Rename and invert sense of Mode predicates.

I find it easier if they describe what's allowed, rather than what's
forbidden. Also, consistent naming makes them easier to understand.
This commit is contained in:
Nicholas Nethercote 2024-01-23 14:25:34 +11:00
parent a1c07214f0
commit 5e5aa6d556

View File

@ -195,29 +195,29 @@ pub fn in_double_quotes(self) -> bool {
} }
} }
/// Non-byte literals should have `\xXX` escapes that are within the ASCII range. /// Are `\x80`..`\xff` allowed?
fn ascii_escapes_should_be_ascii(self) -> bool { fn allow_high_bytes(self) -> bool {
match self { match self {
Char | Str => true, Char | Str => false,
Byte | ByteStr | CStr => false, Byte | ByteStr | CStr => true,
RawStr | RawByteStr | RawCStr => unreachable!(), RawStr | RawByteStr | RawCStr => unreachable!(),
} }
} }
/// Whether characters within the literal must be within the ASCII range. /// Are unicode (non-ASCII) chars allowed?
#[inline] #[inline]
fn chars_should_be_ascii(self) -> bool { fn allow_unicode_chars(self) -> bool {
match self { match self {
Byte | ByteStr | RawByteStr => true, Byte | ByteStr | RawByteStr => false,
Char | Str | RawStr | CStr | RawCStr => false, Char | Str | RawStr | CStr | RawCStr => true,
} }
} }
/// Byte literals do not allow unicode escape. /// Are unicode escapes (`\u`) allowed?
fn is_unicode_escape_disallowed(self) -> bool { fn allow_unicode_escapes(self) -> bool {
match self { match self {
Byte | ByteStr => true, Byte | ByteStr => false,
Char | Str | CStr => false, Char | Str | CStr => true,
RawByteStr | RawStr | RawCStr => unreachable!(), RawByteStr | RawStr | RawCStr => unreachable!(),
} }
} }
@ -255,25 +255,21 @@ fn scan_escape<T: From<char> + From<u8>>(
let value = (hi * 16 + lo) as u8; let value = (hi * 16 + lo) as u8;
return if mode.ascii_escapes_should_be_ascii() && !value.is_ascii() { return if !mode.allow_high_bytes() && !value.is_ascii() {
Err(EscapeError::OutOfRangeHexEscape) Err(EscapeError::OutOfRangeHexEscape)
} else { } else {
// This may be a high byte, but that will only happen if `T` is // This may be a high byte, but that will only happen if `T` is
// `MixedUnit`, because of the `ascii_escapes_should_be_ascii` // `MixedUnit`, because of the `allow_high_bytes` check above.
// check above.
Ok(T::from(value as u8)) Ok(T::from(value as u8))
}; };
} }
'u' => return scan_unicode(chars, mode.is_unicode_escape_disallowed()).map(T::from), 'u' => return scan_unicode(chars, mode.allow_unicode_escapes()).map(T::from),
_ => return Err(EscapeError::InvalidEscape), _ => return Err(EscapeError::InvalidEscape),
}; };
Ok(T::from(res)) Ok(T::from(res))
} }
fn scan_unicode( fn scan_unicode(chars: &mut Chars<'_>, allow_unicode_escapes: bool) -> Result<char, EscapeError> {
chars: &mut Chars<'_>,
is_unicode_escape_disallowed: bool,
) -> Result<char, EscapeError> {
// We've parsed '\u', now we have to parse '{..}'. // We've parsed '\u', now we have to parse '{..}'.
if chars.next() != Some('{') { if chars.next() != Some('{') {
@ -301,7 +297,7 @@ fn scan_unicode(
// Incorrect syntax has higher priority for error reporting // Incorrect syntax has higher priority for error reporting
// than unallowed value for a literal. // than unallowed value for a literal.
if is_unicode_escape_disallowed { if !allow_unicode_escapes {
return Err(EscapeError::UnicodeEscapeInByte); return Err(EscapeError::UnicodeEscapeInByte);
} }
@ -327,12 +323,8 @@ fn scan_unicode(
} }
#[inline] #[inline]
fn ascii_check(c: char, chars_should_be_ascii: bool) -> Result<char, EscapeError> { fn ascii_check(c: char, allow_unicode_chars: bool) -> Result<char, EscapeError> {
if chars_should_be_ascii && !c.is_ascii() { if allow_unicode_chars || c.is_ascii() { Ok(c) } else { Err(EscapeError::NonAsciiCharInByte) }
Err(EscapeError::NonAsciiCharInByte)
} else {
Ok(c)
}
} }
fn unescape_char_or_byte(chars: &mut Chars<'_>, mode: Mode) -> Result<char, EscapeError> { fn unescape_char_or_byte(chars: &mut Chars<'_>, mode: Mode) -> Result<char, EscapeError> {
@ -341,7 +333,7 @@ fn unescape_char_or_byte(chars: &mut Chars<'_>, mode: Mode) -> Result<char, Esca
'\\' => scan_escape(chars, mode), '\\' => scan_escape(chars, mode),
'\n' | '\t' | '\'' => Err(EscapeError::EscapeOnlyChar), '\n' | '\t' | '\'' => Err(EscapeError::EscapeOnlyChar),
'\r' => Err(EscapeError::BareCarriageReturn), '\r' => Err(EscapeError::BareCarriageReturn),
_ => ascii_check(c, mode.chars_should_be_ascii()), _ => ascii_check(c, mode.allow_unicode_chars()),
}?; }?;
if chars.next().is_some() { if chars.next().is_some() {
return Err(EscapeError::MoreThanOneChar); return Err(EscapeError::MoreThanOneChar);
@ -356,7 +348,7 @@ fn unescape_non_raw_common<F, T: From<char> + From<u8>>(src: &str, mode: Mode, c
F: FnMut(Range<usize>, Result<T, EscapeError>), F: FnMut(Range<usize>, Result<T, EscapeError>),
{ {
let mut chars = src.chars(); let mut chars = src.chars();
let chars_should_be_ascii = mode.chars_should_be_ascii(); // get this outside the loop let allow_unicode_chars = mode.allow_unicode_chars(); // get this outside the loop
// The `start` and `end` computation here is complicated because // The `start` and `end` computation here is complicated because
// `skip_ascii_whitespace` makes us to skip over chars without counting // `skip_ascii_whitespace` makes us to skip over chars without counting
@ -381,7 +373,7 @@ fn unescape_non_raw_common<F, T: From<char> + From<u8>>(src: &str, mode: Mode, c
} }
'"' => Err(EscapeError::EscapeOnlyChar), '"' => Err(EscapeError::EscapeOnlyChar),
'\r' => Err(EscapeError::BareCarriageReturn), '\r' => Err(EscapeError::BareCarriageReturn),
_ => ascii_check(c, chars_should_be_ascii).map(T::from), _ => ascii_check(c, allow_unicode_chars).map(T::from),
}; };
let end = src.len() - chars.as_str().len(); let end = src.len() - chars.as_str().len();
callback(start..end, res); callback(start..end, res);
@ -423,7 +415,7 @@ fn check_raw_common<F>(src: &str, mode: Mode, callback: &mut F)
F: FnMut(Range<usize>, Result<char, EscapeError>), F: FnMut(Range<usize>, Result<char, EscapeError>),
{ {
let mut chars = src.chars(); let mut chars = src.chars();
let chars_should_be_ascii = mode.chars_should_be_ascii(); // get this outside the loop let allow_unicode_chars = mode.allow_unicode_chars(); // get this outside the loop
// The `start` and `end` computation here matches the one in // The `start` and `end` computation here matches the one in
// `unescape_non_raw_common` for consistency, even though this function // `unescape_non_raw_common` for consistency, even though this function
@ -432,7 +424,7 @@ fn check_raw_common<F>(src: &str, mode: Mode, callback: &mut F)
let start = src.len() - chars.as_str().len() - c.len_utf8(); let start = src.len() - chars.as_str().len() - c.len_utf8();
let res = match c { let res = match c {
'\r' => Err(EscapeError::BareCarriageReturnInRawString), '\r' => Err(EscapeError::BareCarriageReturnInRawString),
_ => ascii_check(c, chars_should_be_ascii), _ => ascii_check(c, allow_unicode_chars),
}; };
let end = src.len() - chars.as_str().len(); let end = src.len() - chars.as_str().len();
callback(start..end, res); callback(start..end, res);