Open missinglink opened 9 years ago
The # is a separate between language names (Brussel-Hoofstad and Bruxelles-Capitale)
The actionable here is to search for and split on the #
character and import all names, in lieu of language keys let's just say the first one maps to name.default
in pelias and the others map to name.alt1
etc. (as per geonames)
For Sweden we will need to probably import qs_adm0
as the 'main' english name but also qs_a0
as the Swedish name cc/ @nvkelso
{
"type": "Feature",
"properties": {
"qs_adm0_a3": "SWE",
"qs_adm0": "Sweden",
"qs_level": "adm0",
"qs_iso_cc": "SE",
"qs_a0": "Sverige",
"qs_a0_alt": "Sverige",
"qs_a0_lc": "SE0000",
"qs_source": "EuroGlobalMap",
"qs_pop": null,
"qs_id": null,
"qs_gn_id": null,
"qs_woe_id": null,
"qs_scale": null,
"quad_count": 17919,
"photo_sum": 620170,
"photo_max": 121764,
"localhoods": 96969,
"local_sum": 119403432,
"local_max": 15300552,
"popularity": 620170,
"lat": 62.3599815259509,
"lon": 16.9122043617686
},
"geometry": [<<<truncated>>>]
}
Sounds like we're just going to solve this when we move from Quattroshapes to Who's on First.
Agree that makes the most sense.
On Wed, Sep 23, 2015 at 1:05 PM, David Riordan notifications@github.com wrote:
Sounds like we're just going to solve this when we move from Quattroshapes to Who's on First.
— Reply to this email directly or view it on GitHub https://github.com/pelias/quattroshapes/issues/24#issuecomment-142714062 .
@thisisaaronland Is Who's on First still using "language separators#séparateurs linguistiques" for multilingual countries?
I am not sure I understand the question but if it is:
"Does Who's On First encode multiple language names for a place type in a single string?"
Then the answer is:
"Only as a by-product of a past import and those records will be updated to follow the conventions for naming things. [1]"
Does that answer your question?
That sure does answer the question.
On 1/27/16 15:55, Aaron Straup Cope wrote:
I am not sure I understand the question but if it is:
"Does Who's On First encode multiple language names for a place type in a single string?"
Then the answer is:
"Only as a by-product of a past import and those records will be updated to follow the conventions for naming things. [1]"
Does that answer your question?
[1] https://github.com/whosonfirst/whosonfirst-names
— Reply to this email directly or view it on GitHub https://github.com/pelias/quattroshapes/issues/24#issuecomment-175848564.
David Riordan | Product Manager - Search | Mapzen dave.riordan@mapzen.com | gpg 235D 9DC9 5EF6 277C @riordan | +1 (203)-521-1222 https://mapzen.com | @mapzen
Fixed by WOF, but should be resolved in QS
Side note: QS is no longer maintained, the WOF fixes won't be back ported.
@nvkelso, sorry the last comment wasn't very clearly worded. I meant that it would still need to be resolved in QS manually in the future.
Issue still present in WoF (record).
Should be resolved with concordance + WoE words import.
Is there an open ticket or pull request (in WOF-space) that I've simply managed to space on?
I think there was a ticket to change the locality (which was done), but turns out more same-named elements in the hierarchy also need updating?
On Tue, Apr 5, 2016 at 12:38 PM, Aaron Straup Cope <notifications@github.com
wrote:
Is there an open ticket or pull request (in WOF-space) that I've simply managed to space on?
— You are receiving this because you were mentioned. Reply to this email directly or view it on GitHub https://github.com/pelias/quattroshapes/issues/24#issuecomment-205887208
While we don't import Quattroshapes directly any more, this is still an issue that has been brought into WOF: http://pelias.github.io/compare/#/v1/search%3Ftext=Brussel-Hoofstad
Move issue to pelias/pelias
What should we do with the segment after the
#
?