jcs.org
Serenity Operating System
1/*
2 * Copyright (c) 2022-2023, the SerenityOS developers.
3 *
4 * SPDX-License-Identifier: BSD-2-Clause
5 */
6
7#include <AK/Base64.h>
8#include <AK/ByteBuffer.h>
9#include <AK/CharacterTypes.h>
10#include <AK/Error.h>
11#include <AK/StringBuilder.h>
12#include <AK/StringView.h>
13#include <AK/Vector.h>
14#include <LibWeb/Infra/Base64.h>
15#include <LibWeb/Infra/CharacterTypes.h>
16
17namespace Web::Infra {
18
19// https://infra.spec.whatwg.org/#forgiving-base64
20ErrorOr<ByteBuffer> decode_forgiving_base64(StringView input)
21{
22 // 1. Remove all ASCII whitespace from data.
23 // FIXME: It is possible to avoid copying input here, it's just a bit tricky to remove the equal signs
24 StringBuilder builder;
25 for (auto character : input) {
26 if (!is_ascii_whitespace(character))
27 TRY(builder.try_append(character));
28 }
29 auto data = builder.string_view();
30
31 // 2. If data’s code point length divides by 4 leaving no remainder, then:
32 if (data.length() % 4 == 0) {
33 // If data ends with one or two U+003D (=) code points, then remove them from data.
34 if (data.ends_with("=="sv))
35 data = data.substring_view(0, data.length() - 2);
36 else if (data.ends_with('='))
37 data = data.substring_view(0, data.length() - 1);
38 }
39
40 // 3. If data’s code point length divides by 4 leaving a remainder of 1, then return failure.
41 if (data.length() % 4 == 1)
42 return Error::from_string_literal("Invalid input length in forgiving base64 decode");
43
44 // 4. If data contains a code point that is not one of
45 // U+002B (+), U+002F (/), ASCII alphanumeric
46 // then return failure.
47 for (auto point : data) {
48 if (point != '+' && point != '/' && !is_ascii_alphanumeric(point))
49 return Error::from_string_literal("Invalid character in forgiving base64 decode");
50 }
51
52 // 5. Let output be an empty byte sequence.
53 // 6. Let buffer be an empty buffer that can have bits appended to it.
54 Vector<u8> output;
55 u32 buffer = 0;
56 auto accumulated_bits = 0;
57
58 auto add_to_buffer = [&](u8 number) {
59 VERIFY(number < 64);
60 u32 buffer_mask = number;
61
62 if (accumulated_bits == 0)
63 buffer_mask <<= 18;
64 else if (accumulated_bits == 6)
65 buffer_mask <<= 12;
66 else if (accumulated_bits == 12)
67 buffer_mask <<= 6;
68 else if (accumulated_bits == 18)
69 buffer_mask <<= 0;
70
71 buffer |= buffer_mask;
72
73 accumulated_bits += 6;
74 };
75
76 auto append_bytes = [&]() {
77 output.append(static_cast<u8>((buffer & 0xff0000) >> 16));
78 output.append(static_cast<u8>((buffer & 0xff00) >> 8));
79 output.append(static_cast<u8>(buffer & 0xff));
80
81 buffer = 0;
82 accumulated_bits = 0;
83 };
84
85 auto alphabet_lookup_table = AK::base64_lookup_table();
86
87 // 7. Let position be a position variable for data, initially pointing at the start of data.
88 // 8. While position does not point past the end of data:
89 for (auto point : data) {
90 // 1. Find the code point pointed to by position in the second column of Table 1: The Base 64 Alphabet of RFC 4648.
91 // Let n be the number given in the first cell of the same row. [RFC4648]
92 auto n = alphabet_lookup_table[point];
93 VERIFY(n >= 0);
94
95 // 2. Append the six bits corresponding to n, most significant bit first, to buffer.
96 add_to_buffer(static_cast<u8>(n));
97
98 // 3. buffer has accumulated 24 bits,
99 if (accumulated_bits == 24) {
100 // interpret them as three 8-bit big-endian numbers.
101 // Append three bytes with values equal to those numbers to output, in the same order, and then empty buffer
102 append_bytes();
103 }
104 }
105
106 // 9. If buffer is not empty, it contains either 12 or 18 bits.
107 VERIFY(accumulated_bits == 0 || accumulated_bits == 12 || accumulated_bits == 18);
108
109 // If it contains 12 bits, then discard the last four and interpret the remaining eight as an 8-bit big-endian number.
110 if (accumulated_bits == 12)
111 output.append(static_cast<u8>((buffer & 0xff0000) >> 16));
112
113 // If it contains 18 bits, then discard the last two and interpret the remaining 16 as two 8-bit big-endian numbers.
114 // Append the one or two bytes with values equal to those one or two numbers to output, in the same order.
115 if (accumulated_bits == 18) {
116 output.append(static_cast<u8>((buffer & 0xff0000) >> 16));
117 output.append(static_cast<u8>((buffer & 0xff00) >> 8));
118 }
119
120 return ByteBuffer::copy(output);
121}
122
123}