def test_obsolete():
cc = coco.CountryConverter()
assert len(cc.data) == 250
cc = coco.CountryConverter(include_obsolete=False)
assert len(cc.data) == 250
cc = coco.CountryConverter(include_obsolete=True)
assert len(cc.data) == 256
def test_additional_country_file():
converter_basic = coco.CountryConverter()
converter_extended = coco.CountryConverter(additional_data=custom_data)
assert converter_basic.convert("Congo") == "COG"
assert converter_extended.convert("Congo") == "COD"
assert converter_extended.convert("wirtland",
to="name_short") == "Wirtland"
def test_additional_country_file():
converter_basic = coco.CountryConverter()
converter_extended = coco.CountryConverter(additional_data=custom_data)
assert converter_basic.convert('Congo') == 'COG'
assert converter_extended.convert('Congo') == 'COD'
assert converter_extended.convert('wirtland',
to='name_short') == 'Wirtland'
def test_parser():
sys.argv = ["AT", "US"]
args = _parse_arg(coco.CountryConverter().valid_class)
assert args.src == None # noqa
assert args.to == None # noqa
sys.argv = ["EXIO1"]
args = _parse_arg(coco.CountryConverter().valid_class)
assert args.src == None # noqa
assert args.to == None # noqa
def test_parser():
sys.argv = ['AT', 'US']
args = _parse_arg(coco.CountryConverter().valid_class)
assert args.src == None # noqa
assert args.to == None # noqa
sys.argv = ['EXIO1']
args = _parse_arg(coco.CountryConverter().valid_class)
assert args.src == None # noqa
assert args.to == None # noqa
def test_get_correspondence_dict_numeric_replace():
"""Numeric replacement test of get_correspondence_dict method"""
classA = "EXIO1"
classB = "OECD"
cc = coco.CountryConverter()
corr_str = cc.get_correspondence_dict(classA=classA,
classB=classB,
replace_numeric=True)
assert type(corr_str) == dict
assert len(corr_str) == 44
assert corr_str["JP"] == ["OECD"]
assert corr_str["ZA"] == [None]
assert None in corr_str["WW"]
assert "OECD" in corr_str["WW"]
assert len(corr_str["WW"]) == 2
corr_num = cc.get_correspondence_dict(classA=classA,
classB=classB,
replace_numeric=False)
assert type(corr_num) == dict
assert len(corr_num) == 44
assert corr_num["JP"] == [1964]
assert pd.isna(corr_num["ZA"])
assert 2010 in corr_num["WW"]
assert 1961 in corr_num["WW"]
assert len(corr_num["WW"]) == 4
def add_country(update, context):
"""Adds country to job queue."""
CC = coco.CountryConverter()
global COLLECTING_DATA
chat_id = update.message.chat_id
message = update.message.text
try:
if message.lower() == 'usa':
message = 'United states'
Country(CC.convert(message, to='ISO2'))
except ValueError:
update.message.reply_text('Such a name does not exist')
return ConversationHandler.END
try:
if message in COLLECTING_DATA[chat_id]:
update.message.reply_text('You have already '
'selected this country!')
else:
COLLECTING_DATA[chat_id].append(message)
except KeyError:
COLLECTING_DATA[chat_id] = [message]
text = 'You have added a country!\nYour current list of countries:\n'
for country in COLLECTING_DATA[chat_id]:
text += country + ', '
text = text[:-2] + '.'
update.message.reply_text(text)
return ConversationHandler.END
def coco(name):
"""
'U.S.','United States of America' -> 'United States', etc.
"""
# Remove a leading "the" from names. Note that even for non-countries this is desirable for readability.
name = re.sub(r'^(the\s)', '', name, flags=re.IGNORECASE)
# Remove ending "'s" from names, which is present for some
name = re.sub(r'''('s)$''', '', name)
# Custom replacements that country_converter can't seem to handle
custom_dic = {
'US': 'United States',
'UK': 'United Kingdom',
'EU': 'European Union',
'E.U.': 'European Union'
}
if name in custom_dic:
name = custom_dic[name]
logger.disabled = True # Fixes annoying warnings given by the country_converter module
country = country_converter.CountryConverter().convert(name,
to='name_short',
not_found=None)
logger.disabled = False
# If multiple countries are identified, a list of them is returned
# we'll just take the first country identified in this case
if type(country) == list:
country = country[0]
return country
def test_alternative_names(get_regex_test_data):
converter = coco.CountryConverter(include_obsolete=True)
not_found_id = "XXX"
for row in get_regex_test_data.data.iterrows():
name_test = row[1].name_test
name_short = row[1].name_short
name_result = converter.convert(
name_test,
src="regex",
to="name_short",
not_found=not_found_id,
enforce_list=False,
)
assert len(name_result) > 2, (
"File {0} - row {1}: Name {2} matched several "
"regular expressions: {3}".format(get_regex_test_data.data_name,
row[0], name_test,
" ,".join(name_result)))
if name_short != not_found_id:
assert name_result != not_found_id, (
"File {0} - row {1}: Name {2} did not match any "
"regular expression".format(get_regex_test_data.data_name,
row[0], name_test))
assert name_result == name_short, (
"File {0} - row {1}: Name {2} did match the "
"wrong regular expression: {3}".format(
get_regex_test_data.data_name, row[0], name_test, name_result))
def patchCountries(df, patches=None):
if False:
if patches is None: patches = {
'F.S. Micronesia':'Micronesia',
'Micronesia, Federated States of':'Micronesia',
'Holy See':'Vatican City',
'Congo, Republic of the':'Congo',
'Congo Brazzaville':'Congo',
'Democratic Republic of the Congo':'DR Congo',
'Congo, Democratic Republic of the':'DR Congo',
'Congo Kinshasa':'DR Congo',
'The Gambia':'Gambia',
"People's Republic of China":'China',
'Bahamas, The':'Bahamas'
}
applyToCol(df, 'country', lambda x: patches.get(x, x))
else:
cc: coco.CountryCoverter = coco.CountryConverter()
countries = list(df.country)
conv = cc.convert(countries, to = 'name_short', not_found = None, src = 'regex')
df.country = conv
# cc.convert()
return df
def test_get_correspondance_dict_numeric_replace():
""" Numeric replacement test of get_correspondance_dict method
"""
classA = 'EXIO1'
classB = 'OECD'
cc = coco.CountryConverter()
corr_str = cc.get_correspondance_dict(classA=classA,
classB=classB,
replace_numeric=True)
assert type(corr_str) == dict
assert len(corr_str) == 44
assert corr_str['JP'] == ['OECD']
assert corr_str['ZA'] == [None]
assert None in corr_str['WW']
assert 'OECD' in corr_str['WW']
assert len(corr_str['WW']) == 2
corr_num = cc.get_correspondance_dict(classA=classA,
classB=classB,
replace_numeric=False)
assert type(corr_num) == dict
assert len(corr_num) == 44
assert corr_num['JP'] == [1964]
assert pd.np.isnan(corr_num['ZA'])
assert 2010 in corr_num['WW']
assert 1961 in corr_num['WW']
assert len(corr_num['WW']) == 4
def test_alternative_names(get_regex_test_data):
converter = coco.CountryConverter(include_obsolete=True)
not_found_id = 'XXX'
for row in get_regex_test_data.data.iterrows():
name_test = row[1].name_test
name_short = row[1].name_short
name_result = converter.convert(name_test,
src='regex',
to='name_short',
not_found=not_found_id,
enforce_list=False)
assert len(name_result) > 2, (
'File {0} - row {1}: Name {2} matched several '
'regular expressions: {3}'.format(get_regex_test_data.data_name,
row[0], name_test,
' ,'.join(name_result)))
if name_short != not_found_id:
assert name_result != not_found_id, (
'File {0} - row {1}: Name {2} did not match any '
'regular expression'.format(get_regex_test_data.data_name,
row[0], name_test))
assert name_result == name_short, (
'File {0} - row {1}: Name {2} did match the '
'wrong regular expression: {3}'.format(
get_regex_test_data.data_name, row[0], name_test, name_result))
def test_EU_output():
cc = coco.CountryConverter()
EU28 = cc.EU28as("ISO2")
assert len(EU28 == 28)
assert cc.convert("Croatia", to="ISO2") in EU28.ISO2.values
EU27 = cc.EU27as("ISO2")
assert len(EU27 == 27)
assert cc.convert("Croatia", to="ISO2") not in EU27.ISO2.values
def test_EU_output():
cc = coco.CountryConverter()
EU28 = cc.EU28as('ISO2')
assert len(EU28 == 28)
assert cc.convert('Croatia', to='ISO2') in EU28.ISO2.values
EU27 = cc.EU27as('ISO2')
assert len(EU27 == 27)
assert cc.convert('Croatia', to='ISO2') not in EU27.ISO2.values
def test_MESSAGE_output():
cc = coco.CountryConverter()
ms = cc.MESSAGE
mi = cc.MESSAGEas(to="ISO3").set_index("original")
assert len(ms) == 12
assert "PAO" in ms.values
assert "SAS" in ms.values
assert mi.loc["AUT", "aggregated"] == "WEU"
def test_MESSAGE_output():
cc = coco.CountryConverter()
ms = cc.MESSAGE
mi = cc.MESSAGEas(to='ISO3').set_index('original')
assert len(ms) == 12
assert 'PAO' in ms.values
assert 'SAS' in ms.values
assert mi.loc['AUT', 'aggregated'] == 'WEU'
def test_BRIC_output():
cc = coco.CountryConverter()
bs = cc.BRIC
bi = cc.BRICas(to='ISO2')
bn = cc.BRICas(to=None)
assert len(bs) == 4 == len(bi) == len(bn)
assert 'Brazil' in bs.values
assert 'Brazil' in bn.values
assert 'CN' in bi.values
def test_APEC_output():
cc = coco.CountryConverter()
aa = cc.APEC
ai = cc.APECas(to='ISO2')
an = cc.APECas(to=None)
assert len(aa) == 21 == len(ai) == len(an)
assert 'Taiwan' in aa.values
assert 'Russia' in an.values
assert 'RU' in ai.values
def test_BASIC_output():
cc = coco.CountryConverter()
ba = cc.BASIC
bi = cc.BASICas(to='ISO2')
bn = cc.BASICas(to=None)
assert len(ba) == 4 == len(bi) == len(bn)
assert 'Brazil' in ba.values
assert 'Brazil' in bn.values
assert 'IN' in bi.values
def test_CIS_output():
cc = coco.CountryConverter()
ca = cc.CIS
ci = cc.CISas(to='ISO2')
cn = cc.CISas(to=None)
assert len(ca) == 8 == len(ci) == len(cn)
assert 'Belarus' in ca.values
assert 'Armenia' in cn.values
assert 'RU' in ci.values
def shorter_names(column):
"""
Takes in a country column names and changes all countries to their
shorter names.
Returns the column with the shorter names.
"""
cc = coco.CountryConverter()
column = column.map(lambda x: cc.convert(names=x, to='name_short'))
return column
def test_Cecilia_output():
cc = coco.CountryConverter()
cs = cc.Cecilia2050
ci = cc.Cecilia2050as(to='ISO3').set_index('original')
assert len(cs) == 4
assert 'RoW' in ci.values
assert 'EU' in ci.values
assert ci.loc['AUT', 'aggregated'] == 'EU'
assert ci.loc['AFG', 'aggregated'] == 'RoW'
def test_WIOD_output():
cc = coco.CountryConverter()
ws = cc.WIOD
wi = cc.WIODas(to="ISO2").set_index("original")
assert len(ws) == 41
assert "RoW" in ws.values
assert "NLD" in ws.values
assert wi.loc["AF", "aggregated"] == "RoW"
assert wi.loc["AT", "aggregated"] == "AUT"
def test_Cecilia_output():
cc = coco.CountryConverter()
cs = cc.Cecilia2050
ci = cc.Cecilia2050as(to="ISO3").set_index("original")
assert len(cs) == 4
assert "RoW" in ci.values
assert "EU" in ci.values
assert ci.loc["AUT", "aggregated"] == "EU"
assert ci.loc["AFG", "aggregated"] == "RoW"
def test_CIS_output():
cc = coco.CountryConverter()
ca = cc.CIS
ci = cc.CISas(to="ISO2")
cn = cc.CISas(to=None)
assert len(ca) == 8 == len(ci) == len(cn)
assert "Belarus" in ca.values
assert "Armenia" in cn.values
assert "RU" in ci.values
def test_BASIC_output():
cc = coco.CountryConverter()
ba = cc.BASIC
bi = cc.BASICas(to="ISO2")
bn = cc.BASICas(to=None)
assert len(ba) == 4 == len(bi) == len(bn)
assert "Brazil" in ba.values
assert "Brazil" in bn.values
assert "IN" in bi.values
def test_APEC_output():
cc = coco.CountryConverter()
aa = cc.APEC
ai = cc.APECas(to="ISO2")
an = cc.APECas(to=None)
assert len(aa) == 21 == len(ai) == len(an)
assert "Taiwan" in aa.values
assert "Russia" in an.values
assert "RU" in ai.values
def test_Eora_output():
cc = coco.CountryConverter()
es = cc.Eora
ei = cc.Eoraas(to="ISO2").set_index("original")
assert len(es) == 238
assert "AUT" in es.values
assert "AFG" in es.values
assert ei.loc["AF", "aggregated"] == "AFG"
assert ei.loc["AT", "aggregated"] == "AUT"
def test_BRIC_output():
cc = coco.CountryConverter()
bs = cc.BRIC
bi = cc.BRICas(to="ISO2")
bn = cc.BRICas(to=None)
assert len(bs) == 4 == len(bi) == len(bn)
assert "Brazil" in bs.values
assert "Brazil" in bn.values
assert "CN" in bi.values
def test_Eora_output():
cc = coco.CountryConverter()
es = cc.Eora
ei = cc.Eoraas(to='ISO2').set_index('original')
assert len(es) == 238
assert 'AUT' in es.values
assert 'AFG' in es.values
assert ei.loc['AF', 'aggregated'] == 'AFG'
assert ei.loc['AT', 'aggregated'] == 'AUT'