-
Notifications
You must be signed in to change notification settings - Fork 2
/
CharRefTokenizer.swift
228 lines (222 loc) · 8.65 KB
/
CharRefTokenizer.swift
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
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
import Collections
private import HTMLEntities
private enum CharRefState {
case initial
case named
case ambiguousAmpersand
case numeric
case hexadecimalStart(uppercase: Bool)
case decimalStart
case hexadecimal
case decimal
case numericEnd
}
private enum CharRefProcessResult: ~Copyable {
case done([Unicode.Scalar])
case doneNone
case progress
}
struct CharRefTokenizer {
private var state: CharRefState = .initial
private var num: Int = 0
private var numTooBig: Bool = false
private var nameBuffer: String = ""
private var lastMatch: (endIndex: String.Index, c1: Unicode.Scalar, c2: Unicode.Scalar)?
private let isInAttr: Bool
init(inAttr isInAttr: Bool) {
self.isInAttr = isInAttr
}
mutating func tokenize(tokenizer: inout Tokenizer<some TokenSink>, input: inout Deque<Character>) -> [Unicode.Scalar]? {
repeat {
switch self.step(tokenizer: &tokenizer, input: &input) {
case .done(let scalars): return scalars
case .doneNone: return nil
case .progress: break
}
} while true
}
private mutating func step(tokenizer: inout Tokenizer<some TokenSink>, input: inout Deque<Character>) -> CharRefProcessResult {
switch self.state {
case .initial:
switch tokenizer.peek(input)?.firstScalar {
case ("0"..."9")?, ("A"..."Z")?, ("a"..."z")?:
self.state = .named
return .progress
case "#":
tokenizer.discardChar(&input)
self.state = .numeric
return .progress
case _: return .done(["&"])
}
case .named:
guard let c = tokenizer.peek(input) else {
tokenizer.processCharRef("&")
input.prepend(contentsOf: self.nameBuffer)
return .doneNone
}
tokenizer.discardChar(&input)
self.nameBuffer.append(c)
if let (c1, c2) = processedNamedChars[self.nameBuffer] {
if c1 != "\0" {
self.lastMatch = (self.nameBuffer.endIndex, c1, c2)
}
return .progress
} else if let (endIndex, c1, c2) = self.lastMatch {
// swift-format-ignore: NeverForceUnwrap
let lastChar = self.nameBuffer[..<endIndex].last!.firstScalar
let nextChar = self.nameBuffer[endIndex].firstScalar
switch (isInAttr, lastChar, nextChar) {
case (_, ";", _): break
case (true, _, "="), (true, _, "0"..."9"), (true, _, "A"..."Z"), (true, _, "a"..."z"):
tokenizer.processCharRef("&")
input.prepend(contentsOf: self.nameBuffer)
return .doneNone
case _: tokenizer.emitError(.missingSemicolon)
}
return if c2 != "\0" {
.done([c1, c2])
} else {
.done([c1])
}
} else {
self.state = .ambiguousAmpersand
return .progress
}
case .ambiguousAmpersand:
guard let c = tokenizer.peek(input) else { return .doneNone }
switch c.firstScalar {
case "0"..."9", "A"..."Z", "a"..."z":
tokenizer.discardChar(&input)
self.nameBuffer.append(c)
return .progress
case ";": tokenizer.emitError(.unknownNamedCharRef)
case _: break
}
tokenizer.processCharRef("&")
input.prepend(contentsOf: self.nameBuffer)
return .doneNone
case .numeric:
switch tokenizer.peek(input) {
case "X":
tokenizer.discardChar(&input)
self.state = .hexadecimalStart(uppercase: true)
return .progress
case "x":
tokenizer.discardChar(&input)
self.state = .hexadecimalStart(uppercase: false)
return .progress
case _:
self.state = .decimalStart
return .progress
}
case .hexadecimalStart(let uppercase):
switch tokenizer.peek(input)?.firstScalar {
case ("0"..."9")?, ("A"..."F")?, ("a"..."f")?:
self.state = .hexadecimal
return .progress
case _:
tokenizer.emitError(.absenceDigits)
return .done(["&", "#", uppercase ? "X" : "x"])
}
case .decimalStart:
switch tokenizer.peek(input)?.firstScalar {
case ("0"..."9")?:
self.state = .decimal
return .progress
case _:
tokenizer.emitError(.absenceDigits)
return .done(["&", "#"])
}
case .hexadecimal:
if let firstScalar = tokenizer.peek(input)?.firstScalar {
switch firstScalar {
case "0"..."9":
tokenizer.discardChar(&input)
self.num &*= 16
if self.num > 0x10FFFF {
self.numTooBig = true
}
self.num &+= Int(firstScalar.value &- 0x30)
return .progress
case "A"..."F":
tokenizer.discardChar(&input)
self.num &*= 16
if self.num > 0x10FFFF {
self.numTooBig = true
}
self.num &+= Int(firstScalar.value &- 0x37)
return .progress
case "a"..."f":
tokenizer.discardChar(&input)
self.num &*= 16
if self.num > 0x10FFFF {
self.numTooBig = true
}
self.num &+= Int(firstScalar.value &- 0x57)
return .progress
case ";":
tokenizer.discardChar(&input)
self.state = .numericEnd
return .progress
case _: break
}
}
tokenizer.emitError(.missingSemicolon)
self.state = .numericEnd
return .progress
case .decimal:
if let firstScalar = tokenizer.peek(input)?.firstScalar {
switch firstScalar {
case "0"..."9":
tokenizer.discardChar(&input)
self.num &*= 10
if self.num > 0x10FFFF {
self.numTooBig = true
}
self.num &+= Int(firstScalar.value &- 0x30)
return .progress
case ";":
tokenizer.discardChar(&input)
self.state = .numericEnd
return .progress
case _: break
}
}
tokenizer.emitError(.missingSemicolon)
self.state = .numericEnd
return .progress
case .numericEnd:
// swift-format-ignore: NeverForceUnwrap
@inline(__always)
func conv(_ n: Int) -> Unicode.Scalar { .init(n)! }
switch self.num {
case 0x00:
tokenizer.emitError(.nullCharRef)
return .done(["\u{FFFD}"])
case let n where n > 0x10FFFF || self.numTooBig:
tokenizer.emitError(.charRefOutOfRange)
return .done(["\u{FFFD}"])
case 0xD800...0xDBFF, 0xDC00...0xDFFF:
tokenizer.emitError(.surrogateCharRef)
return .done(["\u{FFFD}"])
case 0xFDD0...0xFDEF, 0xFFFE, 0xFFFF, 0x1FFFE, 0x1FFFF, 0x2FFFE, 0x2FFFF,
0x3FFFE, 0x3FFFF, 0x4FFFE, 0x4FFFF, 0x5FFFE, 0x5FFFF, 0x6FFFE, 0x6FFFF,
0x7FFFE, 0x7FFFF, 0x8FFFE, 0x8FFFF, 0x9FFFE, 0x9FFFF, 0xAFFFE, 0xAFFFF,
0xBFFFE, 0xBFFFF, 0xCFFFE, 0xCFFFF, 0xDFFFE, 0xDFFFF, 0xEFFFE, 0xEFFFF,
0xFFFFE, 0xFFFFF, 0x10FFFE, 0x10FFFF:
tokenizer.emitError(.noncharacterCharRef)
return .done([conv(self.num)])
case 0x0D, 0x01...0x08, 0x0B, 0x0D...0x1F, 0x7F:
tokenizer.emitError(.controlCharRef)
return .done([conv(self.num)])
case 0x80...0x9F:
tokenizer.emitError(.controlCharRef)
return switch replacements[self.num &- 0x80] {
case "\0": .done([conv(self.num)])
case let c: .done([c])
}
case let n: return .done([conv(n)])
}
}
}
}