Modify

Opened 15 years ago

Closed 13 years ago

#2387 closed defect (fixed)

oneway=no shown with arrow

Reported by: smsm1 Owned by: ulfl
Priority: major Milestone:
Component: Internal mappaint style Version:
Keywords: Cc: unriccio@…, kay_D, nakor.wp@…

Description

It seems that oneway=no is shown as though it is a one way road, which isn't true and can be confusing when editing.

Attachments (2)

oneway.osm (2.0 KB ) - added by anonymous 14 years ago.
used oneway values
oneway.png (4.2 KB ) - added by anonymous 14 years ago.
used oneway values - rendering with josm 2466

Download all attachments as: .zip

Change History (15)

comment:1 by stoecker, 15 years ago

Resolution: wontfix
Status: newclosed

Yes. JOSM only cares for the key itself and not its value for speed reasons. Anyway this is no big issue as oneway=no is default, so it needs not be entered at all.

comment:2 by unriccio@…, 15 years ago

Cc: unriccio@… added
Resolution: wontfix
Status: closedreopened

that's wrong because all "_link" ways imply oneway=yes, and one have to explicitly tag the oneway=no ways:
http://wiki.openstreetmap.org/wiki/Tag:highway%3Dmotorway_link

comment:3 by ramzai, 14 years ago

JOSM only cares for the key itself and not its value for speed reasons.

Also there is oneway=-1, that reverses direction and that makes such oneways to be shown in the wrong direction in JOSM.

comment:4 by jttt, 14 years ago

Performance is no longer an issue because JOSM now keeps hasDirectionKeys as flag that is recalculated only when primitive keys are changed. It should be even possible (fast enough) to use pattern from Search action to decide whether arrow should be shown

by anonymous, 14 years ago

Attachment: oneway.osm added

used oneway values

by anonymous, 14 years ago

Attachment: oneway.png added

used oneway values - rendering with josm 2466

comment:5 by kay_D, 14 years ago

Cc: kay_D added

What happens with the direction of the way, can it still be seen? I use :left and :right a lot and they depend on the direction. If (for oneway=-1) the arrows are reversed visually, I see potential for tagging errors.

comment:6 by jttt, 14 years ago

See also #2809

comment:7 by Nakor, 14 years ago

Cc: nakor.wp@… added

comment:8 by jttt, 14 years ago

Ways with oneway=no no longer have arrows, oneway=-1 is still broken.

What should we do with _link highways? Show arrows unless oneway=no?

Currently the search pattern to identify ways is arrows is as follows: oneway? | incline=* | incline_steep=* | aerialway=* | waterway=stream | waterway=river | waterway=canal | waterway=drain | "piste:type"=downhill | "piste:type"=sled | man_made="piste:halfpipe"

comment:9 by bastiK, 14 years ago

(In [2890]) Reverse Arrows for 'oneway=-1' and similar (see #2387).
More efficient calculation of arrow geometry.
Test file for Arrow direction added.

comment:10 by bastiK, 14 years ago

I like to add incline=-10% (some negative value) to the reversed arrow search pattern. Currently I don't know how. (Maybe regex?)

comment:11 by augustus.kling@…, 14 years ago

I suggest not showing arrows for “_link” highways at the moment because the implicated tags are unclear. Different languages of the OSM wiki are stating contradictory implications.

The implied tags for “highway=motorway_link” from the wiki:
German: oneway=yes
English: oneway=yes, access=no, motorcar=yes, hgv=yes, surface=paved
Italian: oneway=yes, access=no, motorcar=yes, hgv=yes, surface=paved
Russian: oneway=yes, access=no, motorcar=yes, hgv=yes, surface=paved
Ukrainian: oneway=yes

I am aware of the fact that oneway=yes is given in all cases for highway=motorway_link. However for highway=trunk_link, highway=primary_link or highway=secondary_link this is not implicated in all languages. Instead oneway=* is mentioned.

I suggest, cleaning up the wiki first so that it only implies oneway=yes for all “_link” highways. After that was done, arrows should be shown for “_link” highways unless oneway=no is given.

In my opinion guessing vehicle type restrictions or surfaces should be left to the routers/renders and not be implied globally as implications may differ across regions. Clearly, implying access=no makes no sense but JOSM's bug tracker is not the right place for tagging scheme discussions.

in reply to:  10 comment:12 by jttt, 14 years ago

Replying to bastiK:

I like to add incline=-10% (some negative value) to the reversed arrow search pattern. Currently I don't know how. (Maybe regex?)

I think you can't currently do that. Search expression should get some way to mark that pattern is regular expression or case insensitive, without using checkboxs..

comment:13 by skyper, 13 years ago

Resolution: fixed
Status: reopenedclosed

The main issue of the ticket is fixed (oneway <-> arrows).

The incline problem should go into another ticket.

Note: I think it is dangerous to render oneway and incline with on function. What do you do with a oneway road leading downhill ?

r4157

Modify Ticket

Change Properties
Set your email in Preferences
Action
as closed The owner will remain ulfl.
as The resolution will be set.
The resolution will be deleted. Next status will be 'reopened'.

Add Comment


E-mail address and name can be saved in the Preferences .
 
Note: See TracTickets for help on using tickets.