casefold/normalized_folding_test.mbt1///|2test "canonical caseless matching normalizes before and after folding" {3 inspect(@casefold.canonical_caseless_match("Å", "a\u{030A}"), content="true")4 inspect(@casefold.canonical_case_fold("Å"), content="a\u{030A}")5 inspect(6 @casefold.canonical_caseless_match("Straße", "STRASSE"),7 content="true",8 )9 // Initial NFD moves iota subscript after the accent before it becomes a letter.10 inspect(11 @casefold.canonical_caseless_match("α\u{0345}\u{0301}", "άι"),12 content="true",13 )14 // Final NFD reorders the accent introduced by folding U+0390.15 inspect(16 @casefold.canonical_caseless_match(17 "\u{0390}\u{0323}", "ι\u{0323}\u{0308}\u{0301}",18 ),19 content="true",20 )21 inspect(@casefold.canonical_caseless_match("A", "a"), content="false")22 inspect(@casefold.canonical_caseless_match("é", "e"), content="false")23 inspect(@casefold.canonical_caseless_match("a\u{00AD}", "a"), content="false")24 inspect(@casefold.canonical_case_fold(""), content="")25}2627///|28test "NFKC_Casefold maps compatibility characters and removes ignorables" {29 inspect(30 @casefold.nfkc_casefold("AKffi Straße ㎒"),31 content="akffi strasse mhz",32 )33 inspect(34 @casefold.nfkc_casefold("A\u{00AD}\u{200D}\u{FE0F}\u{E0100}"),35 content="a",36 )37 inspect(@casefold.nfkc_casefold("\u{034F}\u{E0000}\u{E0FFF}"), content="")38 inspect(@casefold.nfkc_casefold("A\u{00AD}\u{030A}"), content="å")39 inspect(@casefold.nfkc_casefold("\u{1100}\u{1161}"), content="가")40 inspect(41 @casefold.nfkc_casefold("Iİıi Ꭰꭰ"),42 content="ii\u{0307}ıi ᎠᎠ",43 )44 inspect(@casefold.nfkc_casefold(""), content="")45 inspect(46 @casefold.nfkc_casefold("中文😀\u{10FFFF}"),47 content="中文😀\u{10FFFF}",48 )49}5051///|52test "identifier matching includes initial canonical decomposition" {53 inspect(54 @casefold.identifier_caseless_match("A\u{00AD}", "a"),55 content="true",56 )57 inspect(58 @casefold.identifier_caseless_match("α\u{0345}\u{0301}", "άι"),59 content="true",60 )61 inspect(@casefold.identifier_caseless_match("é", "e"), content="false")62 inspect(@casefold.identifier_caseless_match("\u{00AD}", ""), content="true")63}64
///|
test "canonical caseless matching normalizes before and after folding" {
inspect(@casefold.canonical_caseless_match("Å", "a\u{030A}"), content="true")
inspect(@casefold.canonical_case_fold("Å"), content="a\u{030A}")
inspect(
@casefold.canonical_caseless_match("Straße", "STRASSE"),
content="true",
)
// Initial NFD moves iota subscript after the accent before it becomes a letter.
inspect(
@casefold.canonical_caseless_match("α\u{0345}\u{0301}", "άι"),
content="true",
)
// Final NFD reorders the accent introduced by folding U+0390.
inspect(
@casefold.canonical_caseless_match(
"\u{0390}\u{0323}", "ι\u{0323}\u{0308}\u{0301}",
),
content="true",
)
inspect(@casefold.canonical_caseless_match("A", "a"), content="false")
inspect(@casefold.canonical_caseless_match("é", "e"), content="false")
inspect(@casefold.canonical_caseless_match("a\u{00AD}", "a"), content="false")
inspect(@casefold.canonical_case_fold(""), content="")
}
///|
test "NFKC_Casefold maps compatibility characters and removes ignorables" {
inspect(
@casefold.nfkc_casefold("AKffi Straße ㎒"),
content="akffi strasse mhz",
)
inspect(
@casefold.nfkc_casefold("A\u{00AD}\u{200D}\u{FE0F}\u{E0100}"),
content="a",
)
inspect(@casefold.nfkc_casefold("\u{034F}\u{E0000}\u{E0FFF}"), content="")
inspect(@casefold.nfkc_casefold("A\u{00AD}\u{030A}"), content="å")
inspect(@casefold.nfkc_casefold("\u{1100}\u{1161}"), content="가")
inspect(
@casefold.nfkc_casefold("Iİıi Ꭰꭰ"),
content="ii\u{0307}ıi ᎠᎠ",
)
inspect(@casefold.nfkc_casefold(""), content="")
inspect(
@casefold.nfkc_casefold("中文😀\u{10FFFF}"),
content="中文😀\u{10FFFF}",
)
}
///|
test "identifier matching includes initial canonical decomposition" {
inspect(
@casefold.identifier_caseless_match("A\u{00AD}", "a"),
content="true",
)
inspect(
@casefold.identifier_caseless_match("α\u{0345}\u{0301}", "άι"),
content="true",
)
inspect(@casefold.identifier_caseless_match("é", "e"), content="false")
inspect(@casefold.identifier_caseless_match("\u{00AD}", ""), content="true")
}