StringDistances.jl/test/modifiers.jl

using StringDistances, Test

@testset "Modifiers" begin

	# Hamming
	@test compare("", "abc", Hamming()) ≈ 0.0 atol = 1e-4
	@test compare("acc", "abc", Hamming()) ≈ 2/3 atol = 1e-4
	@test compare("saturday", "sunday", Hamming()) ≈ 1/8 atol = 1e-4
	@test compare("New York Yankees",  "Yankees", Partial(Hamming())) ≈ 1
	@test compare("New York Yankees",  "", Partial(Hamming())) ≈ 1
	compare("aüa", "aua", Hamming())

	# Qgram
	@test compare("", "abc", QGram(1)) ≈ 0.0 atol = 1e-4
	@test compare("abc", "cba", QGram(1)) ≈ 1.0 atol = 1e-4
	@test compare("abc", "ccc", QGram(1)) ≈ 1/3 atol = 1e-4
	compare("aüa", "aua", TokenMax(QGram(2)))
	@test compare("", "abc", Jaccard(2)) ≈ 0.0 atol = 1e-4
	@test compare("martha", "martha", Jaccard(2)) ≈ 1.0 atol = 1e-4
	@test compare("martha", "martha", Jaccard(2)) ≈ 1.0 atol = 1e-4
	@test compare("aa", "aa ", Partial(Jaccard(2))) ≈ 1.0
	@test compare("martha", "martha", Cosine(2)) ≈ 1.0 atol = 1e-4
	@test compare("martha", "martha", Overlap(2)) ≈ 1.0 atol = 1e-4
	@test compare("martha", "martha", SorensenDice(2)) ≈ 1.0 atol = 1e-4

	# Jaro
	compare("aüa", "aua", Jaro())

	#Levenshtein
	compare("aüa", "aua", Levenshtein())
	compare("aüa", "aua", DamerauLevenshtein())

	# Winkler
	@test compare("martha", "marhta", Winkler(Jaro(), 0.1, 0.0, 4)) ≈ 0.9611 atol = 1e-4
	@test compare("dwayne", "duane", Winkler(Jaro(), 0.1, 0.0, 4)) ≈ 0.84 atol = 1e-4
	@test compare("dixon", "dicksonx", Winkler(Jaro(), 0.1, 0.0, 4)) ≈ 0.81333 atol = 1e-4
	@test compare("william", "williams", Winkler(Jaro(), 0.1, 0.0, 4)) ≈ 0.975 atol = 1e-4
	@test compare("", "foo", Winkler(Jaro(), 0.1, 0.0, 4)) ≈ 0.0 atol = 1e-4
	@test compare("a", "a", Winkler(Jaro(), 0.1, 0.0, 4)) ≈ 1.0 atol = 1e-4
	@test compare("abc", "xyz", Winkler(Jaro(), 0.1, 0.0, 4)) ≈ 0.0 atol = 1e-4

	# RatcliffObershelp
	@test compare("New York Mets vs Atlanta Braves", "", RatcliffObershelp())  ≈ 0.0
	@test round(Int, 100 * compare("为人子女者要堂堂正正做人，千万不可作奸犯科，致使父母蒙羞", "此前稍早些时候中国商务部发布消息称，中美经贸高级别磋商双方牵头人通话，中方就美拟9月1日加征关税进行了严正交涉。", RatcliffObershelp())) == 5
	compare("aüa", "aua", TokenMax(RatcliffObershelp()))

	@test compare("New York Yankees",  "Yankees", Partial(RatcliffObershelp())) ≈ 1.0
	@test compare("New York Yankees",  "", Partial(RatcliffObershelp())) ≈ 0.0
	@test compare("mariners vs angels", "los angeles angels at seattle mariners", Partial(RatcliffObershelp())) ≈ 0.444444444444
	@test compare("HSINCHUANG", "SINJHUAN", Partial(RatcliffObershelp())) ≈ 0.875
	@test compare("HSINCHUANG", "LSINJHUANG DISTRIC", Partial(RatcliffObershelp())) ≈ 0.8
	@test compare("HSINCHUANG", "SINJHUANG DISTRICT", Partial(RatcliffObershelp())) ≈ 0.8
	@test compare("HSINCHUANG",  "SINJHUANG", Partial(RatcliffObershelp())) ≈ 0.8888888888888
	@test compare("New York Mets vs Atlanta Braves", "Atlanta Braves vs New York Mets", TokenSort(RatcliffObershelp()))  ≈ 1.0
	@test compare("mariners vs angels", "los angeles angels of anaheim at seattle mariners", TokenSet(RatcliffObershelp())) ≈ 1.0 - 0.09090909090909094
	@test compare("New York Mets vs Atlanta Braves", "", TokenSort(RatcliffObershelp()))  ≈ 0.0
	@test compare("mariners vs angels", "", TokenSet(RatcliffObershelp())) ≈ 0.0
	@test compare("mariners vs angels", "los angeles angels at seattle mariners", TokenSet(Partial(RatcliffObershelp()))) ≈ 1.0
	@test compare("mariners", "mariner", TokenMax(RatcliffObershelp())) ≈ 0.933333333333333
	@test round(Int, 100 * compare("为人子女者要堂堂正正做人，千万不可作奸犯科，致使父母蒙羞", "此前稍早些时候中国商务部发布消息称，中美经贸高级别磋商双方牵头人通话，中方就美拟9月1日加征关税进行了严正交涉。", RatcliffObershelp())) == 5
	@test round(Int, 100 * compare("为人子女者要堂堂正正做人，千万不可作奸犯科，致使父母蒙羞", "此前稍早些时候中国商务部发布消息称，中美经贸高级别磋商双方牵头人通话，中方就美拟9月1日加征关税进行了严正交涉。", Partial(RatcliffObershelp()))) == 7
	@test round(Int, 100 * compare("mariners", "mariner are playing tomorrow", TokenMax(RatcliffObershelp()))) == 79
	@test round(Int, 100 * compare("mariners", "mariner are playing tomorrow mariner are playing tomorrow mariner are playing tomorrow mariner are playing tomorrow", Partial(RatcliffObershelp()))) == 88
	@test round(Int, 100 * compare("mariners", "mariner are playing tomorrow mariner are playing tomorrow mariner are playing tomorrow mariner are playing tomorrow", TokenSort(RatcliffObershelp()))) == 11
	@test round(Int, 100 * compare("mariners", "are mariner playing tomorrow", RatcliffObershelp())) == 39
	@test round(Int, 100 * compare("mariners", "are mariner playing tomorrow", Partial(RatcliffObershelp()))) == 88
	@test round(Int, 100 * compare("mariners", "mariner are playing tomorrow mariner are playing tomorrow mariner are playing tomorrow mariner are playing tomorrow", TokenSet(RatcliffObershelp()))) == 39
	@test round(Int, 100 * compare("mariners", "mariner are playing tomorrow mariner are playing tomorrow mariner are playing tomorrow mariner are playing tomorrow", TokenSet(Partial(RatcliffObershelp())))) == 88
	# not exactly the same because tokenmax has uses the max of rounded tokenset etc
	@test round(Int, 100 * compare("mariners", "mariner are playing tomorrow mariner are playing tomorrow mariner are playing tomorrow mariner are playing tomorrow", TokenMax(RatcliffObershelp()))) == 52


	# check missing
	@test compare("ok", missing, Levenshtein()) === missing

	# check min
	strings = [
	("martha", "marhta"),
	("dwayne", "duane") ,
	("dixon", "dicksonx"),
	("william", "williams"),
	("", "foo"),
	("a", "a"),
	("abc", "xyz"),
	("abc", "ccc"),
	("kitten", "sitting"),
	("saturday", "sunday"),
	("hi, my name is", "my name is"),
	("alborgów", "amoniak"),
	("cape sand recycling ", "edith ann graham"),
	( "jellyifhs", "jellyfish"),
	("ifhs", "fish"),
	("leia", "leela"),
	]
	for dist in (Levenshtein, DamerauLevenshtein)
		for i in eachindex(strings)
			if compare(strings[i]..., dist()) <  1 / 3
				@test compare(strings[i]..., dist() ; min_score = 1/ 3) ≈ 0.0
			else
				@test compare(strings[i]..., dist() ; min_score = 1/ 3) ≈ compare(strings[i]..., dist())
			end
		end
	end

	# check find_best and find_all
	@test find_best("New York", ["NewYork", "Newark", "San Francisco"], Levenshtein()) == "NewYork"
	@test find_best("New York", ["NewYork", "Newark", "San Francisco"], Jaro()) == "NewYork"
	@test find_all("New York", ["NewYork", "Newark", "San Francisco"], Levenshtein()) == ["NewYork"]
	@test find_all("New York", ["NewYork", "Newark", "San Francisco"], Jaro()) == ["NewYork", "Newark"]

end


#= Python code
from fuzzywuzzy import fuzz
fuzz.ratio("为人子女者要堂堂正正做人，千万不可作奸犯科，致使父母蒙羞", "此前稍早些时候中国商务部发布消息称，中美经贸高级别磋商双方牵头人通话，中方就美拟9月1日加征关税进行了严正交涉。")
fuzz.partial_ratio("为人子女者要堂堂正正做人，千万不可作奸犯科，致使父母蒙羞", "此前稍早些时候中国商务部发布消息称，中美经贸高级别磋商双方牵头人通话，中方就美拟9月1日加征关税进行了严正交涉。")
=#