I set up the restrictions so that who comes from the side lane encounters an *only_straight_on *relation which terminates on the side lane you circled in red. The relation should force them to continue "straight on" to the side lane. The same goes for who comes from the main lane, which is forced (by another relation) to continue *only straight on *the main lane. In your picture, you should see (if I'm not mistaken) that the *from* member is in fact the side lane on the bottom part. On Monday, 2 November 2020 at 12:16:24 UTC+1 CP wrote:
> I think I now understand what your issue is. Are you sure it's not a > modelling issue? > > > > > To me it looks like you've explicitly *allowed* the right-hand turn. (but > maybe I'm not understanding these restrictions correctly.) > If I were to model it, I'd pick the main lane, and put a "no_right_turn" > to the side lane. > My brain is starting to short-circuit now ;-) > > > Op 02-11-2020 om 10:41 schreef Davide F: > > > This is from brouter.de, but the issue is the same with OsmAnd. > Coming from the green bubble on the lower side and driving northwards, the > street splits in two, the orange "main lane" on the left and the white > "side lane" on the right. Between the rail bridges, the two lanes rejoin > with one another, but the signage on the road forbids whoever is in the > side lane to cross over to the main lane, and vice versa, so it's as if the > two lanes kept on being two separate ways, as far as normal routing should > be concerned. > This is important in the case you want, for example, to drive to the red > bubble at the top, since you must then take the side lane *before the > first bridge*, and you are forbidden to do so between the bridges as the > router tells you. > I tried to model this restriction as I wrote in the original post, but > OsmAnd seems to ignore that, and suggests the route in the picture, which > you can't follow in real life. > > On Friday, 30 October 2020 at 19:10:06 UTC+1 CP wrote: > >> Do you have a screenshot of the faulty routing? I don't think I >> understand what exactly the issue is. A screenshot would help. >> >> >> >> Op 30-10-2020 om 17:48 schreef Davide F: >> >> I'm sure the map is up to date. >> >> On Thursday, 29 October 2020 at 13:56:13 UTC+1 tomcroc...@gmail.com >> wrote: >> >>> Do you have osmand live and is your downloaded map up to date? If not, >>> the change won't be downloadable for a week or two. >>> >>> On Thu, 29 Oct 2020 at 10:16, Davide F <dv.fe...@gmail.com> wrote: >>> >>>> I don't think so (but I'm not sure). That way you highlighted is a >>>> 'via' member, and it is indeed the same in both relations, but the 'from' >>>> members are different, so there should be no ambiguity. There would be no >>>> problem if the 'via' segment were collapsed to a single point. >>>> >>>> On Tuesday, 27 October 2020 at 17:26:54 UTC+1 CP wrote: >>>> >>>>> >>>>> >>>>> Op 27-10-2020 om 15:35 schreef Davide F: >>>>> >>>>> I recently added to OSM a turn restriction relation ( >>>>> https://www.openstreetmap.org/relation/11790375) which has two ways, >>>>> instead of a single point, as 'via' members. This usage is documented on >>>>> OSMWiki, so I didn't make anything up, however OsmAnd seems to ignore >>>>> this >>>>> relation while calculating the route. Do you know if this is supported by >>>>> the app? >>>>> >>>>> >>>>> Caveat emptor: I'm no hero with turn restrictions. >>>>> >>>>> If I look at that little piece of main road... >>>>> >>>>> >>>>> >>>>> ... and then in the OSM edit page at the bottom left, you will see >>>>> that there's both a restriction "Only straight on" straight onward >>>>> *and* a restriction "Only straight on" to the right. So I think >>>>> they're cancelling each other out. Could that be the problem? >>>>> >>>>> CP >>>>> >>>>> -- >>>> You received this message because you are subscribed to the Google >>>> Groups "OsmAnd" group. >>>> To unsubscribe from this group and stop receiving emails from it, send >>>> an email to osmand+un...@googlegroups.com. >>>> To view this discussion on the web visit >>>> https://groups.google.com/d/msgid/osmand/a320040a-2cbb-4f34-af2c-05c3d0f905d7n%40googlegroups.com >>>> >>>> <https://groups.google.com/d/msgid/osmand/a320040a-2cbb-4f34-af2c-05c3d0f905d7n%40googlegroups.com?utm_medium=email&utm_source=footer> >>>> . >>>> >>> -- >> You received this message because you are subscribed to the Google Groups >> "OsmAnd" group. >> To unsubscribe from this group and stop receiving emails from it, send an >> email to osmand+un...@googlegroups.com. >> >> To view this discussion on the web visit >> https://groups.google.com/d/msgid/osmand/ad89e474-b2a7-4d09-8abb-8970073a0db1n%40googlegroups.com >> >> <https://groups.google.com/d/msgid/osmand/ad89e474-b2a7-4d09-8abb-8970073a0db1n%40googlegroups.com?utm_medium=email&utm_source=footer> >> . >> >> >> -- > You received this message because you are subscribed to the Google Groups > "OsmAnd" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to osmand+un...@googlegroups.com. > > To view this discussion on the web visit > https://groups.google.com/d/msgid/osmand/0dde5d98-ca39-42e0-8299-09844283c7c7n%40googlegroups.com > > <https://groups.google.com/d/msgid/osmand/0dde5d98-ca39-42e0-8299-09844283c7c7n%40googlegroups.com?utm_medium=email&utm_source=footer> > . > > > -- You received this message because you are subscribed to the Google Groups "OsmAnd" group. To unsubscribe from this group and stop receiving emails from it, send an email to osmand+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/osmand/b9577ef5-c447-47e1-8021-f63a7a367007n%40googlegroups.com.
