ucd/case_mapping/case_mapping_test.mbt1///|2test "lookup_simple: unmapped characters are returned as they are" {3 inspect(@case_mapping.lookup_simple_uppercase('1'), content="1")4 inspect(@case_mapping.lookup_simple_lowercase('a'), content="a")5 inspect(@case_mapping.lookup_simple_titlecase('\u{4E2D}'), content="中")6}78///|9test "lookup_simple: supplementary planes" {10 // Deseret and Adlam have case pairs in plane 1.11 inspect(12 @case_mapping.lookup_simple_uppercase('\u{10428}').to_int(),13 content="66560",14 )15 inspect(16 @case_mapping.lookup_simple_lowercase('\u{1E921}').to_int(),17 content="125251",18 )19 // The last code point in the trie, then code points beyond it.20 inspect(21 @case_mapping.lookup_simple_uppercase('\u{1FFFF}').to_int(),22 content="131071",23 )24 inspect(25 @case_mapping.lookup_simple_uppercase('\u{20000}').to_int(),26 content="131072",27 )28 inspect(29 @case_mapping.lookup_simple_lowercase('\u{10FFFF}').to_int(),30 content="1114111",31 )32}3334///|35test "lookup_full: None without a mapping, the simple target otherwise" {36 debug_inspect(@case_mapping.lookup_full_uppercase('1'), content="None")37 debug_inspect(@case_mapping.lookup_full_uppercase('A'), content="None")38 debug_inspect(@case_mapping.lookup_full_uppercase('a'), content="Some(['A'])")39 debug_inspect(40 @case_mapping.lookup_full_lowercase('\u{00C9}'),41 content="Some(['é'])",42 )43 debug_inspect(44 @case_mapping.lookup_full_titlecase('\u{20000}'),45 content="None",46 )47}4849///|50test "lookup_full: special mappings replace the simple one" {51 debug_inspect(52 @case_mapping.lookup_full_uppercase('\u{00DF}'),53 content="Some(['S', 'S'])",54 )55 debug_inspect(56 @case_mapping.lookup_full_titlecase('\u{00DF}'),57 content="Some(['S', 's'])",58 )59 // U+00DF is its own lowercase, so nothing is listed for it.60 debug_inspect(@case_mapping.lookup_full_lowercase('\u{00DF}'), content="None")61 guard @case_mapping.lookup_full_lowercase('\u{0130}')62 is Some(['i', '\u{0307}']) else {63 fail("U+0130 should lowercase to i followed by U+0307")64 }65 // U+1F80 has a simple uppercase (U+1F88) and a different full one.66 inspect(67 @case_mapping.lookup_simple_uppercase('\u{1F80}').to_int(),68 content="8072",69 )70 debug_inspect(71 @case_mapping.lookup_full_uppercase('\u{1F80}'),72 content="Some(['Ἀ', 'Ι'])",73 )74}7576///|77test "full strings: the character itself, the simple target or the special mapping" {78 inspect(@case_mapping.full_uppercase('1'), content="1")79 inspect(@case_mapping.full_uppercase('a'), content="A")80 inspect(@case_mapping.full_lowercase('Z'), content="z")81 inspect(@case_mapping.full_titlecase('\u{00E9}'), content="É")82 inspect(@case_mapping.full_uppercase('\u{4E2D}'), content="中")83 inspect(@case_mapping.full_uppercase('\u{00DF}'), content="SS")84 inspect(@case_mapping.full_titlecase('\u{FB03}'), content="Ffi")85 inspect(@case_mapping.full_lowercase('\u{0130}'), content="i\u{0307}")86 inspect(87 @case_mapping.full_lowercase('\u{10400}') == "\u{10428}",88 content="true",89 )90 inspect(91 @case_mapping.full_uppercase('\u{10FFFF}') == "\u{10FFFF}",92 content="true",93 )94}9596///|97test "uppercase and lowercase: whole strings" {98 inspect(@case_mapping.uppercase(""), content="")99 inspect(100 @case_mapping.uppercase("Hello, World 123"),101 content="HELLO, WORLD 123",102 )103 inspect(104 @case_mapping.lowercase("Hello, World 123"),105 content="hello, world 123",106 )107 inspect(108 @case_mapping.uppercase("Ünïcödé Çàfé"),109 content="ÜNÏCÖDÉ ÇÀFÉ",110 )111 inspect(112 @case_mapping.lowercase("ΟΔΥΣΣΕΥΣ"),113 content="οδυσσευς",114 )115 inspect(116 @case_mapping.uppercase("世界こんにちは"),117 content="世界こんにちは",118 )119}120121///|122test "uppercase and lowercase: expansions" {123 inspect(@case_mapping.uppercase("Straße"), content="STRASSE")124 inspect(@case_mapping.uppercase("\u{FB03}x"), content="FFIX")125 inspect(@case_mapping.uppercase("ʼn"), content="ʼN")126 // U+1F80 has a simple uppercase, U+1F88, and a different full one.127 inspect(@case_mapping.uppercase("\u{1F80}"), content="ἈΙ")128 inspect(@case_mapping.lowercase("İSTANBUL"), content="i\u{0307}stanbul")129 // Each mapping is the same as for the character alone.130 let text = "aßb\u{FB03}c\u{0130}dDže\u{10428}f"131 let upper = StringBuilder()132 let lower = StringBuilder()133 for c in text {134 upper.write_string(@case_mapping.full_uppercase(c))135 lower.write_string(@case_mapping.full_lowercase(c))136 }137 @debug.assert_eq(@case_mapping.uppercase(text), upper.to_string())138 @debug.assert_eq(@case_mapping.lowercase(text), lower.to_string())139}140141///|142test "uppercase and lowercase: default context without language tailoring" {143 inspect(@case_mapping.lowercase("ΣΑΣ"), content="σας")144 inspect(@case_mapping.lowercase("ΣΑΣΑ"), content="σασα")145 // A single-character mapping has no surrounding context.146 inspect(@case_mapping.full_lowercase('Σ'), content="σ")147 // Language-specific mappings are not applied.148 inspect(@case_mapping.uppercase("i"), content="I")149 inspect(@case_mapping.lowercase("I"), content="i")150}151152///|153test "uppercase and lowercase: supplementary planes and surrogates" {154 @debug.assert_eq(@case_mapping.uppercase("a\u{10428}b"), "A\u{10400}B")155 @debug.assert_eq(156 @case_mapping.lowercase("\u{1E900}\u{1E901}"),157 "\u{1E922}\u{1E923}",158 )159 @debug.assert_eq(@case_mapping.uppercase("\u{20000}a"), "\u{20000}A")160 // Unpaired surrogates are copied as they are, mapped or not around them.161 let lone = String::from_array([162 'a',163 (0xD800).unsafe_to_char(),164 'b',165 (0xDC00).unsafe_to_char(),166 'ß',167 (0xD801).unsafe_to_char(),168 ])169 let mapped = @case_mapping.uppercase(lone)170 @debug.assert_eq(mapped.length(), 7)171 @debug.assert_eq(mapped.code_unit_at(1), (0xD800 : UInt16))172 @debug.assert_eq(mapped.code_unit_at(3), (0xDC00 : UInt16))173 @debug.assert_eq(mapped.code_unit_at(6), (0xD801 : UInt16))174 inspect(mapped[0:1], content="A")175 inspect(mapped[2:3], content="B")176 inspect(mapped[4:6], content="SS")177}178179///|180test "uppercase and lowercase: views" {181 let text = "xx Straße Ünï yy"182 inspect(@case_mapping.uppercase(text[3:13]), content="STRASSE ÜNÏ")183 inspect(@case_mapping.lowercase(text[3:9]), content="straße")184 // Nothing to change in a view of part of a string.185 inspect(@case_mapping.uppercase(text[3:4].to_upper()), content="S")186 inspect(@case_mapping.lowercase(text[0:3]), content="xx ")187}188189///|190test "uppercase and lowercase: every scalar maps as it does alone" {191 // Also checks that ASCII maps as `AsciiCase` says: only letters change.192 let all = StringBuilder()193 let all_upper = StringBuilder()194 let all_lower = StringBuilder()195 for cp in 0..<=0x10FFFF {196 if cp >= 0xD800 && cp <= 0xDFFF {197 continue198 }199 let c = cp.unsafe_to_char()200 let upper = @case_mapping.full_uppercase(c)201 let lower = @case_mapping.full_lowercase(c)202 let input = c.to_string()203 @debug.assert_eq(@case_mapping.uppercase(input), upper, msg="at \{cp}")204 @debug.assert_eq(@case_mapping.lowercase(input), lower, msg="at \{cp}")205 all.write_string(input)206 all_upper.write_string(upper)207 all_lower.write_string(lower)208 }209 let all = all.to_string()210 assert_true(@case_mapping.uppercase(all) == all_upper.to_string())211 assert_true(@case_mapping.lowercase(all) == all_lower.to_string())212}213
///|
test "lookup_simple: unmapped characters are returned as they are" {
inspect(@case_mapping.lookup_simple_uppercase('1'), content="1")
inspect(@case_mapping.lookup_simple_lowercase('a'), content="a")
inspect(@case_mapping.lookup_simple_titlecase('\u{4E2D}'), content="中")
}
///|
test "lookup_simple: supplementary planes" {
// Deseret and Adlam have case pairs in plane 1.
inspect(
@case_mapping.lookup_simple_uppercase('\u{10428}').to_int(),
content="66560",
)
inspect(
@case_mapping.lookup_simple_lowercase('\u{1E921}').to_int(),
content="125251",
)
// The last code point in the trie, then code points beyond it.
inspect(
@case_mapping.lookup_simple_uppercase('\u{1FFFF}').to_int(),
content="131071",
)
inspect(
@case_mapping.lookup_simple_uppercase('\u{20000}').to_int(),
content="131072",
)
inspect(
@case_mapping.lookup_simple_lowercase('\u{10FFFF}').to_int(),
content="1114111",
)
}
///|
test "lookup_full: None without a mapping, the simple target otherwise" {
debug_inspect(@case_mapping.lookup_full_uppercase('1'), content="None")
debug_inspect(@case_mapping.lookup_full_uppercase('A'), content="None")
debug_inspect(@case_mapping.lookup_full_uppercase('a'), content="Some(['A'])")
debug_inspect(
@case_mapping.lookup_full_lowercase('\u{00C9}'),
content="Some(['é'])",
)
debug_inspect(
@case_mapping.lookup_full_titlecase('\u{20000}'),
content="None",
)
}
///|
test "lookup_full: special mappings replace the simple one" {
debug_inspect(
@case_mapping.lookup_full_uppercase('\u{00DF}'),
content="Some(['S', 'S'])",
)
debug_inspect(
@case_mapping.lookup_full_titlecase('\u{00DF}'),
content="Some(['S', 's'])",
)
// U+00DF is its own lowercase, so nothing is listed for it.
debug_inspect(@case_mapping.lookup_full_lowercase('\u{00DF}'), content="None")
guard @case_mapping.lookup_full_lowercase('\u{0130}')
is Some(['i', '\u{0307}']) else {
fail("U+0130 should lowercase to i followed by U+0307")
}
// U+1F80 has a simple uppercase (U+1F88) and a different full one.
inspect(
@case_mapping.lookup_simple_uppercase('\u{1F80}').to_int(),
content="8072",
)
debug_inspect(
@case_mapping.lookup_full_uppercase('\u{1F80}'),
content="Some(['Ἀ', 'Ι'])",
)
}
///|
test "full strings: the character itself, the simple target or the special mapping" {
inspect(@case_mapping.full_uppercase('1'), content="1")
inspect(@case_mapping.full_uppercase('a'), content="A")
inspect(@case_mapping.full_lowercase('Z'), content="z")
inspect(@case_mapping.full_titlecase('\u{00E9}'), content="É")
inspect(@case_mapping.full_uppercase('\u{4E2D}'), content="中")
inspect(@case_mapping.full_uppercase('\u{00DF}'), content="SS")
inspect(@case_mapping.full_titlecase('\u{FB03}'), content="Ffi")
inspect(@case_mapping.full_lowercase('\u{0130}'), content="i\u{0307}")
inspect(
@case_mapping.full_lowercase('\u{10400}') == "\u{10428}",
content="true",
)
inspect(
@case_mapping.full_uppercase('\u{10FFFF}') == "\u{10FFFF}",
content="true",
)
}
///|
test "uppercase and lowercase: whole strings" {
inspect(@case_mapping.uppercase(""), content="")
inspect(
@case_mapping.uppercase("Hello, World 123"),
content="HELLO, WORLD 123",
)
inspect(
@case_mapping.lowercase("Hello, World 123"),
content="hello, world 123",
)
inspect(
@case_mapping.uppercase("Ünïcödé Çàfé"),
content="ÜNÏCÖDÉ ÇÀFÉ",
)
inspect(
@case_mapping.lowercase("ΟΔΥΣΣΕΥΣ"),
content="οδυσσευς",
)
inspect(
@case_mapping.uppercase("世界こんにちは"),
content="世界こんにちは",
)
}
///|
test "uppercase and lowercase: expansions" {
inspect(@case_mapping.uppercase("Straße"), content="STRASSE")
inspect(@case_mapping.uppercase("\u{FB03}x"), content="FFIX")
inspect(@case_mapping.uppercase("ʼn"), content="ʼN")
// U+1F80 has a simple uppercase, U+1F88, and a different full one.
inspect(@case_mapping.uppercase("\u{1F80}"), content="ἈΙ")
inspect(@case_mapping.lowercase("İSTANBUL"), content="i\u{0307}stanbul")
// Each mapping is the same as for the character alone.
let text = "aßb\u{FB03}c\u{0130}dDže\u{10428}f"
let upper = StringBuilder()
let lower = StringBuilder()
for c in text {
upper.write_string(@case_mapping.full_uppercase(c))
lower.write_string(@case_mapping.full_lowercase(c))
}
@debug.assert_eq(@case_mapping.uppercase(text), upper.to_string())
@debug.assert_eq(@case_mapping.lowercase(text), lower.to_string())
}
///|
test "uppercase and lowercase: default context without language tailoring" {
inspect(@case_mapping.lowercase("ΣΑΣ"), content="σας")
inspect(@case_mapping.lowercase("ΣΑΣΑ"), content="σασα")
// A single-character mapping has no surrounding context.
inspect(@case_mapping.full_lowercase('Σ'), content="σ")
// Language-specific mappings are not applied.
inspect(@case_mapping.uppercase("i"), content="I")
inspect(@case_mapping.lowercase("I"), content="i")
}
///|
test "uppercase and lowercase: supplementary planes and surrogates" {
@debug.assert_eq(@case_mapping.uppercase("a\u{10428}b"), "A\u{10400}B")
@debug.assert_eq(
@case_mapping.lowercase("\u{1E900}\u{1E901}"),
"\u{1E922}\u{1E923}",
)
@debug.assert_eq(@case_mapping.uppercase("\u{20000}a"), "\u{20000}A")
// Unpaired surrogates are copied as they are, mapped or not around them.
let lone = String::from_array([
'a',
(0xD800).unsafe_to_char(),
'b',
(0xDC00).unsafe_to_char(),
'ß',
(0xD801).unsafe_to_char(),
])
let mapped = @case_mapping.uppercase(lone)
@debug.assert_eq(mapped.length(), 7)
@debug.assert_eq(mapped.code_unit_at(1), (0xD800 : UInt16))
@debug.assert_eq(mapped.code_unit_at(3), (0xDC00 : UInt16))
@debug.assert_eq(mapped.code_unit_at(6), (0xD801 : UInt16))
inspect(mapped[0:1], content="A")
inspect(mapped[2:3], content="B")
inspect(mapped[4:6], content="SS")
}
///|
test "uppercase and lowercase: views" {
let text = "xx Straße Ünï yy"
inspect(@case_mapping.uppercase(text[3:13]), content="STRASSE ÜNÏ")
inspect(@case_mapping.lowercase(text[3:9]), content="straße")
// Nothing to change in a view of part of a string.
inspect(@case_mapping.uppercase(text[3:4].to_upper()), content="S")
inspect(@case_mapping.lowercase(text[0:3]), content="xx ")
}
///|
test "uppercase and lowercase: every scalar maps as it does alone" {
// Also checks that ASCII maps as `AsciiCase` says: only letters change.
let all = StringBuilder()
let all_upper = StringBuilder()
let all_lower = StringBuilder()
for cp in 0..<=0x10FFFF {
if cp >= 0xD800 && cp <= 0xDFFF {
continue
}
let c = cp.unsafe_to_char()
let upper = @case_mapping.full_uppercase(c)
let lower = @case_mapping.full_lowercase(c)
let input = c.to_string()
@debug.assert_eq(@case_mapping.uppercase(input), upper, msg="at \{cp}")
@debug.assert_eq(@case_mapping.lowercase(input), lower, msg="at \{cp}")
all.write_string(input)
all_upper.write_string(upper)
all_lower.write_string(lower)
}
let all = all.to_string()
assert_true(@case_mapping.uppercase(all) == all_upper.to_string())
assert_true(@case_mapping.lowercase(all) == all_lower.to_string())
}