Harmonize city naming

[MoritzStefaner]

City names should be harmonized.

Source	Spelling
GrowBikeNet filenames	amsterdam, andorra-la-vella
FixBikeNet / LinkBikeNet filenames	Brussels, Budapest
SuperBlockify filenames	Amsterdam_betweenness.gpkg
rec_table.csv	andorra-la-vella
european_capitals.csv	Andorra la Vella

Multi-token names are where this breaks. Our parser reads the city as everything before the first
hyphen, so andorra-la-vella-city_boundary.geojson yields city andorra and layer
la-vella-city_boundary. The boundary is then unrecognised and the import aborts. We work around
that with an alias table in two scripts plus a rename pass over pulled files.

Ask: one lowercase ASCII slug per city, identical across algorithms, filenames and CSVs. Also
consider writing it into each file as a city property, so filenames stop being the only source of truth.

[mszell]

Thx, good to know about the hyphens. I wanted to improve our slugifier anyway, will do so. We use then the same function everywhere.

So, just to be clear, your ask for andorra would be to name it: andorralavella, correct?
Given the country issue, is it ok then to name it andorralavella_ad?

[mszell]

Also consider writing it into each file as a city property

You mean the geojson exports?

[MoritzStefaner]

Given the country issue, is it ok then to name it andorralavella_ad?

Yes, whatever the ID for that city is in the cities.csv table is fine. And let's keep in mind that those IDs should not contain the finename separator.

[MoritzStefaner]

Also consider writing it into each file as a city property

You mean the geojson exports?

Yes, I thought it might be safer/cleaner to add the parameters as properties, but apparently it is not allowed to have properties on a FeatureCollection, so maybe let's stay with the filename based encoding.

[mszell]

As per our new NAMING_CONVENTIONS, lets use everywhere the slugifier of growbikenet for this (not yet in release, but live on main). ..To be centralized into bikenetlib.