Closed ethervoid closed 7 years ago
@rafatower pointed that if the source are polygons could be problems to get an output:
When reporting issues it is quite important to specify what input data you use, at least if it is points or polygons (even better: provide a .carto file), because that affects the DO ability to return meaningful data.
From the "Data Observatory API Architecture Guide":
Unfortunately we may have some trouble reflecting that in the UI.
This could be an issue because of the above-mentioned dataset, walmart
, only have points so it seems like something is not right.
I've executed the obs_getdata
query and I'm getting this messages:
NOTICE: 00000: cdb_dataservices_server._obs_getdata(5): [obs_2017_05_04_f20ee1131c] REMOTE NOTICE: Cannot perform calculation over polygon for us.zillow.SingleFamilyResidenceRental_Zri//us.census.tiger.zcta5/2016-08
Seems like it's thinking we're using polygons when the dataset is made of points. I'll keep digging
I've figured out what the problem is. The geomvals
generated with NULLs are interfering and making the OBS_GetData function think that is dealing with Polygons.
These are the geomvals
with NULL values:
{"(,1058)","(,1645)","(0101000020E61000009510ACAA97DB55C0274EEE7728D44340,977)","(,1017)","(0101000020E610000016A1D80A9A4E55C0BA9F53909F054140,699)","(0101000020E6100000D95E0B7A6FDD54C0F9669B1BD31F4140,3)","(0101000020E6100000EF92382BA26854C0FB0626378A5A4140,605)","(0101000020E610000064E597C1189556C03140A209144F4340,3005)","(0101000020E6100000E23E726BD2F55DC0F58079C894674240,1769)","(0101000020E610000055A69883A0E057C0D787F546ADD44040,145)","(0101000020E6100000DB317557761A57C0267156444D904140,82)","(0101000020E6100000E7C2482F6A9C54C0B62DCA6C90034440,1550)","(,2568)","(0101000020E61000005D70067FBFEC55C0CE1C925A285D4340,527)","(,1212)","(0101000020E61000002272FA7ABEFA54C0732B84D558D84340,1420)","(0101000020E61000005DA8FC6B798957C09B8F6B43C52A4240,1)","(0101000020E61000005853591476DD57C0BB287AE063AC3D40,2505)","(0101000020E6100000766EDA8CD33D57C01A1538D9061A4240,2)","(0101000020E610000079B29B19FD2B58C01EFCC401F4873E40,544)","(0101000020E6100000D9791B9B1D055EC0213EB0E3BF7C4240,1539)","(,1586)","(0101000020E61000005BB4006DABD057C0F73DEAAF57764340,464)","(0101000020E6100000790261A758C352C0FD6A0E10CCF74340,3099)","(0101000020E61000006342CC2555D451C0F05014E813C14440,2623)","(0101000020E6100000BA8784EFFD9B54C0FCE25295B61A4340,2553)","(0101000020E6100000E90B21E7FD1154C05E0F26C5C7A34340,1501)","(0101000020E6100000E318C91EA13D52C0917EFB3A70CE4540,2348)","(,2590)","(,2043)"}
and these are the geomvals
without NULL values:
{"(0101000020E61000009510ACAA97DB55C0274EEE7728D44340,977)","(0101000020E610000016A1D80A9A4E55C0BA9F53909F054140,699)","(0101000020E6100000D95E0B7A6FDD54C0F9669B1BD31F4140,3)","(0101000020E6100000EF92382BA26854C0FB0626378A5A4140,605)","(0101000020E610000064E597C1189556C03140A209144F4340,3005)","(0101000020E6100000E23E726BD2F55DC0F58079C894674240,1769)","(0101000020E610000055A69883A0E057C0D787F546ADD44040,145)","(0101000020E6100000DB317557761A57C0267156444D904140,82)","(0101000020E6100000E7C2482F6A9C54C0B62DCA6C90034440,1550)","(0101000020E61000005D70067FBFEC55C0CE1C925A285D4340,527)","(0101000020E61000002272FA7ABEFA54C0732B84D558D84340,1420)","(0101000020E61000005DA8FC6B798957C09B8F6B43C52A4240,1)","(0101000020E61000005853591476DD57C0BB287AE063AC3D40,2505)","(0101000020E6100000766EDA8CD33D57C01A1538D9061A4240,2)","(0101000020E610000079B29B19FD2B58C01EFCC401F4873E40,544)","(0101000020E6100000D9791B9B1D055EC0213EB0E3BF7C4240,1539)","(0101000020E61000005BB4006DABD057C0F73DEAAF57764340,464)","(0101000020E6100000790261A758C352C0FD6A0E10CCF74340,3099)","(0101000020E61000006342CC2555D451C0F05014E813C14440,2623)","(0101000020E6100000BA8784EFFD9B54C0FCE25295B61A4340,2553)","(0101000020E6100000E90B21E7FD1154C05E0F26C5C7A34340,1501)","(0101000020E6100000E318C91EA13D52C0917EFB3A70CE4540,2348)"}
So the way to deal with this in to remove NULL values here and here
Doing the query filtering the NULL values I was able to get some results:
singlefamilyresidencerental_zri_2016_08
-----------------------------------------
1076
934
954
1041
1213
1301
890
741
941
955
1187
1173
1604
1489
2490
2329
1103
1914
[NULL]
[NULL]
[NULL]
[NULL]
[NULL]
[NULL]
[NULL]
[NULL]
[NULL]
[NULL]
[NULL]
[NULL]
(30 rows)
PR that fix this problem cartodb/camshaft#312
Analysis Only Returns Nulls
I ran an analysis that looked at Zillow Rental Value Index Single Family (2016-8 timespan) and Non-US Citizen Population (2011-2015 timespan) on the Wal-Mart Dataset. Only NULL values where returned.
I tried this for a couple of different maps, datasets, and analysis and was never able to get proper values to appear in the new column.
Reported by @kevin-reilly