Author |
Message |
24/04/2023 16:56:27
|
dk-geonames
Joined: 16/04/2023 18:21:09
Messages: 2
Offline
|
There don't seem to be any entries for SG within the admin1CodesASCII.txt file.
Based on a geoname admin1 grouping for SG country code I would expect the following entries:
SG.00
SG.01
SG.02
SG.03
SG.04
SG.05
Also, there are null values for admin1 within the geoname records for SG (perhaps others too); is it fair to say that all feature_class = 'P' should be contained within at least a single admin1 level, or is that not a correct assessment?
Thank you for looking.
dk
|
|
|
24/04/2023 17:17:41
|
marc
Joined: 08/12/2005 07:39:47
Messages: 4442
Offline
|
Hi Dk
Singapore does not have admin divisions as such. There are several different kind of divisions for statistics and other topics. The most known are probably the community development councils, which you may refer to. We have modeled them as regions:
https://www.geonames.org/search.html?q=sg+development+council&country=sg
Best Regards
Marc
|
|
|
|
01/05/2023 09:19:27
|
dk-geonames
Joined: 16/04/2023 18:21:09
Messages: 2
Offline
|
Thank you for your feedback, Marc.
Since they are modelled as regions, but masquerade as "admin1" codes; should they not appear within the admin1Codes data? Or alternatively, within some alternatively mapped export?
I see the regions you refer to listed as 'ADMD' feature_class in the main table, however I do not see the ISO 3166-2 value anywhere. (Though it is visible and referenced on the link you provided.)
There are some entries within the geonames table that contain admin1 values for some of these "regions". eg. 03 == North West
At the moment, I am working around this situation by simply excluding the "admin1" layer for countries without any entries and they are falling under an "All Regions" data flow. I could look for 'ADMD' entries in the main table in these cases, or create a separate table à la admin1Codes for them... but without the "admin1" values aligned, it couldn't work.
Am I missing something?
Thanks!
|
|
|
|