summaryrefslogtreecommitdiffstats
path: root/compiler/rustc_lexer/src/unescape.rs
blob: 8507ca9d89ed776b77228fec41521ab835ce75e3 (plain)
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
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
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
//! Utilities for validating string and char literals and turning them into
//! values they represent.

use std::ops::Range;
use std::str::Chars;

#[cfg(test)]
mod tests;

/// Errors and warnings that can occur during string unescaping.
#[derive(Debug, PartialEq, Eq)]
pub enum EscapeError {
    /// Expected 1 char, but 0 were found.
    ZeroChars,
    /// Expected 1 char, but more than 1 were found.
    MoreThanOneChar,

    /// Escaped '\' character without continuation.
    LoneSlash,
    /// Invalid escape character (e.g. '\z').
    InvalidEscape,
    /// Raw '\r' encountered.
    BareCarriageReturn,
    /// Raw '\r' encountered in raw string.
    BareCarriageReturnInRawString,
    /// Unescaped character that was expected to be escaped (e.g. raw '\t').
    EscapeOnlyChar,

    /// Numeric character escape is too short (e.g. '\x1').
    TooShortHexEscape,
    /// Invalid character in numeric escape (e.g. '\xz')
    InvalidCharInHexEscape,
    /// Character code in numeric escape is non-ascii (e.g. '\xFF').
    OutOfRangeHexEscape,

    /// '\u' not followed by '{'.
    NoBraceInUnicodeEscape,
    /// Non-hexadecimal value in '\u{..}'.
    InvalidCharInUnicodeEscape,
    /// '\u{}'
    EmptyUnicodeEscape,
    /// No closing brace in '\u{..}', e.g. '\u{12'.
    UnclosedUnicodeEscape,
    /// '\u{_12}'
    LeadingUnderscoreUnicodeEscape,
    /// More than 6 characters in '\u{..}', e.g. '\u{10FFFF_FF}'
    OverlongUnicodeEscape,
    /// Invalid in-bound unicode character code, e.g. '\u{DFFF}'.
    LoneSurrogateUnicodeEscape,
    /// Out of bounds unicode character code, e.g. '\u{FFFFFF}'.
    OutOfRangeUnicodeEscape,

    /// Unicode escape code in byte literal.
    UnicodeEscapeInByte,
    /// Non-ascii character in byte literal, byte string literal, or raw byte string literal.
    NonAsciiCharInByte,

    /// After a line ending with '\', the next line contains whitespace
    /// characters that are not skipped.
    UnskippedWhitespaceWarning,

    /// After a line ending with '\', multiple lines are skipped.
    MultipleSkippedLinesWarning,
}

impl EscapeError {
    /// Returns true for actual errors, as opposed to warnings.
    pub fn is_fatal(&self) -> bool {
        !matches!(
            self,
            EscapeError::UnskippedWhitespaceWarning | EscapeError::MultipleSkippedLinesWarning
        )
    }
}

/// Takes a contents of a literal (without quotes) and produces a
/// sequence of escaped characters or errors.
/// Values are returned through invoking of the provided callback.
pub fn unescape_literal<F>(src: &str, mode: Mode, callback: &mut F)
where
    F: FnMut(Range<usize>, Result<char, EscapeError>),
{
    match mode {
        Mode::Char | Mode::Byte => {
            let mut chars = src.chars();
            let res = unescape_char_or_byte(&mut chars, mode == Mode::Byte);
            callback(0..(src.len() - chars.as_str().len()), res);
        }
        Mode::Str | Mode::ByteStr => unescape_str_or_byte_str(src, mode == Mode::ByteStr, callback),
        Mode::RawStr | Mode::RawByteStr => {
            unescape_raw_str_or_raw_byte_str(src, mode == Mode::RawByteStr, callback)
        }
    }
}

/// Takes a contents of a char literal (without quotes), and returns an
/// unescaped char or an error.
pub fn unescape_char(src: &str) -> Result<char, EscapeError> {
    unescape_char_or_byte(&mut src.chars(), false)
}

/// Takes a contents of a byte literal (without quotes), and returns an
/// unescaped byte or an error.
pub fn unescape_byte(src: &str) -> Result<u8, EscapeError> {
    unescape_char_or_byte(&mut src.chars(), true).map(byte_from_char)
}

/// What kind of literal do we parse.
#[derive(Debug, Clone, Copy, PartialEq)]
pub enum Mode {
    Char,
    Str,
    Byte,
    ByteStr,
    RawStr,
    RawByteStr,
}

impl Mode {
    pub fn in_double_quotes(self) -> bool {
        match self {
            Mode::Str | Mode::ByteStr | Mode::RawStr | Mode::RawByteStr => true,
            Mode::Char | Mode::Byte => false,
        }
    }

    pub fn is_byte(self) -> bool {
        match self {
            Mode::Byte | Mode::ByteStr | Mode::RawByteStr => true,
            Mode::Char | Mode::Str | Mode::RawStr => false,
        }
    }
}

fn scan_escape(chars: &mut Chars<'_>, is_byte: bool) -> Result<char, EscapeError> {
    // Previous character was '\\', unescape what follows.
    let res = match chars.next().ok_or(EscapeError::LoneSlash)? {
        '"' => '"',
        'n' => '\n',
        'r' => '\r',
        't' => '\t',
        '\\' => '\\',
        '\'' => '\'',
        '0' => '\0',

        'x' => {
            // Parse hexadecimal character code.

            let hi = chars.next().ok_or(EscapeError::TooShortHexEscape)?;
            let hi = hi.to_digit(16).ok_or(EscapeError::InvalidCharInHexEscape)?;

            let lo = chars.next().ok_or(EscapeError::TooShortHexEscape)?;
            let lo = lo.to_digit(16).ok_or(EscapeError::InvalidCharInHexEscape)?;

            let value = hi * 16 + lo;

            // For a non-byte literal verify that it is within ASCII range.
            if !is_byte && !is_ascii(value) {
                return Err(EscapeError::OutOfRangeHexEscape);
            }
            let value = value as u8;

            value as char
        }

        'u' => {
            // We've parsed '\u', now we have to parse '{..}'.

            if chars.next() != Some('{') {
                return Err(EscapeError::NoBraceInUnicodeEscape);
            }

            // First character must be a hexadecimal digit.
            let mut n_digits = 1;
            let mut value: u32 = match chars.next().ok_or(EscapeError::UnclosedUnicodeEscape)? {
                '_' => return Err(EscapeError::LeadingUnderscoreUnicodeEscape),
                '}' => return Err(EscapeError::EmptyUnicodeEscape),
                c => c.to_digit(16).ok_or(EscapeError::InvalidCharInUnicodeEscape)?,
            };

            // First character is valid, now parse the rest of the number
            // and closing brace.
            loop {
                match chars.next() {
                    None => return Err(EscapeError::UnclosedUnicodeEscape),
                    Some('_') => continue,
                    Some('}') => {
                        if n_digits > 6 {
                            return Err(EscapeError::OverlongUnicodeEscape);
                        }

                        // Incorrect syntax has higher priority for error reporting
                        // than unallowed value for a literal.
                        if is_byte {
                            return Err(EscapeError::UnicodeEscapeInByte);
                        }

                        break std::char::from_u32(value).ok_or_else(|| {
                            if value > 0x10FFFF {
                                EscapeError::OutOfRangeUnicodeEscape
                            } else {
                                EscapeError::LoneSurrogateUnicodeEscape
                            }
                        })?;
                    }
                    Some(c) => {
                        let digit: u32 =
                            c.to_digit(16).ok_or(EscapeError::InvalidCharInUnicodeEscape)?;
                        n_digits += 1;
                        if n_digits > 6 {
                            // Stop updating value since we're sure that it's incorrect already.
                            continue;
                        }
                        value = value * 16 + digit;
                    }
                };
            }
        }
        _ => return Err(EscapeError::InvalidEscape),
    };
    Ok(res)
}

#[inline]
fn ascii_check(c: char, is_byte: bool) -> Result<char, EscapeError> {
    if is_byte && !c.is_ascii() {
        // Byte literal can't be a non-ascii character.
        Err(EscapeError::NonAsciiCharInByte)
    } else {
        Ok(c)
    }
}

fn unescape_char_or_byte(chars: &mut Chars<'_>, is_byte: bool) -> Result<char, EscapeError> {
    let c = chars.next().ok_or(EscapeError::ZeroChars)?;
    let res = match c {
        '\\' => scan_escape(chars, is_byte),
        '\n' | '\t' | '\'' => Err(EscapeError::EscapeOnlyChar),
        '\r' => Err(EscapeError::BareCarriageReturn),
        _ => ascii_check(c, is_byte),
    }?;
    if chars.next().is_some() {
        return Err(EscapeError::MoreThanOneChar);
    }
    Ok(res)
}

/// Takes a contents of a string literal (without quotes) and produces a
/// sequence of escaped characters or errors.
fn unescape_str_or_byte_str<F>(src: &str, is_byte: bool, callback: &mut F)
where
    F: FnMut(Range<usize>, Result<char, EscapeError>),
{
    let mut chars = src.chars();

    // The `start` and `end` computation here is complicated because
    // `skip_ascii_whitespace` makes us to skip over chars without counting
    // them in the range computation.
    while let Some(c) = chars.next() {
        let start = src.len() - chars.as_str().len() - c.len_utf8();
        let res = match c {
            '\\' => {
                match chars.clone().next() {
                    Some('\n') => {
                        // Rust language specification requires us to skip whitespaces
                        // if unescaped '\' character is followed by '\n'.
                        // For details see [Rust language reference]
                        // (https://doc.rust-lang.org/reference/tokens.html#string-literals).
                        skip_ascii_whitespace(&mut chars, start, callback);
                        continue;
                    }
                    _ => scan_escape(&mut chars, is_byte),
                }
            }
            '\n' => Ok('\n'),
            '\t' => Ok('\t'),
            '"' => Err(EscapeError::EscapeOnlyChar),
            '\r' => Err(EscapeError::BareCarriageReturn),
            _ => ascii_check(c, is_byte),
        };
        let end = src.len() - chars.as_str().len();
        callback(start..end, res);
    }

    fn skip_ascii_whitespace<F>(chars: &mut Chars<'_>, start: usize, callback: &mut F)
    where
        F: FnMut(Range<usize>, Result<char, EscapeError>),
    {
        let tail = chars.as_str();
        let first_non_space = tail
            .bytes()
            .position(|b| b != b' ' && b != b'\t' && b != b'\n' && b != b'\r')
            .unwrap_or(tail.len());
        if tail[1..first_non_space].contains('\n') {
            // The +1 accounts for the escaping slash.
            let end = start + first_non_space + 1;
            callback(start..end, Err(EscapeError::MultipleSkippedLinesWarning));
        }
        let tail = &tail[first_non_space..];
        if let Some(c) = tail.chars().nth(0) {
            // For error reporting, we would like the span to contain the character that was not
            // skipped. The +1 is necessary to account for the leading \ that started the escape.
            let end = start + first_non_space + c.len_utf8() + 1;
            if c.is_whitespace() {
                callback(start..end, Err(EscapeError::UnskippedWhitespaceWarning));
            }
        }
        *chars = tail.chars();
    }
}

/// Takes a contents of a string literal (without quotes) and produces a
/// sequence of characters or errors.
/// NOTE: Raw strings do not perform any explicit character escaping, here we
/// only produce errors on bare CR.
fn unescape_raw_str_or_raw_byte_str<F>(src: &str, is_byte: bool, callback: &mut F)
where
    F: FnMut(Range<usize>, Result<char, EscapeError>),
{
    let mut chars = src.chars();

    // The `start` and `end` computation here matches the one in
    // `unescape_str_or_byte_str` for consistency, even though this function
    // doesn't have to worry about skipping any chars.
    while let Some(c) = chars.next() {
        let start = src.len() - chars.as_str().len() - c.len_utf8();
        let res = match c {
            '\r' => Err(EscapeError::BareCarriageReturnInRawString),
            _ => ascii_check(c, is_byte),
        };
        let end = src.len() - chars.as_str().len();
        callback(start..end, res);
    }
}

#[inline]
pub fn byte_from_char(c: char) -> u8 {
    let res = c as u32;
    debug_assert!(res <= u8::MAX as u32, "guaranteed because of Mode::ByteStr");
    res as u8
}

fn is_ascii(x: u32) -> bool {
    x <= 0x7F
}