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
|
// Copyright (c) 2013-2021 Bruyère Jean-Philippe <jp_bruyere@hotmail.com>
//
// This code is licensed under the MIT license (MIT) (http://opensource.org/licenses/MIT)
using System;
using Crow.Text;
using System.Collections.Generic;
namespace Crow
{
public class Tokenizer {
enum States
{
Init,//first statement of prolog, xmldecl should only apear in this state
prolog,//misc before doctypedecl
ProcessingInstrucitons,
DTD,
DTDObject,//doctype finished
Xml,
StartTag,//inside start tag
Content,//after start tag with no closing slash
EndTag
}
States curState;
List<Token> Toks;
int startOfTok;
public Tokenizer () {}
void skipWhiteSpaces (ref SpanCharReader reader) {
while(!reader.EndOfSpan) {
switch (reader.Peak) {
case '\x85':
case '\x2028':
case '\xA':
reader.Read();
addTok (ref reader, TokenType.LineBreak);
break;
case '\xD':
reader.Read();
if (reader.IsNextCharIn ('\xA', '\x85'))
reader.Read();
addTok (ref reader, TokenType.LineBreak);
break;
case '\x20':
case '\x9':
char c = reader.Read();
while (reader.TryPeak (c))
reader.Read();
addTok (ref reader, c == '\x20' ? TokenType.WhiteSpace : TokenType.Tabulation);
break;
default:
return;
}
}
}
bool readName (ref SpanCharReader reader) {
if (reader.EndOfSpan)
return false;
char c = reader.Peak;
if (char.IsLetter(c) || c == '_' || c == ':') {
reader.Advance ();
while (reader.TryPeak (ref c)) {
if (!(char.IsLetterOrDigit(c) || c == '.' || c == '-' || c == '\xB7'))
return true;
reader.Advance ();
}
return true;
}
return false;
}
void addTok (ref SpanCharReader reader, TokenType tokType) {
if (reader.CurrentPosition == startOfTok)
return;
Toks.Add (new Token(tokType, startOfTok, reader.CurrentPosition));
startOfTok = reader.CurrentPosition;
}
public Token[] Tokenize (string source) {
SpanCharReader reader = new SpanCharReader(source);
startOfTok = 0;
int curObjectLevel = 0;
curState = States.Init;
Toks = new List<Token>(100);
while(!reader.EndOfSpan) {
skipWhiteSpaces (ref reader);
if (reader.EndOfSpan)
break;
switch (reader.Peak) {
case '<':
reader.Advance ();
if (reader.TryPeak ('?')) {
reader.Advance ();
addTok (ref reader, TokenType.PI_Start);
readName (ref reader);
addTok (ref reader, TokenType.PI_Target);
curState = States.ProcessingInstrucitons;
} else if (reader.TryPeak ('!')) {
reader.Advance ();
if (reader.TryPeak ("--")) {
reader.Advance (2);
addTok (ref reader, TokenType.BlockCommentStart);
if (reader.TryReadUntil ("-->")) {
addTok (ref reader, TokenType.BlockComment);
reader.Advance (3);
addTok (ref reader, TokenType.BlockCommentEnd);
} else if (reader.TryPeak ("-->")) {
reader.Advance (3);
addTok (ref reader, TokenType.BlockCommentEnd);
}
} else {
addTok (ref reader, TokenType.DTDObjectOpen);
if (readName (ref reader)) {
addTok (ref reader, TokenType.Keyword);
curState = States.DTDObject;
}
}
} else if (reader.TryPeak('/')) {
reader.Advance ();
addTok (ref reader, TokenType.EndElementOpen);
if (readName (ref reader)) {
addTok (ref reader, TokenType.ElementName);
if (reader.TryPeak('>')) {
reader.Advance ();
addTok (ref reader, TokenType.ClosingSign);
if (--curObjectLevel > 0)
curState = States.Content;
else
curState = States.Xml;
}
}
}else{
addTok (ref reader, TokenType.ElementOpen);
if (readName (ref reader)) {
addTok (ref reader, TokenType.ElementName);
curState = States.StartTag;
}
}
break;
case '?':
reader.Advance ();
if (reader.TryPeak ('>')){
reader.Advance ();
addTok (ref reader, TokenType.PI_End);
}else
addTok (ref reader, TokenType.Unknown);
curState = States.prolog;
break;
case '\'':
case '"':
char q = reader.Read();
addTok (ref reader, TokenType.AttributeValueOpen);
if (reader.TryReadUntil (q)) {
addTok (ref reader, TokenType.AttributeValue);
reader.Advance ();
addTok (ref reader, TokenType.AttributeValueClose);
} else
addTok (ref reader, TokenType.AttributeValue);
break;
case '=':
reader.Advance();
addTok (ref reader, TokenType.EqualSign);
break;
case '>':
reader.Advance();
addTok (ref reader, TokenType.ClosingSign);
curObjectLevel++;
curState = States.Content;
break;
case '/':
reader.Advance();
if (reader.TryRead ('>')) {
addTok (ref reader, TokenType.EmptyElementClosing);
if (--curObjectLevel > 0)
curState = States.Content;
else
curState = States.Xml;
}else
addTok (ref reader, TokenType.Unknown);
break;
default:
if (curState == States.StartTag || curState == States.ProcessingInstrucitons) {
if (readName(ref reader))
addTok (ref reader, TokenType.AttributeName);
else if (reader.TryAdvance())
addTok (ref reader, TokenType.Unknown);
} else {
reader.TryReadUntil ('<');
addTok (ref reader, TokenType.Content);
}
break;
}
}
return Toks.ToArray();
}
}
}
|
