Tagged: 

Viewing 14 posts - 31 through 44 (of 44 total)
  • Author
    Posts
  • #25486
    ChristianK
    ChristianK
    Keymaster

    OK, next try – preserving oneway=yes setting cycleway=lane to cycleway=dummy
    adding cycleway_lane=cw_lane_oneway

    [/crayon]

    This should be OK?

    Here is the complete tagtransform_file:
    https://www.openandromaps.org/wp-content/snippets/makes/tt_andromaps.xml

    Here ist the map of tilburg:
    https://drive.google.com/open?id=1QwCP4JdOV_AozsmntQhJP1wK3D5qqpmU

    Here is the preprocessed OSM file as it is used for rendering the map-file:
    https://drive.google.com/open?id=1acqJEpsoBHfuO3P8Pk1R7aegpK7Asy8p

    LG, Christian

    2 users thanked author for this post.
    #25490
    Tobias
    Tobias
    Moderator

    Works for me :-)

    #25492
    Tobias
    Tobias
    Moderator

    @John posts:
    I think that the tag transforms are correct, what is missing are some addtional tags and transforms:

    cycleway_lane=opposite_lane (tag transfrom from: cycleway=opposite_lane)
    cycleway_lane=opposite_lane_left (tag transfrom from: cycleway:left=opposite_lane)
    cycleway_lane=opposite_lane_right (tag transfrom from: cycleway:right=opposite_lane)
    cycleway_track=opposite_track (tag transfrom from: cycleway=opposite_track)
    cycleway_track=opposite_track_left (tag transfrom from: cycleway:left=opposite_track)
    cycleway_track=opposite_track_right (tag transfrom from: cycleway:right=opposite_track)

    I wouldn’t transform them to regular cycleway_lanes/tracks, because of right- or left-handed traffic and one might to mark them differently or not render them at all.

    Also:
    cycleway:right:oneway=no
    cycleway:left:oneway=no
    cycleway:oneway=no
    tagtransform to to:
    cycleway:oneway=cw_ow_no

    However since I wrote this in the post referred to above, the Wiki recommendation has changed to (eg) cycleway:right=lane + cycleway:right:oneway=no which is a further and unwelcome complication.

    I think these tags make much more sense than oneway:bicycle=no. See in the WIki: https://wiki.openstreetmap.org/wiki/Key:oneway:bicycle
    “Defines the oneway rules for cyclists where they are different from oneway=*.”
    (Edit: when oneway:bicycle=no would be used, the two way road the cycleway is part of would be falsely rendered as oneway.)

    • This reply was modified 1 month, 2 weeks ago by Tobias Tobias.
    • This reply was modified 1 month ago by Tobias Tobias.
    1 user thanked author for this post.
    #25517
    R5O.nl
    R5O.nl
    Participant

    I now see that the Sept maps are releases (tnx for that!) –> what should I expect in this map + Elevate_4.2.5, regarding OneWay displaying?
    Should this now be on the same level as I saw in the ‘Tilburg’ map that was shared in this discussion?
    And thanks to all the guys participating here: I did not expect to spawn this large (and good!) discussion ;)

    • This reply was modified 1 month, 2 weeks ago by R5O.nl R5O.nl.
    • This reply was modified 1 month, 2 weeks ago by R5O.nl R5O.nl.
    #25524
    ChristianK
    ChristianK
    Keymaster

    No its not the same level, the regular map update has the copy_unmatched bug.
    Sorry for this – a regular update for all maps need 3 weeks, I can’t stop this.
    .. and Netherlands was one of the maps redered prior of resolving this bug.

    Actual asia and russia is prozessed on the rendermachine.
    Once the september update is completed I will work on the changes @tobias proposed.

    However, the actual update includes changes in coverage of lots of maps and the inclusion of relations at mapboarders. This means that Steffen have made a special Version of OSMCONVERT64 for OAM that makes the –complete-relations really working.
    This is a major upgrade for the OAM-Maps and keeps me bussy cause there are a lot of side-effects that have to be resolved with this case.

    2 users thanked author for this post.
    #25571
    JohnPercy
    JohnPercy
    Participant

    when oneway:bicycle=no would be used, the two way road the cycleway is part of would be falsely rendered as oneway.

    I’m not sure of the reasoning behind this statement, Won’t the rendering depend on the theme?
    I agree however that specifying that oneway tag applies only to the bike lane by using (eg) cycleway:right:oneway=no avoids this potential ambiguity.

    #25619
    Tobias
    Tobias
    Moderator

    when oneway:bicycle=no would be used, the two way road the cycleway is part of would be falsely rendered as oneway.

    I’m not sure of the reasoning behind this statement, Won’t the rendering depend on the theme?

    I agree however that specifying that oneway tag applies only to the bike lane by using (eg) cycleway:right:oneway=no avoids this potential ambiguity.

    Of course it depends on the theme, but I thought that if one would only have a rule with oneway:bicycle=no and use a symbol as the one in your post above (the two colored two headed arrow) for oneway:bicycle=no, the grey part of the arrow would stand for the main road (this would be totally valid according to the wiki). But of course it’s more secure to use nested rules as in Elevate which also test for oneway=yes.

    #25623
    JohnPercy
    JohnPercy
    Participant

    My scheme is designed in effect to use nested rules and test for oneway=yes.
    IF oneway:bicycle=no AND oneway=yes THEN two-colour blue/grey doubleheaded arrow
    IF cyclelane is on one side only THEN IF oneway:bicycle=no AND oneway=no|~ THEN blue doubleheaded arrow

    • This reply was modified 1 month ago by JohnPercy JohnPercy.
    #25650
    Tobias
    Tobias
    Moderator

    As I said nested rules are more secure. But it would also theoretically be totally okay to use the two headed arrow without testing for oneway=yes/no. oneway:bicycle=no added to a highway marks an exception: “in opposition to other vehicles this is no oneway” (if you take the tag literally), If applied to a road with two-sided traffic, this wouldn’t be an exception. Also this tag applies to the main tag, e.g. highway=primary, and not to cycleway=lane.

    #25786
    Tobias
    Tobias
    Moderator

    When adding this to Elevate, I stumbled upon this

    I think we need the same rules for oneway=-1, resulting in:

    k=”cycleway_lane” v=”cw_lane_oneway_rev”

    k=”cycleway_track” v=”cw_track_oneway_rev”

    …and those will be rendered on the other side.

    Those tags are not contained in tag-mapping (but in tag-transform), so it’s not working (yet). Can you have a look, Christian?

    #25811
    ChristianK
    ChristianK
    Keymaster

    Can you have a look, Christian?

    Done

    Sorry @tobias, I will re-render Bayern.

    #25813
    Tobias
    Tobias
    Moderator

    Nordrhein Westfalen would be better for this actually, thanks!

    1 user thanked author for this post.
    #25841
    ChristianK
    ChristianK
    Keymaster

    Nordrhein Westfalen

    .. ready for download plus Netherlands and Bayern too

    2 users thanked author for this post.
    #25843
    Tobias
    Tobias
    Moderator

    Nordrhein Westfalen

    .. ready for download plus Netherlands and Bayern too

    Thanks, everything works fine now!

    1 user thanked author for this post.
Viewing 14 posts - 31 through 44 (of 44 total)

You must be logged in to reply to this topic.