclists (either they
    actually don't have to, or there are quicker possibilities (e.g.
    using the footway crossing) or simply knowing that noone cares what
    cyclists do in that case...<br>
    Especially I would say no cyclists respects right turn restriction
    in drive-right countries (and some vice-versa for drive-left). <br>
    Therefore for now I still prefer to completly disrespect them in the
    map.<br>
    <br>
    <br>
    <blockquote cite="mid:DUB112-W128AB73EDB353D97940F9BA9E5A0@phx.gbl"
      type="cite">
      <div dir="ltr"><br>
        Gerd<br>
        <br>
        <div>
          <hr id="stopSpelling">Date: Fri, 25 Apr 2014 09:30:54 +0800<br>
          From: <a class="moz-txt-link-abbreviated" href="mailto:extremecarver@gmail.com">extremecarver@gmail.com</a><br>
          To: <a class="moz-txt-link-abbreviated" href="mailto:mkgmap-dev@lists.mkgmap.org.uk">mkgmap-dev@lists.mkgmap.org.uk</a><br>
          Subject: Re: [mkgmap-dev] oneway reverse patch<br>
          <br>
          <p dir="ltr">Oh forgot to mention. I also sometimes set oneway
            yes. Then continue and oneway -1 then contune. All non
            routable. I don't use continue with actions so I depend on
            strict work down of lines file as routable non oneway road
            copies follow</p>
          <div class="ecxgmail_quote">On Apr 25, 2014 9:25 AM, "Felix
            Hartmann" &lt;<a moz-do-not-send="true"
              href="mailto:extremecarver@gmail.com">extremecarver@gmail.com</a>&gt;
            wrote:<br>
            <blockquote class="ecxgmail_quote" style="border-left:1px
              #ccc solid;padding-left:1ex;">
              <p dir="ltr">Oh and I of course also add multiple routable
                lines sometimes. Up to 6 per way in osm as 7 produces
                crashes on garmins side. Often one routable line has
                direction but no oneway while the copy is higher
                importance for routing and oneway (and might be
                reversed). I need to reverse sometimes to save on the
                number of lines in type file too - other times its about
                routing!</p>
              <div class="ecxgmail_quote">On Apr 25, 2014 9:19 AM,
                "Felix Hartmann" &lt;<a moz-do-not-send="true"
                  href="mailto:extremecarver@gmail.com" target="_blank">extremecarver@gmail.com</a>&gt;
                wrote:<br>
                <blockquote class="ecxgmail_quote"
                  style="border-left:1px #ccc solid;padding-left:1ex;">
                  <p dir="ltr">Yes I both add opposite oneway on the
                    same road as well as only one oneway or no oneway.
                    It is highly important that the order is followed
                    strictly and continue vs continue with actions is
                    strictly enacted in order. All reversing should
                    happen at the time its placed in the style. !!</p>
                  <p dir="ltr">Sometimes I reverse a way 3 times during
                    processing. All I can say right now its a bit of a
                    mess</p>
                  <div class="ecxgmail_quote">On Apr 25, 2014 2:24 AM,
                    "Gerd Petermann" &lt;<a moz-do-not-send="true"
                      href="mailto:gpetermann_muenchen@hotmail.com"
                      target="_blank">gpetermann_muenchen@hotmail.com</a>&gt;
                    wrote:<br>
                    <blockquote class="ecxgmail_quote"
                      style="border-left:1px #ccc
                      solid;padding-left:1ex;">
                      <div>
                        <div dir="ltr">Hi Felix,<br>
                          <br>
                          if your style interprets a tag like a
                          oneway=yes you should add<br>
                          that tag. This might prevent problems caused
                          by the RoadMerger,<br>
                          which might reverse lines which are not
                          oneways.<br>
                          If you find or set oneway=-1, the current
                          implementation<br>
                          will reverse the way. <br>
                          If you add multiple routable lines for one OSM<br>
                          way, one with, one without the oneway tag,<br>
                          you will see unpredictable directions if such
                          a way <br>
                          is modified by the WrongAngleFixer and the
                          type<br>
                          is direction dependent.<br>
                          The WrongAngleFixer assumes that the points in
                          <br>
                          all ways with the same OSM id are equal, so<br>
                          it optimizes one way and copies the points to
                          the others.<br>
                          This will fail if they have different oneway
                          attributes.<br>
                          If you think this could be the cause of the
                          problem,<br>
                          I should be able to provide a fix.<br>
                          <br>
                          Gerd<br>
                          <div><br>
                            <hr>Date: Fri, 25 Apr 2014 01:00:18 +0800<br>
                            From: <a moz-do-not-send="true"
                              href="mailto:extremecarver@gmail.com"
                              target="_blank">extremecarver@gmail.com</a><br>
                            To: <a moz-do-not-send="true"
                              href="mailto:mkgmap-dev@lists.mkgmap.org.uk"
                              target="_blank">mkgmap-dev@lists.mkgmap.org.uk</a><br>
                            Subject: Re: [mkgmap-dev] oneway reverse
                            patch<br>
                            <br>
                            <div dir="ltr">
                              <div>Ups - but forgot to say. I think in
                                99% of all cases - cycleway:left and
                                cycleway:right are used on streets which
                                feature oneway=yes... less often
                                oneway=-1 and even less often no oneway
                                at all.<br>
                              </div>
                              Streets with oneway=yes are fine. I'm
                              talking about no oneway tag from OSM data
                              at all, or oneway=-1 set in style (but no
                              oneway from OSM data) or no oneway at all.
                              Only on those there are problems - so
                              you're not likely to notice them I
                              think....<br>
                            </div>
                            <div><br>
                              <br>
                              <div>On 25 April 2014 00:39, Minko <span
                                  dir="ltr">&lt;<a
                                    moz-do-not-send="true"
                                    href="mailto:ligfietser@online.nl"
                                    target="_blank">ligfietser@online.nl</a>&gt;</span>
                                wrote:<br>
                                <blockquote style="border-left:1px #ccc
                                  solid;padding-left:1ex;">
                                  Yes I render cycleway:left and
                                  cycleway:right too.<br>
                                  And as you say, they are always on the
                                  wrong side on the GPS.<br>
                                  Interesting to know that Garmin uses
                                  asymmetric lines independent of the
                                  road direction. If we only could find
                                  out how they do this...<br>
                                  <br>
                                  Felix wrote<br>
                                  <div>&gt; I don't think that Minkos
                                    style shows cycleways on the
                                    left/right side<br>
                                    &gt; of a road - am I wrong? -
                                    Anyhow they would usually have a
                                    oneway tag<br>
                                    &gt; already in OSM data.<br>
                                    &gt; Also definitely no
                                    uphill/downhill arrows - which
                                    nearly never<br>
                                    &gt; actually have a oneway tag (and
                                    only sometimes I add one during<br>
                                    &gt; processing).<br>
                                  </div>
                                  <div>
                                    <div>_______________________________________________<br>
                                      mkgmap-dev mailing list<br>
                                      <a moz-do-not-send="true"
                                        href="mailto:mkgmap-dev@lists.mkgmap.org.uk"
                                        target="_blank">mkgmap-dev@lists.mkgmap.org.uk</a><br>
                                      <a moz-do-not-send="true"
                                        href="http://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev"
                                        target="_blank">http://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev</a><br>
                                    </div>
                                  </div>
                                </blockquote>
                              </div>
                              <br>
                            </div>
                            <br>
                            _______________________________________________
mkgmap-dev
                            mailing list
                            <a moz-do-not-send="true"
                              href="mailto:mkgmap-dev@lists.mkgmap.org.uk"
                              target="_blank">mkgmap-dev@lists.mkgmap.org.uk</a>
                            <a moz-do-not-send="true"
                              href="http://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev"
                              target="_blank">http://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev</a></div>
                        </div>
                      </div>
                      <br>
                      _______________________________________________<br>
                      mkgmap-dev mailing list<br>
                      <a moz-do-not-send="true"
                        href="mailto:mkgmap-dev@lists.mkgmap.org.uk"
                        target="_blank">mkgmap-dev@lists.mkgmap.org.uk</a><br>
                      <a moz-do-not-send="true"
                        href="http://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev"
                        target="_blank">http://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev</a><br>
                    </blockquote>
                  </div>
                </blockquote>
              </div>
            </blockquote>
          </div>
          <br>
          _______________________________________________
          mkgmap-dev mailing list
          <a class="moz-txt-link-abbreviated" href="mailto:mkgmap-dev@lists.mkgmap.org.uk">mkgmap-dev@lists.mkgmap.org.uk</a>
          <a class="moz-txt-link-freetext" href="http://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev">http://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev</a></div>
      </div>
      <br>
      <fieldset class="mimeAttachmentHeader"></fieldset>
      <br>
      <pre wrap="">_______________________________________________
mkgmap-dev mailing list
<a class="moz-txt-link-abbreviated" href="mailto:mkgmap-dev@lists.mkgmap.org.uk">mkgmap-dev@lists.mkgmap.org.uk</a>
<a class="moz-txt-link-freetext" href="http://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev">http://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev</a></pre>
    </blockquote>
    <br>
  </body>
</html>