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
|
#include "error.hpp"
#include "lexer.hpp"
#include "token.hpp"
#include "token_type.hpp"
#include <exception>
#include <optional>
#include <regex>
#include <string_view>
#include <string>
#include <vector>
namespace math::server {
namespace lexer {
namespace {
std::string_view match_number(const std::string_view& input) {
static constexpr std::regex::flag_type flags =
std::regex_constants::ECMAScript |
std::regex_constants::icase;
// This is a hacky attempt to describe a C-like grammar for floating-point
// numbers using a regex (the tests seem to pass though).
// A proper NFA would be better, I guess.
static const std::regex number_regex{R"REGEX(^(?:\d+(?:\.\d*)?|\.\d+)(e[+-]?(\d*))?)REGEX", flags};
std::cmatch match;
if (!std::regex_search(input.cbegin(), input.cend(), match, number_regex)) {
return {};
}
{
// If we have the numeric part of a number followed by 'e' and no digits,
// 1) that 'e' definitely belongs to this number token,
// 2) the user forgot to type in the required digits.
const auto& exponent = match[1];
const auto& abs_power = match[2];
if (exponent.matched && abs_power.matched && abs_power.length() == 0) {
throw LexerError{"exponent has no digits: " + match[0].str()};
}
}
return {match[0].first, match[0].length()};
}
std::optional<double> parse_number(const std::string_view& input, std::string_view& token) {
const auto view = match_number(input);
if (!view.data()) {
return {};
}
try {
const auto result = std::stod(std::string{view});
token = view;
return result;
} catch (const std::exception& e) {
throw LexerError{"internal: couldn't parse number from: " + std::string{view}};
}
return {};
}
bool starts_with(const std::string_view& a, const std::string_view& b) noexcept {
return a.length() >= b.length()
&& a.compare(0, b.length(), b) == 0;
}
std::optional<token::Type> parse_const_token(const std::string_view& input, std::string_view& token) {
for (const auto type : token::const_tokens()) {
const auto str = token::type_to_string(type);
if (starts_with(input, str)) {
token = {input.cbegin(), str.length()};
return {type};
}
}
return {};
}
}
namespace details {
std::optional<double> parse_number(const std::string_view& input) {
std::string_view token;
return lexer::parse_number(input, token);
}
std::optional<token::Type> parse_const_token(const std::string_view& input) {
std::string_view token;
return lexer::parse_const_token(input, token);
}
std::string_view parse_whitespace(const std::string_view& input) {
static const std::regex ws_regex{R"(^\s+)"};
std::cmatch match;
if (std::regex_search(input.cbegin(), input.cend(), match, ws_regex)) {
return {match[0].first, match[0].length()};
}
return {};
}
}
}
Lexer::Lexer(const std::string_view& input)
: Lexer{lexer::Input{input}} {
}
Lexer::Lexer(const lexer::Input& input)
: m_input{input} {
consume_token();
}
bool Lexer::for_each_token(const TokenProcessor& process) {
for (auto token = peek_token(); token.has_value(); drop_token(), token = peek_token()) {
if (!process(*token)) {
return false;
}
}
return true;
}
std::vector<Lexer::ParsedToken> Lexer::get_tokens() {
std::vector<ParsedToken> tokens;
for_each_token([&tokens] (const ParsedToken& token) {
tokens.emplace_back(token);
return true;
});
return tokens;
}
void Lexer::drop_token() {
if (!has_token()) {
throw LexerError{"internal: no tokens to drop"};
}
m_token_buffer = {};
consume_token();
}
std::optional<Lexer::ParsedToken> Lexer::drop_token_of_type(Type type) {
if (!has_token()) {
throw LexerError{"internal: no tokens to drop"};
}
if (m_token_buffer.value().get_type() != type) {
return {};
}
const auto result = m_token_buffer;
drop_token();
return result;
}
void Lexer::consume_whitespace() {
const auto ws = parse_whitespace();
if (!ws.has_value()) {
return;
}
m_input.consume(ws->get_length());
}
void Lexer::consume_token() {
if (m_input.empty()) {
return;
}
consume_whitespace();
if (m_input.empty()) {
return;
}
auto token{parse_token()};
m_input.consume(token.get_length());
m_token_buffer = std::move(token);
}
std::optional<Lexer::ParsedToken> Lexer::parse_whitespace() const {
const auto token_view = lexer::details::parse_whitespace(m_input.get_input());
if (token_view.empty()) {
return {};
}
return ParsedToken{Token{Token::Type::WHITESPACE}, m_input.get_pos(), token_view};
}
std::optional<Lexer::ParsedToken> Lexer::parse_const_token() const {
std::string_view token_view;
const auto type = lexer::parse_const_token(m_input.get_input(), token_view);
if (!type.has_value()) {
return {};
}
return ParsedToken{Token{*type}, m_input.get_pos(), token_view};
}
std::optional<Lexer::ParsedToken> Lexer::parse_number() const {
std::string_view token_view;
const auto number = lexer::parse_number(m_input.get_input(), token_view);
if (!number.has_value()) {
return {};
}
return ParsedToken{Token{*number}, m_input.get_pos(), token_view};
}
Lexer::ParsedToken Lexer::parse_token() const {
if (const auto const_token = parse_const_token(); const_token.has_value()) {
return *const_token;
}
if (const auto number = parse_number(); number.has_value()) {
return *number;
}
throw LexerError{"invalid input at: " + std::string{m_input.get_input()}};
}
}
|
