-
Notifications
You must be signed in to change notification settings - Fork 374
Expand file tree
/
Copy pathencoding.ts
More file actions
212 lines (185 loc) · 6.88 KB
/
Copy pathencoding.ts
File metadata and controls
212 lines (185 loc) · 6.88 KB
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
209
210
211
212
import chardet from "chardet";
/**
* 从 Content-Type header 中解析 charset
*/
export const parseCharsetFromContentType = (ct: string | null): string => {
const m = ct ? /charset=([^;]+)/i.exec(ct) : null;
return m ? m[1].trim().toLowerCase().replace(/['"]/g, "") : "";
};
export const decodeUTF32 = (utf32Bytes: Uint8Array, isLE: boolean = true): string => {
if (!(utf32Bytes instanceof Uint8Array)) {
throw new TypeError("utf32Bytes must be a Uint8Array");
}
const byteLen = utf32Bytes.byteLength;
if (byteLen % 4 !== 0) {
throw new RangeError("UTF-32 byte length must be a multiple of 4");
}
const numCodePoints = byteLen >>> 2;
let u32;
if (isLE) {
u32 = new Uint32Array(utf32Bytes.buffer, utf32Bytes.byteOffset, numCodePoints);
} else {
const view = new DataView(utf32Bytes.buffer, utf32Bytes.byteOffset, byteLen);
u32 = new Uint32Array(numCodePoints);
for (let i = 0, j = 0; i < byteLen; i += 4) {
u32[j++] = view.getUint32(i, false);
}
}
if (u32[0] === 0x0000feff) u32 = u32.subarray(1);
let out = "";
for (let i = 0; i < u32.length; i += 16384) {
out += String.fromCodePoint(...u32.subarray(i, i + 16384));
}
return out;
};
export const bytesDecode = (charset: string, bytes: Uint8Array): string => {
if (charset === "utf-32le") {
return decodeUTF32(bytes, true);
} else if (charset === "utf-32be") {
return decodeUTF32(bytes, false);
} else {
return new TextDecoder(charset).decode(bytes);
}
};
/**
* 检测字节数组的编码
* 优先使用 Content-Type header，失败时使用 chardet（仅对前16KB检测以提升性能）
*/
export const detectEncoding = (data: Uint8Array, contentType: string | null): string => {
// 优先尝试使用 Content-Type header 中的 charset
const headerCharset = parseCharsetFromContentType(contentType);
if (headerCharset) {
try {
// 验证 charset 是否有效
new TextDecoder(headerCharset);
return headerCharset;
} catch (e: any) {
console.warn(`Invalid charset from Content-Type header: ${headerCharset}, error: ${e.message}`);
}
}
// 使用 chardet 检测编码，仅检测前16KB以提升性能
const sampleSize = Math.min(data.length, 16 * 1024); // max 16KB
const sample = data.subarray(0, sampleSize);
const analysedResult = chardet.analyse(sample);
let highestConfidence = 0;
const results = [];
let leastCharLen = Infinity;
for (const entry of analysedResult) {
const encoding = entry.name.toLowerCase();
let decodedText;
try {
// 验证检测到的编码是否有效
decodedText = bytesDecode(encoding, sample);
} catch (_e: any) {
// ignored
}
if (!decodedText) continue;
if (!highestConfidence) {
highestConfidence = entry.confidence;
if (highestConfidence > 90) return encoding;
} else if (highestConfidence > 70 && entry.confidence < 30) {
// 不考虑 confidence 过低的编码
break;
} else if (highestConfidence > 50 && entry.confidence < 20) {
// 不考虑 confidence 过低的编码
break;
}
// 当字元符少，不足以自动判断时，改用文本重复性测试
const chars = new Set(decodedText);
let charLen = chars.size;
if (charLen > leastCharLen) continue;
if (chars.has("\ufffd")) {
// 发现 REPLACEMENT CHARACTER，每个替代符视为独立字符，并至少增加1
const rplCharLen = decodedText.split("\ufffd").length - 1;
charLen += Math.max(rplCharLen, 1);
}
results.push({
encoding,
charLen: charLen,
});
if (charLen < leastCharLen) leastCharLen = charLen;
}
const ret = results.find((e) => e.charLen === leastCharLen);
// 没有有效charset时回退到 UTF-8
return ret?.encoding || "utf-8";
};
const detectBOM = (u8: Uint8Array): string | null => {
// UTF-8 BOM
if (u8[0] === 0xef && u8[1] === 0xbb && u8[2] === 0xbf) return "utf-8";
// UTF-16
if (u8[0] === 0xff && u8[1] === 0xfe) return u8[2] === 0 && u8[3] === 0 ? "utf-32le" : "utf-16le";
if (u8[0] === 0xfe && u8[1] === 0xff) return "utf-16be";
// UTF-32BE
if (u8[0] === 0 && u8[1] === 0 && u8[2] === 0xfe && u8[3] === 0xff) return "utf-32be";
return null;
};
function guessByNullPattern(u8: Uint8Array, size = u8.length): string | null {
if (size < 64) return null;
const n = [0, 0, 0, 0]; // n[0..3] = count of nulls at pos % 4
for (let i = 0; i < size; i++) {
if (u8[i] === 0) n[i & 3]++;
}
const total = n[0] + n[1] + n[2] + n[3];
if (total < size * 0.07) return null;
const density = total / size;
// UTF-32 — expect ~75% nulls (even with many CJK chars still ~70%+)
if (density > 0.54) {
const beScore = n[0] + n[1] + n[2];
const leScore = n[1] + n[2] + n[3];
if (beScore > leScore * 4.5 && beScore > size * 0.3) return "utf-32be";
if (leScore > beScore * 4.5 && leScore > size * 0.3) return "utf-32le";
return null;
}
// UTF-16 — expect ~25–50% nulls depending on script
if (density > 0.1) {
const even = n[0] + n[2];
const odd = n[1] + n[3];
if (even > odd * 4.2) return "utf-16be";
if (odd > even * 4.2) return "utf-16le";
if (even > odd * 2.1 && density > 0.2) return "utf-16be";
if (odd > even * 2.1 && density > 0.2) return "utf-16le";
}
return null;
}
/**
* Reads a Blob or File with reasonably good encoding detection
* Priority: Content-Type header → BOM → strong UTF-16 heuristics → UTF-8 validation → legacy fallback ("windows-1252")
* @param blob
* @returns {Promise<string>}
*/
export const readBlobContent = async (blob: Blob | File | Response, contentType: string | null): Promise<string> => {
const buffer = await blob.arrayBuffer();
const uint8 = new Uint8Array(buffer);
if (uint8.length === 0) {
return "";
}
// 优先尝试使用 Content-Type header 中的 charset
const headerCharset = parseCharsetFromContentType(contentType);
if (headerCharset) {
try {
// 验证 charset 是否有效
return bytesDecode(headerCharset, uint8);
} catch (e: any) {
console.warn(`Invalid charset from Content-Type header: ${headerCharset}, error: ${e.message}`);
}
}
// BOM detection (after Content-Type header)
const bomEncoding = detectBOM(uint8);
if (bomEncoding) return bytesDecode(bomEncoding, uint8);
const checkSize = Math.min(uint8.length, 16 * 1024);
if (uint8.length % 2 === 0) {
// Heuristic detection (first 16 KB)
const heuristicEncoding = guessByNullPattern(uint8, checkSize);
if (heuristicEncoding) return bytesDecode(heuristicEncoding, uint8);
}
// UTF-8 validation → legacy fallback
let encoding = "utf-8";
try {
// Strict mode – throws on invalid sequences
new TextDecoder("utf-8", { fatal: true }).decode(uint8.subarray(0, checkSize));
} catch {
// Invalid UTF-8 → most common real-world fallback
encoding = "windows-1252"; // OR detectEncoding(uint8, null)
}
return bytesDecode(encoding, uint8);
};