Closed nguyandy closed 2 years ago
nguyandy
commented
2 years ago 
ndellicarpini
commented
2 years ago
- ChesROMS Chlorophyll 3D returning blank tiles
This is a unit issue
- traffic lights don't work with climatology layers
This is fixed in an open pyxms pr
- any styling outside of the default does not work with ChesROMS Chlorophyll 3D (this might be related to Changing style on some layers doesn't work #312)
This is a unit issue
ndellicarpini
commented
2 years ago For chlorophyll units, should we just override them to mg/m^3 rather than mg chl/m^3, or should we take some other sort of approach (maybe using some sort of supportsUnitConversion)? I'm unsure how to handle this.
jamesdoyle21
commented
2 years ago mg/m^3 makes more sense to me. That should be plenty of information for the user
ndellicarpini
commented
2 years ago mg/m^3 makes more sense to me. That should be plenty of information for the user
@nguyandy do you think we should handle this in OM, or should we try to get the units to change on the data side (sorry if that doesn't make sense, I'm not entirely sure how units are added to the data).
jamesdoyle21
commented
2 years ago I'm still not getting anything for ChesROMS Chlorophyll 3D
ndellicarpini
commented
2 years ago I'm still not getting anything for ChesROMS Chlorophyll 3D
Both of the chesroms layers are broken (alkalinity doesn't support changing the style) because they use units we don't/won't support. We still need to figure out how to handle layers with oddball units.
nguyandy
commented
2 years ago @nguyandy do you think we should handle this in OM, or should we try to get the units to change on the data side (sorry if that doesn't make sense, I'm not entirely sure how units are added to the data).
Sorry Nick I never saw this comment. Changing the units in the data is probably the "right" way to do this. Changing the units in OM3 would be the fastest and easiest way.
@mpiannucci thoughts?
mpiannucci
commented
2 years ago Is this as simple as adding the oddball units to OM3 unit system? It is extendable right?
nguyandy
commented
2 years ago Should be, I'll give it swing in a bit
nguyandy
commented
2 years ago Ahhh it's all coming back to me now.
So the units we get from ncwms/pyxms is mg_chl/m3
The unit being passed in the OM3 requests is mg chl/m³
To which we get the error
Cannot convert units mg chl/m³ to mg_chl/m3
When we make the changes to OM3 to use the original unit mg_chl/m3, we still get an error from pyxms
Cannot convert units mg_chl/m3 to mg_chl/m3
But when we take out the unit parameter, the request works https://wms.oceansmap.com/wms?service=WMS&request=GetMap&version=1.1.1&layers=CHESROMS_ECB/chlorophyll&styles=raster/algae&colorscalerange=0,30&width=256&height=256&format=image/png&transparent=true&time=2021-07-10T00:00:00.000Z&srs=EPSG:3857&bbox=-8766409.899970295,3757032.814272985,-8140237.7642581295,4383204.9499851465
@ndellicarpini suggested last week that we add a field into the data-catalog, supportsUnitConversion for layers where this might be problematic. Where OM3 will check supportsUnitConversion before tackling on the units param in the request + can notify users if unit conversions don't work for certain layers.
This would fix our issue. Another option would be to update pyxms to bypass/handle these failures - but that means users won't know if unit conversions are working or not.
mpiannucci
commented
2 years ago That sounds like a pyxms error to me, I don't think that should error out. I can take a look
mpiannucci
commented
2 years ago So I just pushed a pyxms fix that will not try and convert units when the users requested units are the same as the source units. This means that for certain weird unit layers we will just not have them toggle to other systems for now and the layers should then work out of the box, just with no conversions