No structural cost value/type needed for population exposure models

[Muharem Hrnjadovic]

[12:24] <vitorsilva> al-maisan: Hi Muharem, do you have a sec?
[12:24] <al-maisan> hello vitorsilva, yes, sure
[12:25] <vitorsilva> I trying to store population using the NRML 0.3 exposure model schema. I set up the assetCategory = population and then each asset was defined as:
[12:25] <vitorsilva> <assetDefinition gml:id="a2">
[12:25] <vitorsilva> <site>
[12:25] <vitorsilva> <gml:Point srsName="epsg:4326">
[12:25] <vitorsilva> <gml:pos>28.6975 40.9825</gml:pos>
[12:25] <vitorsilva> </gml:Point>
[12:25] <vitorsilva> </site>
[12:25] <vitorsilva> <number>1</number>
[12:25] <vitorsilva> <taxonomy>VF1</taxonomy>
[12:25] <vitorsilva> </assetDefinition>
[12:26] <vitorsilva> but I get un error: ValueError: Invalid input: 'REXD(cost=None, cost_type=None, area=None, area_type=None, number_of_units=1.0)'
[12:26] * al-maisan looks
[12:28] <al-maisan> vitorsilva: that error comes from inside the asset value calculation function
[12:28] <al-maisan> vitorsilva: I believe the problem is that the cost type is not set
[12:29] <vitorsilva> ok I can set that up but for population there is no cost type… it's just people count..
[12:30] <al-maisan> I see .. the asset value calculation function is buggy .. it always expects the cost type to be set
[12:30] * al-maisan will file a bug
[12:31] <al-maisan> vitorsilva: what is the correct behaviour in this case>
[12:31] <al-maisan> ?
[12:32] <vitorsilva> stco and area are not required, only the attribute "number"
[12:32] <al-maisan> in that case the number is the per-asset-value?
[12:36] <vitorsilva> yes
[12:36] <al-maisan> vitorsilva: if that is right the easiest work-around would be to set stco=1 stcoType=per_asset
[12:36] <vitorsilva> ok
[12:36] <vitorsilva> al-maisan: is it ok to open a bug on LP to fix this?
[12:36] <al-maisan> sorry about that
[12:37] <al-maisan> vitorsilva: of course!
[12:37] <al-maisan> vitorsilva: Will you file the bug or shall I?
[12:38] <al-maisan> vitorsilva: the NRML file that you expected to work should be attached to the bug in any case
[12:38] <vitorsilva> al-maisan: I guess it will be more efficient if you do it
[12:38] <al-maisan> can you please email me the exposure NRML file and I will do it
[12:38] <vitorsilva> ok, send me the link and I'll attach it
[12:38] <al-maisan> even better

[Vitor Silva (vitor-silva)]

This bug fix should allow the use of the attached model

[Muharem Hrnjadovic (al-maisan)]

[12:50] <al-maisan> vitorsilva: what we discussed above only applies to population exposure models, right?
[12:51] <vitorsilva> yap, only when assetCategory is set to "population"