Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Residential render below trunk link #266

Closed
Rovastar opened this issue Nov 15, 2013 · 16 comments
Closed

Residential render below trunk link #266

Rovastar opened this issue Nov 15, 2013 · 16 comments

Comments

@Rovastar
Copy link
Contributor

Currently residential renders above the trunk link like here.

http://www.openstreetmap.org/#map=19/52.92545/-1.50992

Probably need to look at all the layer ordering for link roads as they I think are pushed to the bottom.

@matthijsmelissen
Copy link
Collaborator

Thanks for reporting that. I will look at it.

@Rovastar
Copy link
Contributor Author

Excellent. As you are looking at roads it makes sense to pick it up. Otherwise I'll get to it at some point.
I'll have a look for any more road anomalies too and post anything I find.

@matkoniecz
Copy link
Contributor

@matthijsmelissen
Copy link
Collaborator

I think that there is unfortunately no clear solution, because in some cases (links ending in a T junction on residential roads or other lower types or roads), we do want the lower type of road to be drawn on top of the link.

@Rovastar
Copy link
Contributor Author

Rovastar commented Feb 9, 2014

Overall I think the situation can be improved with a more simple reordering. All link roads are currently render first and everything else is put on top we just need to move these up a little. I was say, thinking Motorway link above Secondary or maybe Primary roads. Trunk link above tertiary or maybe secondary roads, etc need to think in a bit more detail what the very best ordering is.
Occasionally you want to see other info but the majority of cases should look much better.

@matthijsmelissen
Copy link
Collaborator

That's a heuristic solution that would still break many situations, like here: http://www.openstreetmap.org/#map=17/51.63670/-2.45155

@matthijsmelissen
Copy link
Collaborator

@Rovastar I'm not really sure how to approach this issue. You suggest not worrying about edge cases, but to me, http://www.openstreetmap.org/#map=17/41.05566/14.57152 seems to be an edge case - perhaps even a mistagging, shouldn't the road to the right be primary instead of primary_link? I would be happy with any suggestions / proposals / ideas.

@Rovastar
Copy link
Contributor Author

I suppose I was thinking of cases we can improve first. E.g. Motorway, (trunk, secondary) links are to be rendered above footpath, service roads, living streets, tracks, etc.
let's go all the way up to unclassified as you have seen tertiary cases.
about the edge cases my point is that sometimes things will look different and we should not be worried if overall it is improved.

@matthijsmelissen
Copy link
Collaborator

Yes, that would make sense.

What would be the best way to accomplish this? Creating a separate layer for every road type is not the best solution, I think, so the SQL output should be ordered correctly already. We currently render by z_order. We could add a manual ORDER BY in the SQL, something like

ORDER BY case when highway = 'motorway' then 3 when highway = 'motorway_link' then 2 when highway = 'footway' then 1 [...] end

but that looks ugly, and might come with a huge performance penalty.

I think the best option would be to change osm2pgsql, and populate the z_order field in such a way that footpaths, service roads, living streets, tracks etc. have a lower z_order then link roads.

Could we accomplish this by merely changing the osm2pgsql stylesheet, or would we need to patch osm2pgsql itself?

@matkoniecz
Copy link
Contributor

@matthijsmelissen
Copy link
Collaborator

Resolved by #626.

@matthijsmelissen
Copy link
Collaborator

@mkoniecz's problem is solved but the original one is still open. Reopened for further investigation.

@matthijsmelissen
Copy link
Collaborator

The original issue was not accurately tagged/traced, I improved the tagging. It's not clear if a further fix is necessary or possible. I will therefore close this issue for now. If there are other situations like this, feel free to create a new issue (or reply to this one).

@pnorman
Copy link
Collaborator

pnorman commented Aug 19, 2014

closing as per #266 (comment)

@pnorman pnorman closed this as completed Aug 19, 2014
@Rovastar
Copy link
Contributor Author

I didn't realise that junction had changed. It wasn't tagged incorrectly at the time (it did really cross the two links) but the junction has now it seems changed.

I am not sure this issue is actually closed. Unclassified should go under trunk links. But the whole links thing needs looking at anyway.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
Development

No branches or pull requests

4 participants