grindlemire/go-tui code browser

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208

package tui

import (
	"strings"
	"testing"
	"unicode/utf8"
)

// Test data uses escapes so the intent is unambiguous in source:
//
//	\U0001F680                          rocket
//	\U0001F1FA\U0001F1F8                US flag (regional-indicator pair)
//	\U0001F44D\U0001F3FD                thumbs up + medium skin tone
//	\U0001F468‍\U0001F469...\U0001F466       man ZWJ woman ZWJ girl ZWJ boy (family)
//	❤️                        heavy heart + VS16 (emoji presentation)
//	1️⃣                       keycap digit one
//	é                             "e" + combining acute ("é", decomposed)
//	a​b                            "a" + zero-width space + "b"

// clusterRuneStarts returns the rune index where each grapheme cluster begins,
// followed by the total rune count. For "ab" it returns [0,1,2]; for the family
// emoji (7 runes, one cluster) it returns [0,7].
func clusterRuneStarts(s string) []int {
	starts := []int{0}
	runeIdx := 0
	for len(s) > 0 {
		_, _, size := nextCluster(s)
		if size == 0 {
			break
		}
		// Count runes in this cluster.
		for i := 0; i < size; {
			_, rs := utf8.DecodeRuneInString(s[i:])
			if rs == 0 {
				break
			}
			i += rs
			runeIdx++
		}
		starts = append(starts, runeIdx)
		s = s[size:]
	}
	return starts
}

const (
	emojiRocket  = "\U0001F680"
	emojiFlagUS  = "\U0001F1FA\U0001F1F8"
	emojiThumbs  = "\U0001F44D\U0001F3FD"
	emojiFamily  = "\U0001F468\u200d\U0001F469\u200d\U0001F467\u200d\U0001F466"
	emojiHeart   = "\u2764\U0000FE0F"
	emojiKeycap1 = "1\U0000FE0F\u20E3"
	accentE      = "e\u0301"
)

func TestNextCluster_SegmentationAndWidth(t *testing.T) {
	type tc struct {
		in       string
		clusters []string
		width    int
	}

	tests := map[string]tc{
		"ascii":        {in: "ab", clusters: []string{"a", "b"}, width: 2},
		"cjk":          {in: "你好", clusters: []string{"你", "好"}, width: 4},
		"single emoji": {in: emojiRocket, clusters: []string{emojiRocket}, width: 2},
		// Regional-indicator pair, skin-tone, ZWJ family, heart+VS16, and keycap are
		// each one cluster two columns wide, not the sum of their code points.
		"flag":       {in: emojiFlagUS, clusters: []string{emojiFlagUS}, width: 2},
		"skin tone":  {in: emojiThumbs, clusters: []string{emojiThumbs}, width: 2},
		"zwj family": {in: emojiFamily, clusters: []string{emojiFamily}, width: 2},
		"heart vs16": {in: emojiHeart, clusters: []string{emojiHeart}, width: 2},
		"keycap":     {in: emojiKeycap1, clusters: []string{emojiKeycap1}, width: 2},
		// ASCII fast path must not split a base from a following combining mark:
		// the byte after 'e' is 0xCC (a leading byte), not ASCII, so it decodes.
		"decomposed accent":  {in: accentE, clusters: []string{accentE}, width: 1},
		"decomposed in word": {in: "caf" + accentE, clusters: []string{"c", "a", "f", accentE}, width: 4},
		// A format control with GCB=Control (ZWSP) must NOT glue to its neighbours.
		"zwsp does not glue": {in: "a\u200bb", clusters: []string{"a", "\u200b", "b"}, width: 3},
		// Regional indicators pair at most two at a time; a third stands alone.
		"three regional indicators": {
			in:       emojiFlagUS + "\U0001F1EB",
			clusters: []string{emojiFlagUS, "\U0001F1EB"},
			width:    4,
		},
	}

	for name, tt := range tests {
		t.Run(name, func(t *testing.T) {
			var got []string
			total := 0
			for rest := tt.in; len(rest) > 0; {
				cl, w, size := nextCluster(rest)
				if size == 0 {
					t.Fatalf("nextCluster(%q) returned size 0", rest)
				}
				got = append(got, cl)
				total += w
				rest = rest[size:]
			}

			if len(got) != len(tt.clusters) {
				t.Fatalf("clusters = %q, want %q", got, tt.clusters)
			}
			for i := range got {
				if got[i] != tt.clusters[i] {
					t.Errorf("cluster[%d] = %q, want %q", i, got[i], tt.clusters[i])
				}
			}
			if strings.Join(got, "") != tt.in {
				t.Errorf("clusters %q do not reconstruct input %q", got, tt.in)
			}
			if total != tt.width {
				t.Errorf("summed cluster width = %d, want %d", total, tt.width)
			}
			if sw := stringWidth(tt.in); sw != tt.width {
				t.Errorf("stringWidth(%q) = %d, want %d", tt.in, sw, tt.width)
			}
		})
	}
}

// TestStringWidth_Clusters documents the exact divergence from issue #95: a
// per-code-point sum over-counts these, the grapheme-aware width does not.
func TestStringWidth_Clusters(t *testing.T) {
	type tc struct {
		in    string
		width int
	}

	tests := map[string]tc{
		"flag":         {in: emojiFlagUS, width: 2},
		"skin tone":    {in: emojiThumbs, width: 2},
		"zwj family":   {in: emojiFamily, width: 2},
		"decomposed":   {in: "caf" + accentE, width: 4},
		"flag in text": {in: "x" + emojiFlagUS + "y", width: 4},
	}

	for name, tt := range tests {
		t.Run(name, func(t *testing.T) {
			if got := stringWidth(tt.in); got != tt.width {
				t.Errorf("stringWidth(%q) = %d, want %d", tt.in, got, tt.width)
			}
		})
	}
}

// TestNextClusterBytes_MatchesNextCluster pins that the []byte scanner variant
// segments identically to the string variant (same shared state machine).
func TestNextClusterBytes_MatchesNextCluster(t *testing.T) {
	inputs := map[string]string{
		"ascii":      "ab",
		"cjk":        "你好",
		"rocket":     emojiRocket,
		"flag":       emojiFlagUS,
		"skin tone":  emojiThumbs,
		"zwj family": emojiFamily,
		"heart vs16": emojiHeart,
		"keycap":     emojiKeycap1,
		"accent":     accentE,
		"zwsp":       "a\u200bb",
	}

	for name, in := range inputs {
		t.Run(name, func(t *testing.T) {
			for rest := in; len(rest) > 0; {
				cl, w, size := nextCluster(rest)
				bw, bsize, base := nextClusterBytes([]byte(rest))
				if bw != w || bsize != size {
					t.Fatalf("nextClusterBytes = (w=%d size=%d), nextCluster = (w=%d size=%d)", bw, bsize, w, size)
				}
				wantBase, _ := utf8.DecodeRuneInString(cl)
				if base != wantBase {
					t.Errorf("base rune = %q, want %q", base, wantBase)
				}
				rest = rest[size:]
			}
		})
	}
}

func TestClusterRuneStarts(t *testing.T) {
	type tc struct {
		in   string
		want []int
	}

	tests := map[string]tc{
		"ascii":                 {in: "ab", want: []int{0, 1, 2}},
		"family is one cluster": {in: emojiFamily, want: []int{0, 7}},
		// c, a, f are one rune each; the final "é" is two runes (e + combining).
		"decomposed in word": {in: "caf" + accentE, want: []int{0, 1, 2, 3, 5}},
	}

	for name, tt := range tests {
		t.Run(name, func(t *testing.T) {
			got := clusterRuneStarts(tt.in)
			if len(got) != len(tt.want) {
				t.Fatalf("clusterRuneStarts(%q) = %v, want %v", tt.in, got, tt.want)
			}
			for i := range got {
				if got[i] != tt.want[i] {
					t.Fatalf("clusterRuneStarts(%q) = %v, want %v", tt.in, got, tt.want)
				}
			}
		})
	}
}