Skip to content

Latest commit

 

History

History
418 lines (352 loc) · 14.4 KB

README.md

File metadata and controls

418 lines (352 loc) · 14.4 KB
Raw

Geoconvert converts (obvious) data like zipcode, postcode, address, country name, French department name, Americain state name, Canadian province or territory name to generic data.

  • Languages available for country names: deutsch, english, french, portuguese, spanish.
  • Languages available for capital names: deutsch, english, french.

Supports Python 3.8, 3.9, 3.10 and 3.11.

Build Status codecov

Installation

Install the latest version from Github:

pip install https://github.com/jurismarches/geoconvert/archive/master.zip

TL;DR

>>> from geoconvert.convert import (
... 	address_to_country_code,
... 	address_to_subdivision_code,
... 	address_to_country_and_subdivision_codes,
... )
>>> address_to_country_code("Bienvenue à Londres")  # FR
'GB'
>>> address_to_country_code("Bienvenue à Londres", lang="fr")
'GB'
>>> address_to_country_code("Bienvenue à Londres", lang="en")
>>> address_to_country_code("Welcome to Cyprus")  # EN
'CY'
>>> address_to_country_code("Welcome to Los Angeles, California")
'US'
>>> address_to_country_code("Welcome to Los Angeles, CA")
>>> address_to_country_code("Willkommen bei Kairo")  # DE
'EG'
>>> address_to_country_code("Bem vindo ao Afeganistão")  # PT
'AF'
>>> address_to_country_code("Bienvenidos a Nueva Zelanda")  # ES
'NZ'
>>> address_to_country_code("659 Ocean Ave, Lakewood, New Jersey 08701")
'US'
>>> address_to_country_code("Le Chemin des Garennes, Jersey JE3 2FE, Jersey")
'JE'
>>> address_to_subdivision_code("2 pl. Saint-Pierre, 44000 Nantes, France")  # FR
'44'
>>> address_to_subdivision_code("2 pl. Saint-Pierre, 44000 Nantes", country="FR")
'44'
>>> address_to_subdivision_code("2 pl. Saint-Pierre, 44000 Nantes", country="US")
>>> address_to_subdivision_code("1800 W Erie Ave, Lorain, OH 44052")  # US
'OH'
>>> address_to_subdivision_code("1800 W Erie Ave, Lorain, OH 44052", country="FR")
>>> address_to_subdivision_code("196 Voie Camillien-Houde, Montréal, QC H3H 1A1")  # CA
'QC'
>>> address_to_country_and_subdivision_codes("1170 Cattus Island Blvd, Toms River, New Jersey")  # US
('US', 'NJ')
>>> address_to_country_and_subdivision_codes("Montréal, Québec")  # CA
('CA', 'QC')
>>> address_to_country_and_subdivision_codes("1800 W Erie Ave, Lorain, OH 44052")  # US
('US', 'OH')
>>> address_to_country_and_subdivision_codes("96524 Föritztal OT Neuhaus-Schierschnitz")  # DE not found
(None, None)
>>> address_to_country_and_subdivision_codes("96524 Föritztal OT Neuhaus-Schierschnitz Germany")  # DE found
('DE', 'TH')
>>> address_to_country_and_subdivision_codes("Kairo", lang="de")
('EG', None)
>>> address_to_country_and_subdivision_codes("Kairo", lang="de", country="EG")
('EG', None)
>>> address_to_country_and_subdivision_codes("Kairo", lang="de", country="US")
(None, None)

Usage examples

Geoconvert mainly provides mainly two types of functions:

  • one converts an input containing a country or capital name to a two-letter country code (ISO 3166-1 alpha 2) using a set of available languages.
  • the other converts an address of a given country to a subdivision code (department code for France, province or territory code for Canada, state code for the United States of America).

Some usage examples are shown below. For more examples, take a look at the unit tests.

Finding country codes

The main function you are looking for is the following, converting a country or a capital name to a country code. By default, it uses all the available languages to guess the country code:

>>> from geoconvert.convert import address_to_country_code
>>> address_to_country_code("Bienvenue à Kinshasa")  # FR
'CD'
>>> address_to_country_code("Welcome to Cyprus")  # EN
'CY'
>>> address_to_country_code("Willkommen bei Kairo")  # DE
'EG'
>>> address_to_country_code("Bem vindo ao Afeganistão")  # PT
'AF'
>>> address_to_country_code("Bienvenidos a Nueva Zelanda")  # ES
'NZ'

If you known the language of your input data, you may select that language for more efficiency (available choices: "de" for german, "en" for english, "fr" for french, "pt" for portuguese):

>>> address_to_country_code("Bienvenue à Kinshasa", lang="fr")
'CD'
>>> address_to_country_code("Welcome to Cyprus", lang="en")
'CY'
>>> address_to_country_code("Willkommen bei Kairo", lang="de")
'EG'
>>> address_to_country_code("Bem vindo ao Afeganistão", lang="pt")
'AF'
>>> address_to_country_code("Bienvenidos a Nueva Zelanda", lang="es")
'NZ'

There are more specific functions, for which you can also select a language (all of them are used by default):

>>> from geoconvert.convert import (
... 	country_name_to_country_code, capital_name_to_country_code
... )
>>> country_name_to_country_code(" Le  nigéria c'est trop   sympa", lang="fr")
'NG'
>>> capital_name_to_country_code('Kairo', lang="de")
'EG'

Finding subdivision codes

Geoconvert also gives you the abilty to find the code associated to a smaller part of a given country, namely:

  • the province or territory code for Canada,
  • the department code for France,
  • the department codes of a French region,
  • the state code for the United States of America.

The main function you are looking for is address_to_subdivision_code. It converts an address to a subdivision code. By default, it uses all the available subdivisions to guess the subdivision code:

>>> address_to_subdivision_code("2 pl. Saint-Pierre, 44000 Nantes, France")  # FR
'44'
>>> address_to_subdivision_code("1800 W Erie Ave, Lorain, OH 44052")  # US
'OH'
>>> address_to_subdivision_code("196 Voie Camillien-Houde, Montréal, QC H3H 1A1")  # CA
'QC'

There should be no confusion between French and US postcodes:

>>> address_to_subdivision_code("2 pl. Saint-Pierre, 44000 Nantes", country="US")
>>> address_to_subdivision_code("1800 W Erie Ave, Lorain, OH 44052", country="FR")
>>> address_to_subdivision_code("96524 Föritztal OT Neuhaus-Schierschnitz", country="US")

Brazil

You can find the state code associated with an address, via the postcode, state name (in Portuguese) or code:

>>> from geoconvert.convert import br_address_to_state_code
>>> br_address_to_state_code("Luz, 01120-010")
'SP'
>>> br_address_to_state_code("Piauí")
'PI'
>>> br_address_to_state_code("Dourados, MS")
'MS'

You may use more specific functions, depending on your needs:

>>> from geoconvert.convert import (
... 	br_postcode_to_state_code, br_state_name_to_state_code
... )
>>> br_postcode_to_state_code("Mariana, 04094-050")
'SP'
>>> br_state_name_to_state_code("a capital do estado do maranhao.")
'MA'

Canada

You can find the province or territory code associated with an address, via the postcode, province or territory name (in French or English) or code:

>>> from geoconvert.convert import ca_address_to_province_code
>>> ca_address_to_province_code("H3T 1X6")
'QC'
>>> ca_address_to_province_code("Toronto, Ontario")
'ON'
>>> ca_address_to_province_code("Toronto, ON")
'ON'
>>> ca_address_to_province_code("Québec")
'QC'
>>> ca_address_to_province_code("Terre-Neuve-et-Labrador")
'NL'
>>> ca_address_to_province_code("Newfoundland and Labrador")
'NL'
>>> ca_address_to_province_code("Bldg 158, Iqaluit, NU X0A 0H0, Canada")
'NU'
>>> ca_address_to_province_code("52-58 Franklin Rd, Inuvik, NT X0E 0T0, Canada")
'NT'

You may use more specific functions, depending on your needs:

>>> from geoconvert.convert import (
... 	ca_postcode_to_province_code, ca_province_name_to_province_code
... )
>>> ca_postcode_to_province_code("Iqaluit X0A 0H0")
'NU'
>>> ca_postcode_to_province_code("Inuvik X0E 0T0")
'NT'
>>> ca_province_name_to_province_code("Welcome to Ontario")
'ON'

France

You can find the department code associated with an address or department name:

>>> from geoconvert.convert import fr_address_to_dept_code
>>> fr_address_to_dept_code("Chemin du Solarium Le Haut Vigneau 33175 GRADIGNAN CEDEX")
'33'
>>> fr_address_to_dept_code("Loire-Atlantique")
'44'
>>> fr_address_to_dept_code("LOIRE    -    ATLANTIQUE")
'44'
>>> fr_address_to_dept_code("Loire    -    ")
'42'

You can also derive the main department and the list of all departments which are part of a given region from its name:

>>> from geoconvert.convert import fr_region_name_to_info
>>> # Returns the deparment code where the head of the region is
>>> # and the list of all the departments in the region.
>>> fr_region_name_to_info('Les Pays de la Loire, une superbe région')
('44', ['44', '49', '53', '72', '85'])

You may use more specific functions, depending on your needs:

>>> from geoconvert.convert import (
... 	fr_address_to_dept_code,
... 	fr_dept_name_to_dept_code,
... 	fr_postcode_to_dept_code,
... )
>>> fr_postcode_to_dept_code("Chemin du Solarium Le Haut Vigneau 33175 GRADIGNAN CEDEX")
'33'
>>> fr_dept_name_to_dept_code("Loire    -    Atlantique")
'44'
>>> fr_dept_name_to_dept_code("Loire    -    ")
'42'

United States of America

You can find the state code associated with an address, via the postcode, state name (in English) or code:

>>> from geoconvert.convert import us_address_to_state_code
>>> us_address_to_state_code("Sunnyvale, CA 94085")
'CA'
>>> us_address_to_state_code("New Hampshire")
'NH'
>>> us_address_to_state_code("Los Angeles, CA")
'CA'

You may use more specific functions, depending on your needs:

>>> from geoconvert.convert import (
... 	us_postcode_to_state_code, us_state_name_to_state_code
... )
>>> us_postcode_to_state_code("6931 Rings Rd, Amlin, OH 43002")
'OH'
>>> us_state_name_to_state_code("Welcome to West Virginia")
'WV'

Mixing both country codes and subdivision codes

You can derive both country and subdivision codes at the same time:

>>> from geoconvert.convert import address_to_country_and_subdivision_codes
>>> address_to_country_and_subdivision_codes("Montréal, QC, Canada")
('CA', 'QC')
>>> address_to_country_and_subdivision_codes("Montréal, Québec")
('CA', 'QC')
>>> address_to_country_and_subdivision_codes("Montréal, QC")
(None, None)
>>> address_to_country_and_subdivision_codes("Montréal, QC", country="CA")
('CA', 'QC')
>>> address_to_country_and_subdivision_codes("Montréal", country="CA")
(None, None)
>>> address_to_country_and_subdivision_codes("Montréal, QC", country="US")
(None, None)
>>> address_to_country_and_subdivision_codes("Kairo", lang="de")
('EG', None)
>>> address_to_country_and_subdivision_codes("Kairo", lang="de", country="EG")
('EG', None)
>>> address_to_country_and_subdivision_codes("Kairo", lang="de", country="US")
(None, None)

Result format can be choosen between a tuple and an iso code

>>> address_to_country_and_subdivision_codes("14467 Potsdam")
('DE', 'BB')
>>> address_to_country_and_subdivision_codes("14467 Potsdam", iso_format=True)
'DE-BB'
>>> address_to_country_and_subdivision_codes("Eschborn Germany")
('DE', None)
>>> address_to_country_and_subdivision_codes("Eschborn Germany", iso_format=True)
'DE'

There should be no confusion between french, deutsch and US postcodes:

>>> address_to_country_and_subdivision_codes("2 pl. Saint-Pierre, 44000 Nantes", country="US")
(None, None)
>>> address_to_country_and_subdivision_codes("6931 Rings Rd, Amlin, OH 43002", country="FR")
(None, None)
>>> address_to_country_and_subdivision_codes("Straße 3 53119 Bonn", country="US")
(None, None)

For developers

Tests

Just run pytest:

make test

Do not forget to install the dependencies first, preferably in a vitrual environment:

python3 -m venv .venv
source .venv/bin/activate
python -m pip install -r requirements_test.txt

Releases

To release a new version you must update __version__ on geoconvert/__init__.py and tag the master branch with the same version.

Adding a new language for countries and/or capitals

  • Add the data in dictionaries, where the key is the name, and the value is the corresponding country code.
  • These dictionaries must be named countries_<language_code> in geoconvert/data/countries.py and/or capitals_<language_code> in geoconvert/data/capitals.py. The language code is a two-letter, lowercase code. For instance, for italian, the language code is "it".
  • Use already existing data as a template, especially to handle special_countries.
  • Add the corresponding dictionnary to language_to_country_names and/or language_to_capital_names. This allows the address_to_country_code, country_name_to_country_code, and capital_name_to_country_code functions to use this language.
  • Add the language to the list of available languages at the top of this README.
  • Add examples which are specific for this language in the README, doctests and tests, especially for address_to_country_code and address_to_country_and_subdivision_codes.

Adding subdivisions to a new country

  • Add the data in dictionaries in geoconvert/data/subdivisions/<country_name>.py, where <country_name> is the lowercase country name in English (for instance, geoconvert/data/subdivisions/germany.py for Germany).
  • Add functions that allow you convert an address to a subdivision code :
    • a base function <country_code>_address_to_<subdivision_name>_code that uses the other implemented functions converting an address to a subdivision code (e.g., ca_address_to_province_code)
    • a function converting the full subdivision name to the subdivision code (e.g., ca_province_name_to_province_code)
    • a function converting a postcode from that country to a subdivision (e.g., ca_postcode_to_province_code). This is optionnal, but highly recommended.
    • optionnally, a regex finding the subdivision code directly from the text can be used as the last option in <country_code>_address_to_<subdivision_name>_code.
  • Add the base function to the country_to_subdivision_lookup_function dictionary under the corresponding uppercase country code key. This allows this function to be used by the address_to_subdivision_code and address_to_country_and_subdivision_codes functions.
  • Add the newly available subdivisions to the list at the top of this README.
  • Add examples which are specific for this country in the README, doctests and tests, especially the address_to_subdivision_code and address_to_country_and_subdivision_codes functions. Each added function must be tested as well.