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
346
347
348
349
350
351
352
353
354
//! 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 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 characted (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 charactes 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.
    NonAsciiCharInByte,
    /// Non-ascii character in byte string literal.
    NonAsciiCharInByteString,
}

/// Takes a contents of a char literal (without quotes), and returns an
/// unescaped char or an error
pub fn unescape_char(literal_text: &str) -> Result<char, (usize, EscapeError)> {
    let mut chars = literal_text.chars();
    unescape_char_or_byte(&mut chars, Mode::Char)
        .map_err(|err| (literal_text.len() - chars.as_str().len(), err))
}

/// Takes a contents of a byte literal (without quotes), and returns an
/// unescaped byte or an error.
pub fn unescape_byte(literal_text: &str) -> Result<u8, (usize, EscapeError)> {
    let mut chars = literal_text.chars();
    unescape_char_or_byte(&mut chars, Mode::Byte)
        .map(byte_from_char)
        .map_err(|err| (literal_text.len() - chars.as_str().len(), err))
}

/// Takes a contents of a string literal (without quotes) and produces a
/// sequence of escaped characters or errors.
/// Values are returned through invoking of the provided callback.
pub fn unescape_str<F>(literal_text: &str, callback: &mut F)
where
    F: FnMut(Range<usize>, Result<char, EscapeError>),
{
    unescape_str_or_byte_str(literal_text, Mode::Str, callback)
}

/// Takes a contents of a byte string literal (without quotes) and produces a
/// sequence of bytes or errors.
/// Values are returned through invoking of the provided callback.
pub fn unescape_byte_str<F>(literal_text: &str, callback: &mut F)
where
    F: FnMut(Range<usize>, Result<u8, EscapeError>),
{
    unescape_str_or_byte_str(literal_text, Mode::ByteStr, &mut |range, char| {
        callback(range, char.map(byte_from_char))
    })
}

/// Takes a contents of a raw string literal (without quotes) and produces a
/// sequence of characters or errors.
/// Values are returned through invoking of the provided callback.
/// NOTE: Raw strings do not perform any explicit character escaping, here we
/// only translate CRLF to LF and produce errors on bare CR.
pub fn unescape_raw_str<F>(literal_text: &str, callback: &mut F)
where
    F: FnMut(Range<usize>, Result<char, EscapeError>),
{
    unescape_raw_str_or_byte_str(literal_text, Mode::Str, callback)
}

/// Takes a contents of a raw byte string literal (without quotes) and produces a
/// sequence of bytes or errors.
/// Values are returned through invoking of the provided callback.
/// NOTE: Raw strings do not perform any explicit character escaping, here we
/// only translate CRLF to LF and produce errors on bare CR.
pub fn unescape_raw_byte_str<F>(literal_text: &str, callback: &mut F)
where
    F: FnMut(Range<usize>, Result<u8, EscapeError>),
{
    unescape_raw_str_or_byte_str(literal_text, Mode::ByteStr, &mut |range, char| {
        callback(range, char.map(byte_from_char))
    })
}

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

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

    pub fn in_double_quotes(self) -> bool {
        !self.in_single_quotes()
    }

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

fn scan_escape(first_char: char, chars: &mut Chars<'_>, mode: Mode) -> Result<char, EscapeError> {
    if first_char != '\\' {
        // Previous character was not a slash, and we don't expect it to be
        // an escape-only character.
        return match first_char {
            '\t' | '\n' => Err(EscapeError::EscapeOnlyChar),
            '\r' => Err(EscapeError::BareCarriageReturn),
            '\'' if mode.in_single_quotes() => Err(EscapeError::EscapeOnlyChar),
            '"' if mode.in_double_quotes() => Err(EscapeError::EscapeOnlyChar),
            _ => {
                if mode.is_bytes() && !first_char.is_ascii() {
                    // Byte literal can't be a non-ascii character.
                    return Err(EscapeError::NonAsciiCharInByte);
                }
                Ok(first_char)
            }
        };
    }

    // Previous character is '\\', try to unescape it.

    let second_char = chars.next().ok_or(EscapeError::LoneSlash)?;

    let res = match second_char {
        '"' => '"',
        '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 byte literal verify that it is within ASCII range.
            if !mode.is_bytes() && !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 characrer 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 mode.is_bytes() {
                            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 =
                            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 is incorrect already.
                            continue;
                        }
                        let digit = digit as u32;
                        value = value * 16 + digit;
                    }
                };
            }
        }
        _ => return Err(EscapeError::InvalidEscape),
    };
    Ok(res)
}

fn unescape_char_or_byte(chars: &mut Chars<'_>, mode: Mode) -> Result<char, EscapeError> {
    let first_char = chars.next().ok_or(EscapeError::ZeroChars)?;
    let res = scan_escape(first_char, chars, mode)?;
    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, mode: Mode, callback: &mut F)
where
    F: FnMut(Range<usize>, Result<char, EscapeError>),
{
    assert!(mode.in_double_quotes());
    let initial_len = src.len();
    let mut chars = src.chars();
    while let Some(first_char) = chars.next() {
        let start = initial_len - chars.as_str().len() - first_char.len_utf8();

        let unescaped_char = match first_char {
            '\\' => {
                let second_char = chars.clone().next();
                match second_char {
                    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);
                        continue;
                    }
                    _ => scan_escape(first_char, &mut chars, mode),
                }
            }
            '\n' => Ok('\n'),
            '\t' => Ok('\t'),
            _ => scan_escape(first_char, &mut chars, mode),
        };
        let end = initial_len - chars.as_str().len();
        callback(start..end, unescaped_char);
    }

    fn skip_ascii_whitespace(chars: &mut Chars<'_>) {
        let str = chars.as_str();
        let first_non_space = str
            .bytes()
            .position(|b| b != b' ' && b != b'\t' && b != b'\n' && b != b'\r')
            .unwrap_or(str.len());
        *chars = str[first_non_space..].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 translate CRLF to LF and produce errors on bare CR.
fn unescape_raw_str_or_byte_str<F>(literal_text: &str, mode: Mode, callback: &mut F)
where
    F: FnMut(Range<usize>, Result<char, EscapeError>),
{
    assert!(mode.in_double_quotes());
    let initial_len = literal_text.len();

    let mut chars = literal_text.chars();
    while let Some(curr) = chars.next() {
        let start = initial_len - chars.as_str().len() - curr.len_utf8();

        let result = match curr {
            '\r' => Err(EscapeError::BareCarriageReturnInRawString),
            c if mode.is_bytes() && !c.is_ascii() => Err(EscapeError::NonAsciiCharInByteString),
            c => Ok(c),
        };
        let end = initial_len - chars.as_str().len();

        callback(start..end, result);
    }
}

fn byte_from_char(c: char) -> u8 {
    let res = c as u32;
    assert!(res <= u8::max_value() as u32, "guaranteed because of Mode::Byte(Str)");
    res as u8
}

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