github.com/go-xe2/third@v1.0.3/golang.org/x/text/unicode/norm/iter_test.go (about) 1 // Copyright 2011 The Go Authors. All rights reserved. 2 // Use of this source code is governed by a BSD-style 3 // license that can be found in the LICENSE file. 4 5 package norm 6 7 import ( 8 "strings" 9 "testing" 10 ) 11 12 func doIterNorm(f Form, s string) []byte { 13 acc := []byte{} 14 i := Iter{} 15 i.InitString(f, s) 16 for !i.Done() { 17 acc = append(acc, i.Next()...) 18 } 19 return acc 20 } 21 22 func TestIterNext(t *testing.T) { 23 runNormTests(t, "IterNext", func(f Form, out []byte, s string) []byte { 24 return doIterNorm(f, string(append(out, s...))) 25 }) 26 } 27 28 type SegmentTest struct { 29 in string 30 out []string 31 } 32 33 var segmentTests = []SegmentTest{ 34 {"\u1E0A\u0323a", []string{"\x44\u0323\u0307", "a", ""}}, 35 {rep('a', segSize), append(strings.Split(rep('a', segSize), ""), "")}, 36 {rep('a', segSize+2), append(strings.Split(rep('a', segSize+2), ""), "")}, 37 {rep('a', segSize) + "\u0300aa", 38 append(strings.Split(rep('a', segSize-1), ""), "a\u0300", "a", "a", "")}, 39 40 // U+0f73 is NOT treated as a starter as it is a modifier 41 {"a" + grave(29) + "\u0f73", []string{"a" + grave(29), cgj + "\u0f73"}}, 42 {"a\u0f73", []string{"a\u0f73"}}, 43 44 // U+ff9e is treated as a non-starter. 45 // TODO: should we? Note that this will only affect iteration, as whether 46 // or not we do so does not affect the normalization output and will either 47 // way result in consistent iteration output. 48 {"a" + grave(30) + "\uff9e", []string{"a" + grave(30), cgj + "\uff9e"}}, 49 {"a\uff9e", []string{"a\uff9e"}}, 50 } 51 52 var segmentTestsK = []SegmentTest{ 53 {"\u3332", []string{"\u30D5", "\u30A1", "\u30E9", "\u30C3", "\u30C8\u3099", ""}}, 54 // last segment of multi-segment decomposition needs normalization 55 {"\u3332\u093C", []string{"\u30D5", "\u30A1", "\u30E9", "\u30C3", "\u30C8\u093C\u3099", ""}}, 56 {"\u320E", []string{"\x28", "\uAC00", "\x29"}}, 57 58 // last segment should be copied to start of buffer. 59 {"\ufdfa", []string{"\u0635", "\u0644", "\u0649", " ", "\u0627", "\u0644", "\u0644", "\u0647", " ", "\u0639", "\u0644", "\u064a", "\u0647", " ", "\u0648", "\u0633", "\u0644", "\u0645", ""}}, 60 {"\ufdfa" + grave(30), []string{"\u0635", "\u0644", "\u0649", " ", "\u0627", "\u0644", "\u0644", "\u0647", " ", "\u0639", "\u0644", "\u064a", "\u0647", " ", "\u0648", "\u0633", "\u0644", "\u0645" + grave(30), ""}}, 61 {"\uFDFA" + grave(64), []string{"\u0635", "\u0644", "\u0649", " ", "\u0627", "\u0644", "\u0644", "\u0647", " ", "\u0639", "\u0644", "\u064a", "\u0647", " ", "\u0648", "\u0633", "\u0644", "\u0645" + grave(30), cgj + grave(30), cgj + grave(4), ""}}, 62 63 // Hangul and Jamo are grouped together. 64 {"\uAC00", []string{"\u1100\u1161", ""}}, 65 {"\uAC01", []string{"\u1100\u1161\u11A8", ""}}, 66 {"\u1100\u1161", []string{"\u1100\u1161", ""}}, 67 } 68 69 // Note that, by design, segmentation is equal for composing and decomposing forms. 70 func TestIterSegmentation(t *testing.T) { 71 segmentTest(t, "SegmentTestD", NFD, segmentTests) 72 segmentTest(t, "SegmentTestC", NFC, segmentTests) 73 segmentTest(t, "SegmentTestKD", NFKD, segmentTestsK) 74 segmentTest(t, "SegmentTestKC", NFKC, segmentTestsK) 75 } 76 77 func segmentTest(t *testing.T, name string, f Form, tests []SegmentTest) { 78 iter := Iter{} 79 for i, tt := range tests { 80 iter.InitString(f, tt.in) 81 for j, seg := range tt.out { 82 if seg == "" { 83 if !iter.Done() { 84 res := string(iter.Next()) 85 t.Errorf(`%s:%d:%d: expected Done()==true, found segment %+q`, name, i, j, res) 86 } 87 continue 88 } 89 if iter.Done() { 90 t.Errorf("%s:%d:%d: Done()==true, want false", name, i, j) 91 } 92 seg = f.String(seg) 93 if res := string(iter.Next()); res != seg { 94 t.Errorf(`%s:%d:%d" segment was %+q (%d); want %+q (%d)`, name, i, j, pc(res), len(res), pc(seg), len(seg)) 95 } 96 } 97 } 98 }