I was recently looking at the Indian state of Arunachal Pradesh, and noticed that many rivers had been provided with Chinese names. Soon I found Villages and lakes with Chinese names too. This state has been administered by India, since India's independence, but is claimed by China, as part of Tibet, and therefore part of China. I suspect that this claim is related to the fact that some Chinese names have ended up on objects here. I have been working on moving Chinese names to the This is way over my head, as I know next to nothing about overpass queries or character encoding. Can someone lend a hand? asked 12 Jul '17, 08:32  keithonearth edited 13 Jul '17, 08:19 |
2 Answers:
One of the Overpass-API developers runs a server with prototype support for ICU character ranges in regex: https://www.openstreetmap.org/user/mmd/diary/40197 This makes the query straightforward: answered 12 Jul '17, 18:00  maxerickson |
Osmose-QA have check for this, look at: http://osmose.openstreetmap.fr/en/map/#zoom=7&lat=27.858&lon=94.465&layer=Mapnik&overlays=FFFFFFFFFFFFFFFFFFFFT&item=5070&level=1%2C2%2C3&tags=&fixable= The check matchs the language, or the default for the country for "name" tag, with the content. answered 14 Jul '17, 18:53  frodrigo |
I don't see anything like that in the Overpass documentation. Could you just download objects with name tags and do the analysis yourself offline?
Try to download all names, then sort them. This should lead to all Chinese names being next to each other. Maybe you can use the CSV output to generate a list with
name,type,OSM IDor something similar.I guess you could find some of them with a regular expression that tests for some common characters in names of villages.