use land-color as background for waterway tunnels #2594
Conversation
|
I think for intermittent waterways this goes into the wrong direction. Both tunnels and intermittent waterways are something less than a normal waterway in terms of visibility and importance for the typical map user. Therefore they should IMO be rendered less prominent than normal waterways. Using land color between the dashes will make them more prominent on many backgrounds. The difference is likely fairly small for tunnels so i consider the change acceptable there although both the current and the new proposed styling are not ideal in terms of relative prominence at the moment. I am generally not really satisfied with the dashed rendering of intermittent waterways we have at the moment, especially at the low zoom levels. And at the high zooms we should probably revisit that when we implement intermittent water areas - see #996. Note if we decide to change the water color in a way similar to what i proposed in #1781 (comment) the relative weight of surface waterways vs. tunnels would change (to the better). The side effect you showed would be solved by the layer reordering i have envisaged for the move to water polygons and differentiated color water rendering (i.e. rendering waterways before water areas). |
water.mss
Outdated
| @@ -103,6 +103,20 @@ | |||
| [waterway = 'canal'][zoom >= 12], | |||
| [waterway = 'river'][zoom >= 12], | |||
| [waterway = 'wadi'][zoom >= 13] { | |||
| // the additional line of land color is used to provide a background for dashed appearances | |||
There was a problem hiding this comment.
tab in the middle of the line?
water.mss
Outdated
| [intermittent = 'yes'], | ||
| [waterway = 'wadi'], | ||
| [int_tunnel = 'yes'] { | ||
| line-color: @land-color; |
There was a problem hiding this comment.
Maybe for clarity good to prefix these with land/ or background/ or so?
nebulon42
force-pushed
the
water-tunnel-casing
branch
2 times, most recently
from
bf896a9 to
39d45c7
Compare
|
I have adressed the points raised. For wadis the same as for intermittent waterways applies, therefore I have both removed from this change. Wadis would have looked like this (likely not a wadi but tagged as one): |
nebulon42
changed the title
water.mss
Outdated
| [zoom >= 18] { background/line-width: 12; } | ||
| } | ||
|
|
||
| [int_tunnel = 'yes'] { |
There was a problem hiding this comment.
This selector is the same as the selector above
| @@ -113,6 +125,7 @@ | |||
| [zoom >= 18] { bridgecasing/line-width: 13; } | |||
| } | |||
| } | |||
|
|
|||
There was a problem hiding this comment.
Spurious whitespace
| @@ -166,24 +195,35 @@ | |||
| [waterway = 'stream'][zoom >= 15] { bridgeglow/line-width: 3; } | |||
| } | |||
| } | |||
|
|
|||
There was a problem hiding this comment.
Spurious whitespace in diff
|
Ping @nebulon42 about the technical comments above. |
nebulon42
force-pushed
the
water-tunnel-casing
branch
from
39d45c7 to
970df74
Compare
|
@math1985 I have merged the redundant filters, thanks for noticing. The additional newlines were intentional. I think they improve readability. |
|
Does this solve / change #405 too? |
|
No, because the river line is still rendered above the tunnel. |
Ref #488.
Waterway tunnel dashes might disappear depending on background. By using an additional line with land-color as base this does not happen anymore.
Original location of #488:
before
after
This would also apply to intermittent waterways:
before
after
This can have side-effects when a intermittent river is tagged with a riverbank. Depends if we view that as valid tagging combination.
before
after
Also stream tunnels over forests look a bit strange but there is likely some fine-tuning possible:
