Language

Reference

Human language is a complex subject, and it has a lot of influence on how we search for addresses. Consider a country which uses many languages and we reverse geocode in that country, which language should be returned?

Language Introduction

The following table shows which countries support which languages, what the default language is, and which languages are supported by Naurt’s parser.

Code	Country	Supported Languages	Default Language	Parser Support
AE	United Arab Emirates	English, Arabic	Arabic	English
AT	Austria	German	German	German
AU	Australia	English	English	English
BE	Belgium	French, Dutch, German*	French, Dutch, German*	French, Dutch, German
BR	Brazil	Portuguese	Portuguese	Portuguese
CA	Canada	English*	English*	English*
CH	Switzerland	French, Italian, German*	French, Italian, German*	French, Italian, German
DE	Germany	German	German	German
DK	Denmark	Danish	Danish	Danish
EG	Egypt	English, Arabic	Arabic	English
ES	Spain	Spanish	Spanish	Spanish
FI	Finland	Finnish	Finnish	Finnish
FR	France	French	French	French
GB (UK)	United Kingdom	English	English	English
GG	Guernsey	English	English	English
HK	Hong Kong	Chinese	Chinese
IE	Ireland	English	English	English
IM	Isle of Man	English	English	English
IT	Italy	Italian	Italian	Italian
JE	Jersey	English	English	English
JP	Japan	Japanese	Japanese
KR	South Korea	Korean	Korean
NL	Netherlands	Dutch	Dutch	Dutch
NO	Norway	Norwegian	Norwegian	Norwegian
NZ	New Zealand	English	English	English
PT	Portugal	Portuguese	Portuguese	Portuguese
SA	Saudi Arabia	English, Arabic	Arabic	English
SE	Sweden	Swedish	Swedish	Swedish
SG	Singapore	English	English	English
TW	Taiwan	Chinese	Chinese
US	United States	English	English	English

Canada, Belgium and Switzerland are special cases, please see those sections

Searching with an Address

If you include an address in your search, the general rule is that the returned address will be in the same language as the search language.

Some languages support the parser. If you are not using a parsed request, Naurt will attempt to first parse your address using our parser. In general, a parsed address has much better search accuracy than an unparsed address. If a language does not support a parser, it is because in general a parser is not required. Chinese, for example, generally has unique enough addresses to get very good search performance without the use of a parser.

If any Arabic is used in searching countries with Arabic support, the entire address search will be considered Arabic, and the parser will not be used. The same is true for any Chinese, Korean or Japanese characters.

Reverse Geocode

When reverse geocoding, there may be a need to specify a language desired. To avoid having mixed languages appear when reverse geocoding, Naurt has selected a default language per country. If no language is specified, the default language in the above template will be used. If forward geocoding, then there is no need to specify a language as it will use whichever language was searched with. Naurt does not offer translation services.

The language specification is done using ISO 639 Language Codes. The currently supported languages are

AR - Arabic
AUTO - Default (uses the default language for any country - this option is used if language is not specified)
DA - Danish
DE - German
EN - English
ES - Spanish
FI - Finnish
FR - French
IT - Italian
JA - Japanese
KO - Korean
NL - Dutch
NO - Norwegian
PT - Portuguese
ZH - Chinese

Please be aware that the language field will under some circumstances cause nothing to be found. For example, using the English language tag in Japan nothing will be found, as there are no English addresses in Japan.

This can be specified using the language field in a request

{
    "location": {
        "latitude": 24.49,
        "longitude": 54.41
    },
    "language": "EN"
}

If you do not specify a language, then AUTO will be used. In other words

{
    "location": {
        "latitude": 24.49,
        "longitude": 54.41
    }
}

is equivalent to

{
    "location": {
        "latitude": 24.49,
        "longitude": 54.41
    },
    "language": "AUTO"
}

AUTO will use the default language, as specified in the table above.

Belgium and Switzerland Special Case

Belgium and Switzerland both have three languages.

For Belgium this is French, Dutch (also known as Flemish in Belgium) and German. French is mostly spoken in the South, Dutch in the North, and German in small portions of the East.

In Switzerland the languages are French, German and Italian. French is mostly spoken in the West, German in the North, East and Centre, and Italian in small portions of the South.

In Belgium and Switzerland, the languages are non-overlapping in the Naurt coverage. For example, in the French speaking region of Belgium there are no Dutch or German addresses. Therefore, when reverse geocoding in Belgium or Switzerland our suggestion is to not specify a language and use Auto.

Canada Special Case

Unfortunately, due to upstream address suppliers, Canadian addresses are mixed English/French. In general, addresses in Québec are in French and addresses elsewhere are in English. Due to upstream restrictions, Canada is only classed in the API as having English support. Therefore, when reverse geocoding in Canada, never specify French otherwise no results will be found.