Struct cranelift_isle::lexer::Lexer

source · 
pub struct Lexer<'a> {
    pub filenames: Vec<Arc<str>>,
    pub file_texts: Vec<Arc<str>>,
    /* private fields */
}
Expand description

The lexer.

Breaks source text up into a sequence of tokens (with source positions).

Fields§

§filenames: Vec<Arc<str>>

Arena of filenames from the input source.

Indexed via Pos::file.

§file_texts: Vec<Arc<str>>

Arena of file source texts.

Indexed via Pos::file.

Implementations§

Create a new lexer for the given source contents and filename.

Create a new lexer from the given files.

Examples found in repository?
src/compile.rs (line 22)
18
19
20
21
22
23
24
25

pub fn from_files<P: AsRef<Path>>(
    inputs: impl IntoIterator<Item = P>,
    options: &codegen::CodegenOptions,
) -> Result<String, Errors> {
    let lexer = crate::lexer::Lexer::from_files(inputs)?;
    let defs = crate::parser::parse(lexer)?;
    compile(&defs, options)
}

Get the lexer’s current source position.

Examples found in repository?
src/parser.rs (line 55)
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71

    fn take<F: Fn(&Token) -> bool>(&mut self, f: F) -> Result<Token> {
        if let Some(&(pos, ref peek)) = self.lexer.peek() {
            if !f(peek) {
                return Err(self.error(pos, format!("Unexpected token {:?}", peek)));
            }
            Ok(self.lexer.next()?.unwrap().1)
        } else {
            Err(self.error(self.lexer.pos(), "Unexpected EOF".to_string()))
        }
    }

    fn is<F: Fn(&Token) -> bool>(&self, f: F) -> bool {
        if let Some(&(_, ref peek)) = self.lexer.peek() {
            f(peek)
        } else {
            false
        }
    }

    fn pos(&self) -> Pos {
        self.lexer
            .peek()
            .map_or_else(|| self.lexer.pos(), |(pos, _)| *pos)
    }
More examples
Hide additional examples
src/lexer.rs (line 207)
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
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297

    fn next_token(&mut self) -> Result<Option<(Pos, Token)>> {
        fn is_sym_first_char(c: u8) -> bool {
            match c {
                b'-' | b'0'..=b'9' | b'(' | b')' | b';' | b'<' | b'>' => false,
                c if c.is_ascii_whitespace() => false,
                _ => true,
            }
        }
        fn is_sym_other_char(c: u8) -> bool {
            match c {
                b'(' | b')' | b';' | b'@' | b'<' => false,
                c if c.is_ascii_whitespace() => false,
                _ => true,
            }
        }

        // Skip any whitespace and any comments.
        while self.pos.offset < self.buf.len() {
            if self.buf[self.pos.offset].is_ascii_whitespace() {
                self.advance_pos();
                continue;
            }
            if self.buf[self.pos.offset] == b';' {
                while self.pos.offset < self.buf.len() && self.buf[self.pos.offset] != b'\n' {
                    self.advance_pos();
                }
                continue;
            }
            break;
        }

        if self.pos.offset == self.buf.len() {
            return Ok(None);
        }

        let char_pos = self.pos();
        match self.buf[self.pos.offset] {
            b'(' => {
                self.advance_pos();
                Ok(Some((char_pos, Token::LParen)))
            }
            b')' => {
                self.advance_pos();
                Ok(Some((char_pos, Token::RParen)))
            }
            b'@' => {
                self.advance_pos();
                Ok(Some((char_pos, Token::At)))
            }
            c if is_sym_first_char(c) => {
                let start = self.pos.offset;
                let start_pos = self.pos();
                while self.pos.offset < self.buf.len()
                    && is_sym_other_char(self.buf[self.pos.offset])
                {
                    self.advance_pos();
                }
                let end = self.pos.offset;
                let s = std::str::from_utf8(&self.buf[start..end])
                    .expect("Only ASCII characters, should be UTF-8");
                debug_assert!(!s.is_empty());
                Ok(Some((start_pos, Token::Symbol(s.to_string()))))
            }
            c if (c >= b'0' && c <= b'9') || c == b'-' => {
                let start_pos = self.pos();
                let neg = if c == b'-' {
                    self.advance_pos();
                    true
                } else {
                    false
                };

                let mut radix = 10;

                // Check for hex literals.
                if self.buf.get(self.pos.offset).copied() == Some(b'0')
                    && (self.buf.get(self.pos.offset + 1).copied() == Some(b'x')
                        || self.buf.get(self.pos.offset + 1).copied() == Some(b'X'))
                {
                    self.advance_pos();
                    self.advance_pos();
                    radix = 16;
                }

                // Find the range in the buffer for this integer literal. We'll
                // pass this range to `i64::from_str_radix` to do the actual
                // string-to-integer conversion.
                let mut s = vec![];
                while self.pos.offset < self.buf.len()
                    && ((radix == 10
                        && self.buf[self.pos.offset] >= b'0'
                        && self.buf[self.pos.offset] <= b'9')
                        || (radix == 16
                            && ((self.buf[self.pos.offset] >= b'0'
                                && self.buf[self.pos.offset] <= b'9')
                                || (self.buf[self.pos.offset] >= b'a'
                                    && self.buf[self.pos.offset] <= b'f')
                                || (self.buf[self.pos.offset] >= b'A'
                                    && self.buf[self.pos.offset] <= b'F')))
                        || self.buf[self.pos.offset] == b'_')
                {
                    if self.buf[self.pos.offset] != b'_' {
                        s.push(self.buf[self.pos.offset]);
                    }
                    self.advance_pos();
                }
                let s_utf8 = std::str::from_utf8(&s[..]).unwrap();

                // Support either signed range (-2^127..2^127) or
                // unsigned range (0..2^128).
                let num = i128::from_str_radix(s_utf8, radix)
                    .or_else(|_| u128::from_str_radix(s_utf8, radix).map(|val| val as i128))
                    .map_err(|e| self.error(start_pos, e.to_string()))?;

                let tok = if neg {
                    Token::Int(num.checked_neg().ok_or_else(|| {
                        self.error(start_pos, "integer literal cannot fit in i128")
                    })?)
                } else {
                    Token::Int(num)
                };
                Ok(Some((start_pos, tok)))
            }
            c => Err(self.error(self.pos, format!("Unexpected character '{}'", c))),
        }
    }

Get the next token from this lexer’s token stream, if any.

Examples found in repository?
src/parser.rs (line 53)
48
49
50
51
52
53
54
55
56
57

    fn take<F: Fn(&Token) -> bool>(&mut self, f: F) -> Result<Token> {
        if let Some(&(pos, ref peek)) = self.lexer.peek() {
            if !f(peek) {
                return Err(self.error(pos, format!("Unexpected token {:?}", peek)));
            }
            Ok(self.lexer.next()?.unwrap().1)
        } else {
            Err(self.error(self.lexer.pos(), "Unexpected EOF".to_string()))
        }
    }

Peek ahead at the next token.

Examples found in repository?
src/parser.rs (line 49)
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

    fn take<F: Fn(&Token) -> bool>(&mut self, f: F) -> Result<Token> {
        if let Some(&(pos, ref peek)) = self.lexer.peek() {
            if !f(peek) {
                return Err(self.error(pos, format!("Unexpected token {:?}", peek)));
            }
            Ok(self.lexer.next()?.unwrap().1)
        } else {
            Err(self.error(self.lexer.pos(), "Unexpected EOF".to_string()))
        }
    }

    fn is<F: Fn(&Token) -> bool>(&self, f: F) -> bool {
        if let Some(&(_, ref peek)) = self.lexer.peek() {
            f(peek)
        } else {
            false
        }
    }

    fn pos(&self) -> Pos {
        self.lexer
            .peek()
            .map_or_else(|| self.lexer.pos(), |(pos, _)| *pos)
    }

    fn is_lparen(&self) -> bool {
        self.is(|tok| *tok == Token::LParen)
    }
    fn is_rparen(&self) -> bool {
        self.is(|tok| *tok == Token::RParen)
    }
    fn is_at(&self) -> bool {
        self.is(|tok| *tok == Token::At)
    }
    fn is_sym(&self) -> bool {
        self.is(|tok| tok.is_sym())
    }
    fn is_int(&self) -> bool {
        self.is(|tok| tok.is_int())
    }
    fn is_sym_str(&self, s: &str) -> bool {
        self.is(|tok| match tok {
            &Token::Symbol(ref tok_s) if tok_s == s => true,
            _ => false,
        })
    }

    fn is_const(&self) -> bool {
        self.is(|tok| match tok {
            &Token::Symbol(ref tok_s) if tok_s.starts_with("$") => true,
            _ => false,
        })
    }

    fn lparen(&mut self) -> Result<()> {
        self.take(|tok| *tok == Token::LParen).map(|_| ())
    }
    fn rparen(&mut self) -> Result<()> {
        self.take(|tok| *tok == Token::RParen).map(|_| ())
    }
    fn at(&mut self) -> Result<()> {
        self.take(|tok| *tok == Token::At).map(|_| ())
    }

    fn symbol(&mut self) -> Result<String> {
        match self.take(|tok| tok.is_sym())? {
            Token::Symbol(s) => Ok(s),
            _ => unreachable!(),
        }
    }

    fn int(&mut self) -> Result<i128> {
        match self.take(|tok| tok.is_int())? {
            Token::Int(i) => Ok(i),
            _ => unreachable!(),
        }
    }

    fn parse_defs(mut self) -> Result<Defs> {
        let mut defs = vec![];
        while !self.lexer.eof() {
            defs.push(self.parse_def()?);
        }
        Ok(Defs {
            defs,
            filenames: self.lexer.filenames,
            file_texts: self.lexer.file_texts,
        })
    }

    fn parse_def(&mut self) -> Result<Def> {
        self.lparen()?;
        let pos = self.pos();
        let def = match &self.symbol()?[..] {
            "pragma" => Def::Pragma(self.parse_pragma()?),
            "type" => Def::Type(self.parse_type()?),
            "decl" => Def::Decl(self.parse_decl()?),
            "rule" => Def::Rule(self.parse_rule()?),
            "extractor" => Def::Extractor(self.parse_etor()?),
            "extern" => Def::Extern(self.parse_extern()?),
            "convert" => Def::Converter(self.parse_converter()?),
            s => {
                return Err(self.error(pos, format!("Unexpected identifier: {}", s)));
            }
        };
        self.rparen()?;
        Ok(def)
    }

    fn str_to_ident(&self, pos: Pos, s: &str) -> Result<Ident> {
        let first = s
            .chars()
            .next()
            .ok_or_else(|| self.error(pos, "empty symbol".into()))?;
        if !first.is_alphabetic() && first != '_' && first != '$' {
            return Err(self.error(
                pos,
                format!("Identifier '{}' does not start with letter or _ or $", s),
            ));
        }
        if s.chars()
            .skip(1)
            .any(|c| !c.is_alphanumeric() && c != '_' && c != '.' && c != '$')
        {
            return Err(self.error(
                pos,
                format!(
                    "Identifier '{}' contains invalid character (not a-z, A-Z, 0-9, _, ., $)",
                    s
                ),
            ));
        }
        Ok(Ident(s.to_string(), pos))
    }

    fn parse_ident(&mut self) -> Result<Ident> {
        let pos = self.pos();
        let s = self.symbol()?;
        self.str_to_ident(pos, &s)
    }

    fn parse_const(&mut self) -> Result<Ident> {
        let pos = self.pos();
        let ident = self.parse_ident()?;
        if ident.0.starts_with("$") {
            let s = &ident.0[1..];
            Ok(Ident(s.to_string(), ident.1))
        } else {
            Err(self.error(
                pos,
                "Not a constant identifier; must start with a '$'".to_string(),
            ))
        }
    }

    fn parse_pragma(&mut self) -> Result<Pragma> {
        let ident = self.parse_ident()?;
        // currently, no pragmas are defined, but the infrastructure is useful to keep around
        match ident.0.as_str() {
            pragma => Err(self.error(ident.1, format!("Unknown pragma '{}'", pragma))),
        }
    }

    fn parse_type(&mut self) -> Result<Type> {
        let pos = self.pos();
        let name = self.parse_ident()?;

        let mut is_extern = false;
        let mut is_nodebug = false;

        while self.lexer.peek().map_or(false, |(_pos, tok)| tok.is_sym()) {
            let sym = self.symbol()?;
            if sym == "extern" {
                is_extern = true;
            } else if sym == "nodebug" {
                is_nodebug = true;
            } else {
                return Err(self.error(
                    self.pos(),
                    format!("unknown type declaration modifier: {}", sym),
                ));
            }
        }

        let ty = self.parse_typevalue()?;
        Ok(Type {
            name,
            is_extern,
            is_nodebug,
            ty,
            pos,
        })
    }

Are we at the end of the source input?

Examples found in repository?
src/parser.rs (line 128)
126
127
128
129
130
131
132
133
134
135
136

    fn parse_defs(mut self) -> Result<Defs> {
        let mut defs = vec![];
        while !self.lexer.eof() {
            defs.push(self.parse_def()?);
        }
        Ok(Defs {
            defs,
            filenames: self.lexer.filenames,
            file_texts: self.lexer.file_texts,
        })
    }

Trait Implementations§

Returns a copy of the value. Read more
Performs copy-assignment from source. Read more
Formats the value using the given formatter. Read more

Auto Trait Implementations§

Blanket Implementations§

Gets the TypeId of self. Read more
Immutably borrows from an owned value. Read more
Mutably borrows from an owned value. Read more

Returns the argument unchanged.

Calls U::from(self).

That is, this conversion is whatever the implementation of From<T> for U chooses to do.

The resulting type after obtaining ownership.
Creates owned data from borrowed data, usually by cloning. Read more
Uses borrowed data to replace owned data, usually by cloning. Read more
The type returned in the event of a conversion error.
Performs the conversion.
The type returned in the event of a conversion error.
Performs the conversion.