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
|
/*
Ousía
Copyright (C) 2014 Benjamin Paaßen, Andreas Stöckel
This program is free software: you can redistribute it and/or modify
it under the terms of the GNU General Public License as published by
the Free Software Foundation, either version 3 of the License, or
(at your option) any later version.
This program is distributed in the hope that it will be useful,
but WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
GNU General Public License for more details.
You should have received a copy of the GNU General Public License
along with this program. If not, see <http://www.gnu.org/licenses/>.
*/
#include <sstream>
#include "Tokenizer.hpp"
namespace ousia {
static std::map<char, TokenTreeNode> buildChildren(
const std::map<std::string, int> &inputs)
{
std::map<char, TokenTreeNode> children;
std::map<char, std::map<std::string, int>> nexts;
for (auto &e : inputs) {
const std::string &s = e.first;
const int id = e.second;
if (s.empty()) {
continue;
}
char start = s[0];
const std::string suffix = s.substr(1);
if (nexts.find(start) != nexts.end()) {
nexts[start].insert(std::make_pair(suffix, id));
} else {
nexts.insert(std::make_pair(
start, std::map<std::string, int>{{suffix, id}}));
}
}
for (auto &n : nexts) {
children.insert(std::make_pair(n.first, TokenTreeNode{n.second}));
}
return children;
}
static int buildId(const std::map<std::string, int> &inputs)
{
int tokenId = TOKEN_NONE;
for (auto &e : inputs) {
if (e.first.empty()) {
if (tokenId != TOKEN_NONE) {
throw TokenizerException{std::string{"Ambigous token found: "} +
std::to_string(e.second)};
} else {
tokenId = e.second;
}
}
}
return tokenId;
}
TokenTreeNode::TokenTreeNode(const std::map<std::string, int> &inputs)
: children(buildChildren(inputs)), tokenId(buildId(inputs))
{
}
Tokenizer::Tokenizer(BufferedCharReader &input, const TokenTreeNode &root)
: input(input), root(root)
{
}
bool Tokenizer::prepare()
{
std::stringstream buffer;
char c;
int startColumn = input.getColumn();
int startLine = input.getLine();
bool bufEmpty = true;
while (input.peek(&c)) {
if (root.children.find(c) != root.children.end()) {
// if there might be a special token, keep peeking forward
// until we find the token (or we don't).
TokenTreeNode const *n = &root;
std::stringstream tBuf;
int match = TOKEN_NONE;
while (true) {
tBuf << c;
n = &(n->children.at(c));
if (n->tokenId != TOKEN_NONE) {
match = n->tokenId;
// from here on we found a token. If we have something
// in our buffer already, we end the search now.
if (!bufEmpty) {
break;
} else {
// if we want to return this token ( = we have nothing
// in our buffer yet) we look greedily for the longest
// possible token we can construct.
input.consumePeek();
}
}
if (!input.peek(&c)) {
// if we are at the end we break off the search.
break;
}
if (n->children.find(c) == n->children.end()) {
// if we do not find a possible continuation anymore,
// break off the search.
break;
}
}
//reset the peek pointer to the last valid position.
input.resetPeek();
// check if we did indeed find a special token.
if (match != TOKEN_NONE) {
if (bufEmpty) {
// if we did not have text before, construct that token.
if (doPrepare(
Token{match, tBuf.str(), startColumn, startLine,
input.getColumn(), input.getLine()},
peeked)) {
return true;
} else {
startColumn = input.getColumn();
startLine = input.getLine();
continue;
}
} else {
// otherwise we return the text before the token.
if (doPrepare(Token{TOKEN_TEXT, buffer.str(), startColumn,
startLine, input.getColumn(),
input.getLine()},
peeked)) {
return true;
} else{
//we need to clear the buffer here. After all the token
//corresponding to this buffer segment is already
//constructed.
buffer.str(std::string());
bufEmpty = true;
startColumn = input.getColumn();
startLine = input.getLine();
continue;
}
}
} else{
//if we found nothing, read at least one character.
input.peek(&c);
}
}
buffer << c;
bufEmpty = false;
input.consumePeek();
}
if (!bufEmpty) {
return doPrepare(Token{TOKEN_TEXT, buffer.str(), startColumn, startLine,
input.getColumn(), input.getLine()},
peeked);
}
return false;
}
bool Tokenizer::doPrepare(const Token &t, std::deque<Token> &peeked)
{
peeked.push_back(t);
return true;
}
bool Tokenizer::next(Token &t)
{
if (peeked.empty()) {
if (!prepare()) {
return false;
}
}
t = peeked.front();
peeked.pop_front();
resetPeek();
return true;
}
bool Tokenizer::peek(Token &t)
{
if (peekCursor >= peeked.size()) {
if (!prepare()) {
return false;
}
}
t = peeked[peekCursor];
return true;
}
void Tokenizer::resetPeek() { peekCursor = 0; }
void Tokenizer::consumePeek()
{
while (peekCursor > 0) {
peeked.pop_front();
peekCursor--;
}
}
}
|
