lalrpop 0.23.0

convenient LR(1) parser generator
Documentation 
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
229
230
231
232
233
234
235
236
237
238
239
240
241
242

use crate::grammar::parse_tree::Grammar;
use crate::lexer::dfa::interpret;
use crate::normalize::NormResult;
use crate::normalize::resolve::resolve;
use crate::parser;
use crate::test_util;

fn validate_grammar(grammar: &str) -> NormResult<Grammar> {
    let parsed_grammar = parser::parse_grammar(grammar).expect("parse grammar");
    let parsed_grammar = resolve(parsed_grammar).expect("resolve");
    super::validate(parsed_grammar)
}

fn check_err(expected_err: &str, grammar: &str, span: &str) {
    let err = validate_grammar(grammar).unwrap_err();
    test_util::check_norm_err(expected_err, span, err);
}

fn check_intern_token(grammar: &str, expected_tokens: Vec<(&'static str, &'static str)>) {
    let parsed_grammar = validate_grammar(grammar).expect("validate");
    let intern_token = parsed_grammar.intern_token().expect("intern_token");
    println!("intern_token: {intern_token:?}");
    for (input, expected_user_name) in expected_tokens {
        let actual_user_name =
            interpret::interpret(&intern_token.dfa, input).map(|(index, text)| {
                let user_name = &intern_token.match_entries[index.index()].user_name;
                (user_name.clone(), text)
            });
        let actual_user_name = format!("{actual_user_name:?}");
        if expected_user_name != actual_user_name {
            panic!(
                "input `{input}` matched `{actual_user_name}` but we expected `{expected_user_name}`"
            );
        }
    }
}

#[test]
fn unknown_terminal() {
    check_err(
        r#"terminal `"\+"` does not have a pattern defined for it"#,
        r#"grammar; extern { enum Term { } } X = X "+";"#,
        r#"                                        ~~~ "#,
    );
}

#[test]
fn unknown_id_terminal() {
    check_err(
        r#"terminal `"foo"` does not have a pattern defined for it"#,
        r#"grammar; extern { enum Term { } } X = X "foo";"#,
        r#"                                        ~~~~~ "#,
    );
}

#[test]
fn tick_input_lifetime_already_declared() {
    check_err(
        r#".*the `'input` lifetime is implicit and cannot be declared"#,
        r#"grammar<'input>; X = X "foo";"#,
        r#"~~~~~~~                      "#,
    );
}

#[test]
fn input_parameter_already_declared() {
    check_err(
        r#".*the `input` parameter is implicit and cannot be declared"#,
        r#"grammar(input:u32); X = X "foo";"#,
        r#"~~~~~~~                         "#,
    );
}

#[test]
fn invalid_regular_expression_unterminated_group() {
    check_err(
        r#"unclosed group"#,
        r#"grammar; X = X r"(123";"#,
        r#"               ~~~~~~~ "#,
    );
}

#[test]
fn quoted_literals() {
    check_intern_token(
        r#"grammar; X = X "+" "-" "foo" "(" ")";"#,
        vec![
            ("+", r#"Some(("+", "+"))"#),
            ("-", r#"Some(("-", "-"))"#),
            ("(", r#"Some(("(", "("))"#),
            (")", r#"Some((")", ")"))"#),
            ("foo", r#"Some(("foo", "foo"))"#),
            ("<", r#"None"#),
        ],
    );
}

#[test]
fn regex_literals() {
    check_intern_token(
        r#"grammar; X = X r"[a-z]+" r"[0-9]+";"#,
        vec![
            ("a", r##"Some((r#"[a-z]+"#, "a"))"##),
            ("def", r##"Some((r#"[a-z]+"#, "def"))"##),
            ("1", r##"Some((r#"[0-9]+"#, "1"))"##),
            ("9123456", r##"Some((r#"[0-9]+"#, "9123456"))"##),
        ],
    );
}

/// Basic test for match mappings.
#[test]
// This test requires regex's unicode case support
#[cfg_attr(not(feature = "unicode"), ignore)]
fn match_mappings() {
    check_intern_token(
        r#"grammar; match { r"(?i)begin" => "BEGIN" } else { "abc" => ALPHA } X = "BEGIN" ALPHA;"#,
        vec![
            ("BEGIN", r#"Some(("BEGIN", "BEGIN"))"#),
            ("begin", r#"Some(("BEGIN", "begin"))"#),
            ("abc", r#"Some((ALPHA, "abc"))"#),
        ],
    );
}

/// Match mappings, exercising precedence. Here the ID regex *would*
/// be ambiguous with the begin regex.
#[test]
// This test requires regex's unicode case support
#[cfg_attr(not(feature = "unicode"), ignore)]
fn match_precedence() {
    check_intern_token(
        r#"grammar; match { r"(?i)begin" => "BEGIN" } else { r"\w+" => ID } X = ();"#,
        vec![
            ("BEGIN", r#"Some(("BEGIN", "BEGIN"))"#),
            ("begin", r#"Some(("BEGIN", "begin"))"#),
            ("abc", r#"Some((ID, "abc"))"#),
        ],
    );
}

/// Test that, without a `catch-all`, using unrecognized literals is an error.
#[test]
fn invalid_match_literal() {
    check_err(
        r#"terminal `"foo"` does not have a match mapping defined for it"#,
        r#"grammar; match { r"(?i)begin" => "BEGIN" } X = "foo";"#,
        r#"                                               ~~~~~ "#,
    );
}

/// Test that, without a `catch-all`, using unrecognized literals is an error.
#[test]
fn invalid_match_regex_literal() {
    check_err(
        r##"terminal `r#"foo"#` does not have a match mapping defined for it"##,
        r#"grammar; match { r"(?i)begin" => "BEGIN" } X = r"foo";"#,
        r#"                                               ~~~~~~ "#,
    );
}

/// Test that, with a catch-all, the previous two examples work.
#[test]
// This test requires regex's unicode case support
#[cfg_attr(not(feature = "unicode"), ignore)]
fn match_catch_all() {
    let grammar = r#"grammar; match { r"(?i)begin" => "BEGIN", _ } X = { "foo", r"foo" };"#;
    assert!(validate_grammar(grammar).is_ok())
}

/// Test that a `catch-all` can be use in the first `match` arm.
/// Before the pull request to close [issue 325](https://github.com/lalrpop/lalrpop/issues/325),
/// the usage of the `catch-all` symbol was not allowed in the first arm of a `match` block.
#[test]
fn match_catch_all_in_first_arm() {
    let grammar = r#"
        grammar;
        match {
            r"[a-z]",
            _
        } else {
            r"[[:word:]]+"
        }
        pub Term = {
            Num,
            "(" <Term> ")",
            r"[[:word:]]+" => format!("Id({})", <>),
        };
        Num: String = r"[0-9]+" => <>.to_string();
"#;
    assert!(validate_grammar(grammar).is_ok());
    check_intern_token(
        grammar,
        vec![
            ("x", r##"Some((r#"[a-z]"#, "x"))"##),
            ("xy", r##"Some((r#"[[:word:]]+"#, "xy"))"##),
        ],
    );
}

#[test]
// This test requires regex's unicode case support
#[cfg_attr(not(feature = "unicode"), ignore)]
fn complex_match() {
    let grammar = r#"
        grammar;
        match {
            "abc"        => "ABC",
            r"(?i)begin" => BEGIN
        }

        pub Query: String = {
            "ABC" BEGIN => String::from("Success")
        };
"#;
    assert!(validate_grammar(grammar).is_ok())
}

/// Test that overlapping regular expressions are still forbidden within one level
/// of a match declaration.
#[test]
// This test requires regex's unicode case support
#[cfg_attr(not(feature = "unicode"), ignore)]
fn ambiguity_within_match() {
    check_err(
        r##"ambiguity detected between the terminal `r#"b"#` and the terminal `r#"\(\?i\)b"#`"##,
        r#"grammar; match { r"(?i)b" => "B", r"b" => "b" }"#,
        r#"                                  ~~~~~~~~~~~~ "#,
    );
}

/// Test that using the **exact same regular expression** twice is
/// forbidden, even across multiple levels of the match expression.
/// No good reason to do that.
#[test]
fn same_literal_twice() {
    check_err(
        r##"multiple match entries for `r#"\[bB\]"#`"##,
        r#"grammar; match { r"[bB]" => "B" } else { r"[bB]" => "b" }"#,
        r#"                                         ~~~~~~~~~~~~~~~ "#,
    );
}