@atproto/lex-data 0.0.2 → 0.0.4
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- package/CHANGELOG.md +20 -0
- package/dist/blob.d.ts +6 -0
- package/dist/blob.d.ts.map +1 -1
- package/dist/blob.js +4 -1
- package/dist/blob.js.map +1 -1
- package/dist/cid.d.ts +2 -3
- package/dist/cid.d.ts.map +1 -1
- package/dist/cid.js.map +1 -1
- package/dist/language.d.ts +2 -2
- package/dist/language.d.ts.map +1 -1
- package/dist/language.js +4 -4
- package/dist/language.js.map +1 -1
- package/dist/lex.d.ts.map +1 -1
- package/dist/lex.js +49 -48
- package/dist/lex.js.map +1 -1
- package/dist/object.d.ts +13 -1
- package/dist/object.d.ts.map +1 -1
- package/dist/object.js +15 -2
- package/dist/object.js.map +1 -1
- package/dist/uint8array-base64.d.ts +3 -0
- package/dist/uint8array-base64.d.ts.map +1 -0
- package/dist/uint8array-base64.js +3 -0
- package/dist/uint8array-base64.js.map +1 -0
- package/dist/uint8array-from-base64.d.ts +4 -3
- package/dist/uint8array-from-base64.d.ts.map +1 -1
- package/dist/uint8array-from-base64.js +9 -6
- package/dist/uint8array-from-base64.js.map +1 -1
- package/dist/uint8array-to-base64.d.ts +4 -3
- package/dist/uint8array-to-base64.d.ts.map +1 -1
- package/dist/uint8array-to-base64.js +7 -6
- package/dist/uint8array-to-base64.js.map +1 -1
- package/dist/uint8array.d.ts +6 -3
- package/dist/uint8array.d.ts.map +1 -1
- package/dist/uint8array.js +2 -1
- package/dist/uint8array.js.map +1 -1
- package/package.json +1 -1
- package/src/blob.ts +12 -1
- package/src/cid.ts +7 -7
- package/src/language.test.ts +33 -33
- package/src/language.ts +2 -2
- package/src/lex.test.ts +104 -1
- package/src/lex.ts +50 -43
- package/src/object.ts +16 -4
- package/src/uint8array-base64.ts +2 -0
- package/src/uint8array-from-base64.test.ts +33 -19
- package/src/uint8array-from-base64.ts +19 -6
- package/src/uint8array-to-base64.test.ts +170 -0
- package/src/uint8array-to-base64.ts +18 -8
- package/src/uint8array.ts +13 -5
- package/tsconfig.tests.json +1 -1
|
@@ -1,5 +1,6 @@
|
|
|
1
1
|
import { fromString } from 'uint8arrays/from-string'
|
|
2
2
|
import { NodeJSBuffer } from './lib/nodejs-buffer.js'
|
|
3
|
+
import { Base64Alphabet } from './uint8array-base64.js'
|
|
3
4
|
|
|
4
5
|
const Buffer = NodeJSBuffer
|
|
5
6
|
|
|
@@ -21,14 +22,23 @@ declare global {
|
|
|
21
22
|
|
|
22
23
|
export const fromBase64Native =
|
|
23
24
|
typeof Uint8Array.fromBase64 === 'function'
|
|
24
|
-
? function fromBase64Native(
|
|
25
|
-
|
|
25
|
+
? function fromBase64Native(
|
|
26
|
+
b64: string,
|
|
27
|
+
alphabet: Base64Alphabet = 'base64',
|
|
28
|
+
): Uint8Array {
|
|
29
|
+
return Uint8Array.fromBase64!(b64, {
|
|
30
|
+
alphabet,
|
|
31
|
+
lastChunkHandling: 'loose',
|
|
32
|
+
})
|
|
26
33
|
}
|
|
27
34
|
: null
|
|
28
35
|
|
|
29
36
|
export const fromBase64Node = Buffer
|
|
30
|
-
? function fromBase64Node(
|
|
31
|
-
|
|
37
|
+
? function fromBase64Node(
|
|
38
|
+
b64: string,
|
|
39
|
+
alphabet: Base64Alphabet = 'base64',
|
|
40
|
+
): Uint8Array {
|
|
41
|
+
const bytes = Buffer.from(b64, alphabet)
|
|
32
42
|
verifyBase64ForBytes(b64, bytes)
|
|
33
43
|
// Convert to Uint8Array because even though Buffer is a sub class of
|
|
34
44
|
// Uint8Array, it serializes differently to Uint8Array (e.g. in JSON) and
|
|
@@ -37,8 +47,11 @@ export const fromBase64Node = Buffer
|
|
|
37
47
|
}
|
|
38
48
|
: null
|
|
39
49
|
|
|
40
|
-
export function fromBase64Ponyfill(
|
|
41
|
-
|
|
50
|
+
export function fromBase64Ponyfill(
|
|
51
|
+
b64: string,
|
|
52
|
+
alphabet: Base64Alphabet = 'base64',
|
|
53
|
+
): Uint8Array {
|
|
54
|
+
const bytes = fromString(b64, b64.endsWith('=') ? `${alphabet}pad` : alphabet)
|
|
42
55
|
verifyBase64ForBytes(b64, bytes)
|
|
43
56
|
return bytes
|
|
44
57
|
}
|
|
@@ -0,0 +1,170 @@
|
|
|
1
|
+
import 'core-js/modules/es.uint8-array.from-base64.js'
|
|
2
|
+
import 'core-js/modules/es.uint8-array.to-base64.js'
|
|
3
|
+
import assert from 'node:assert'
|
|
4
|
+
import {
|
|
5
|
+
toBase64Native,
|
|
6
|
+
toBase64Node,
|
|
7
|
+
toBase64Ponyfill,
|
|
8
|
+
} from './uint8array-to-base64.js'
|
|
9
|
+
|
|
10
|
+
for (const toBase64 of [
|
|
11
|
+
toBase64Native,
|
|
12
|
+
toBase64Node,
|
|
13
|
+
toBase64Ponyfill,
|
|
14
|
+
] as const) {
|
|
15
|
+
// Tests should run in NodeJS where implementations are either available or
|
|
16
|
+
// polyfilled (see core-js imports above).
|
|
17
|
+
assert(toBase64 !== null, 'toBase64 implementation should not be null')
|
|
18
|
+
|
|
19
|
+
describe(toBase64.name, () => {
|
|
20
|
+
describe('basic encoding', () => {
|
|
21
|
+
it('encodes empty Uint8Array', () => {
|
|
22
|
+
const encoded = toBase64(new Uint8Array(0))
|
|
23
|
+
expect(typeof encoded).toBe('string')
|
|
24
|
+
expect(encoded).toBe('')
|
|
25
|
+
})
|
|
26
|
+
|
|
27
|
+
it('encodes 10MB', () => {
|
|
28
|
+
const bytes = Buffer.allocUnsafe(10_000_000).fill('🐩')
|
|
29
|
+
const encoded = toBase64(bytes)
|
|
30
|
+
expect(typeof encoded).toBe('string')
|
|
31
|
+
// Verify by decoding back
|
|
32
|
+
const decoded = Buffer.from(encoded, 'base64')
|
|
33
|
+
expect(decoded.equals(bytes)).toBe(true)
|
|
34
|
+
})
|
|
35
|
+
})
|
|
36
|
+
|
|
37
|
+
describe('base64 alphabet (default)', () => {
|
|
38
|
+
for (const string of [
|
|
39
|
+
'',
|
|
40
|
+
'\0\0',
|
|
41
|
+
'\0\0\0',
|
|
42
|
+
'\0\0\0\0',
|
|
43
|
+
'__',
|
|
44
|
+
'é',
|
|
45
|
+
'àç',
|
|
46
|
+
'\0éàç',
|
|
47
|
+
'```\x1b',
|
|
48
|
+
'aaa',
|
|
49
|
+
'Hello, World!',
|
|
50
|
+
'😀😃😄😁😆😅😂🤣😊😇',
|
|
51
|
+
'👩💻👨💻👩🔬👨🔬👩🚀👨🚀',
|
|
52
|
+
'🌍🌎🌏🌐🪐🌟✨⚡🔥💧',
|
|
53
|
+
] as const) {
|
|
54
|
+
const buffer = Buffer.from(string, 'utf8')
|
|
55
|
+
const expected = buffer.toString('base64').replace(/=+$/, '')
|
|
56
|
+
|
|
57
|
+
it(`encodes ${JSON.stringify(string)} as ${JSON.stringify(expected)}`, () => {
|
|
58
|
+
const encoded = toBase64(buffer)
|
|
59
|
+
expect(encoded).toBe(expected)
|
|
60
|
+
})
|
|
61
|
+
}
|
|
62
|
+
})
|
|
63
|
+
|
|
64
|
+
describe('base64url alphabet', () => {
|
|
65
|
+
for (const string of [
|
|
66
|
+
'',
|
|
67
|
+
'\0\0',
|
|
68
|
+
'\0\0\0',
|
|
69
|
+
'\0\0\0\0',
|
|
70
|
+
'__',
|
|
71
|
+
'é',
|
|
72
|
+
'àç',
|
|
73
|
+
'\0éàç',
|
|
74
|
+
'```\x1b',
|
|
75
|
+
'aaa',
|
|
76
|
+
'Hello, World!',
|
|
77
|
+
'😀😃😄😁😆😅😂🤣😊😇',
|
|
78
|
+
'👩💻👨💻👩🔬👨🔬👩🚀👨🚀',
|
|
79
|
+
'🌍🌎🌏🌐🪐🌟✨⚡🔥💧',
|
|
80
|
+
] as const) {
|
|
81
|
+
const buffer = Buffer.from(string, 'utf8')
|
|
82
|
+
const expected = buffer.toString('base64url')
|
|
83
|
+
|
|
84
|
+
it(`encodes ${JSON.stringify(string)} as ${JSON.stringify(expected)}`, () => {
|
|
85
|
+
const encoded = toBase64(buffer, 'base64url')
|
|
86
|
+
expect(encoded).toBe(expected)
|
|
87
|
+
})
|
|
88
|
+
}
|
|
89
|
+
})
|
|
90
|
+
|
|
91
|
+
describe('base64 vs base64url character differences', () => {
|
|
92
|
+
// Test data that produces + and / in standard base64
|
|
93
|
+
// These should become - and _ in base64url
|
|
94
|
+
it('uses + and / for base64 alphabet', () => {
|
|
95
|
+
// 0xfb, 0xff, 0xbf produces "+/+/" in base64
|
|
96
|
+
const bytes = new Uint8Array([0xfb, 0xff, 0xbf])
|
|
97
|
+
const encoded = toBase64(bytes)
|
|
98
|
+
expect(encoded).toContain('+')
|
|
99
|
+
expect(encoded).toContain('/')
|
|
100
|
+
expect(encoded).not.toContain('-')
|
|
101
|
+
expect(encoded).not.toContain('_')
|
|
102
|
+
})
|
|
103
|
+
|
|
104
|
+
it('uses - and _ for base64url alphabet', () => {
|
|
105
|
+
// Same bytes should use - and _ in base64url
|
|
106
|
+
const bytes = new Uint8Array([0xfb, 0xff, 0xbf])
|
|
107
|
+
const encoded = toBase64(bytes, 'base64url')
|
|
108
|
+
expect(encoded).toContain('-')
|
|
109
|
+
expect(encoded).toContain('_')
|
|
110
|
+
expect(encoded).not.toContain('+')
|
|
111
|
+
expect(encoded).not.toContain('/')
|
|
112
|
+
})
|
|
113
|
+
})
|
|
114
|
+
|
|
115
|
+
describe('padding behavior', () => {
|
|
116
|
+
it('omits padding by default for 1-byte input', () => {
|
|
117
|
+
// 1 byte -> 2 base64 chars + 2 padding
|
|
118
|
+
const bytes = new Uint8Array([0x4d]) // 'M' -> 'TQ=='
|
|
119
|
+
const encoded = toBase64(bytes)
|
|
120
|
+
expect(encoded).toBe('TQ')
|
|
121
|
+
expect(encoded).not.toContain('=')
|
|
122
|
+
})
|
|
123
|
+
|
|
124
|
+
it('omits padding by default for 2-byte input', () => {
|
|
125
|
+
// 2 bytes -> 3 base64 chars + 1 padding
|
|
126
|
+
const bytes = new Uint8Array([0x4d, 0x61]) // 'Ma' -> 'TWE='
|
|
127
|
+
const encoded = toBase64(bytes)
|
|
128
|
+
expect(encoded).toBe('TWE')
|
|
129
|
+
expect(encoded).not.toContain('=')
|
|
130
|
+
})
|
|
131
|
+
|
|
132
|
+
it('no padding needed for 3-byte input', () => {
|
|
133
|
+
// 3 bytes -> 4 base64 chars, no padding needed
|
|
134
|
+
const bytes = new Uint8Array([0x4d, 0x61, 0x6e]) // 'Man' -> 'TWFu'
|
|
135
|
+
const encoded = toBase64(bytes)
|
|
136
|
+
expect(encoded).toBe('TWFu')
|
|
137
|
+
})
|
|
138
|
+
|
|
139
|
+
it('includes double padding when omitPadding: false for 1-byte input', () => {
|
|
140
|
+
const bytes = new Uint8Array([0x4d]) // 'M' -> 'TQ=='
|
|
141
|
+
const encoded = toBase64(bytes)
|
|
142
|
+
expect(encoded).toBe('TQ')
|
|
143
|
+
})
|
|
144
|
+
|
|
145
|
+
it('includes single padding when omitPadding: false for 2-byte input', () => {
|
|
146
|
+
const bytes = new Uint8Array([0x4d, 0x61]) // 'Ma' -> 'TWE='
|
|
147
|
+
const encoded = toBase64(bytes)
|
|
148
|
+
expect(encoded).toBe('TWE')
|
|
149
|
+
})
|
|
150
|
+
})
|
|
151
|
+
|
|
152
|
+
describe('Uint8Array subarray handling', () => {
|
|
153
|
+
it('correctly encodes a subarray', () => {
|
|
154
|
+
const fullArray = new Uint8Array([0x00, 0x4d, 0x61, 0x6e, 0x00])
|
|
155
|
+
const subarray = fullArray.subarray(1, 4) // 'Man'
|
|
156
|
+
const encoded = toBase64(subarray)
|
|
157
|
+
expect(encoded).toBe('TWFu')
|
|
158
|
+
})
|
|
159
|
+
|
|
160
|
+
it('correctly encodes a subarray with offset', () => {
|
|
161
|
+
const buffer = new ArrayBuffer(10)
|
|
162
|
+
const fullView = new Uint8Array(buffer)
|
|
163
|
+
fullView.set([0x00, 0x00, 0x4d, 0x61, 0x6e, 0x00, 0x00])
|
|
164
|
+
const subarray = new Uint8Array(buffer, 2, 3) // 'Man' at offset 2
|
|
165
|
+
const encoded = toBase64(subarray)
|
|
166
|
+
expect(encoded).toBe('TWFu')
|
|
167
|
+
})
|
|
168
|
+
})
|
|
169
|
+
})
|
|
170
|
+
}
|
|
@@ -1,5 +1,6 @@
|
|
|
1
1
|
import { toString } from 'uint8arrays/to-string'
|
|
2
2
|
import { NodeJSBuffer } from './lib/nodejs-buffer.js'
|
|
3
|
+
import { Base64Alphabet } from './uint8array-base64.js'
|
|
3
4
|
|
|
4
5
|
const Buffer = NodeJSBuffer
|
|
5
6
|
|
|
@@ -18,16 +19,22 @@ declare global {
|
|
|
18
19
|
|
|
19
20
|
export const toBase64Native =
|
|
20
21
|
typeof Uint8Array.prototype.toBase64 === 'function'
|
|
21
|
-
? function toBase64Native(
|
|
22
|
-
|
|
22
|
+
? function toBase64Native(
|
|
23
|
+
bytes: Uint8Array,
|
|
24
|
+
alphabet: Base64Alphabet = 'base64',
|
|
25
|
+
): string {
|
|
26
|
+
return bytes.toBase64!({ alphabet, omitPadding: true })
|
|
23
27
|
}
|
|
24
28
|
: null
|
|
25
29
|
|
|
26
30
|
export const toBase64Node = Buffer
|
|
27
|
-
? function toBase64Node(
|
|
28
|
-
|
|
29
|
-
|
|
30
|
-
|
|
31
|
+
? function toBase64Node(
|
|
32
|
+
bytes: Uint8Array,
|
|
33
|
+
alphabet: Base64Alphabet = 'base64',
|
|
34
|
+
): string {
|
|
35
|
+
const buffer = bytes instanceof Buffer ? bytes : Buffer.from(bytes)
|
|
36
|
+
const b64 = buffer.toString(alphabet)
|
|
37
|
+
|
|
31
38
|
// @NOTE We strip padding for strict compatibility with
|
|
32
39
|
// uint8arrays.toString behavior. Tests failing because of the presence of
|
|
33
40
|
// padding are not really synonymous with an actual error and we might
|
|
@@ -40,6 +47,9 @@ export const toBase64Node = Buffer
|
|
|
40
47
|
}
|
|
41
48
|
: null
|
|
42
49
|
|
|
43
|
-
export function toBase64Ponyfill(
|
|
44
|
-
|
|
50
|
+
export function toBase64Ponyfill(
|
|
51
|
+
bytes: Uint8Array,
|
|
52
|
+
alphabet: Base64Alphabet = 'base64',
|
|
53
|
+
): string {
|
|
54
|
+
return toString(bytes, alphabet)
|
|
45
55
|
}
|
package/src/uint8array.ts
CHANGED
|
@@ -1,3 +1,4 @@
|
|
|
1
|
+
import { Base64Alphabet } from './uint8array-base64.js'
|
|
1
2
|
import {
|
|
2
3
|
fromBase64Native,
|
|
3
4
|
fromBase64Node,
|
|
@@ -9,6 +10,8 @@ import {
|
|
|
9
10
|
toBase64Ponyfill,
|
|
10
11
|
} from './uint8array-to-base64.js'
|
|
11
12
|
|
|
13
|
+
export type { Base64Alphabet }
|
|
14
|
+
|
|
12
15
|
// @TODO drop dependency on uint8arrays package once Uint8Array.fromBase64 /
|
|
13
16
|
// Uint8Array.prototype.toBase64 is widely supported, and mark fromBase64 /
|
|
14
17
|
// toBase64 as deprecated. We can also drop NodeJS specific implementations
|
|
@@ -19,17 +22,22 @@ import {
|
|
|
19
22
|
*
|
|
20
23
|
* @returns The base64 encoded string
|
|
21
24
|
*/
|
|
22
|
-
export const toBase64: (
|
|
23
|
-
|
|
25
|
+
export const toBase64: (
|
|
26
|
+
bytes: Uint8Array,
|
|
27
|
+
alphabet?: Base64Alphabet,
|
|
28
|
+
) => string = toBase64Native ?? toBase64Node ?? toBase64Ponyfill
|
|
24
29
|
|
|
25
30
|
/**
|
|
26
|
-
* Decodes a base64 string into a Uint8Array.
|
|
31
|
+
* Decodes a base64 string into a Uint8Array. This function supports both padded
|
|
32
|
+
* and unpadded base64 strings.
|
|
27
33
|
*
|
|
28
34
|
* @returns The decoded {@link Uint8Array}
|
|
29
35
|
* @throws If the input is not a valid base64 string
|
|
30
36
|
*/
|
|
31
|
-
export const fromBase64: (
|
|
32
|
-
|
|
37
|
+
export const fromBase64: (
|
|
38
|
+
b64: string,
|
|
39
|
+
alphabet?: Base64Alphabet,
|
|
40
|
+
) => Uint8Array = fromBase64Native ?? fromBase64Node ?? fromBase64Ponyfill
|
|
33
41
|
|
|
34
42
|
if (toBase64 === toBase64Ponyfill || fromBase64 === fromBase64Ponyfill) {
|
|
35
43
|
/*#__PURE__*/
|