summaryrefslogtreecommitdiffstats
path: root/third_party/rust/litrs/src/char
diff options
context:
space:
mode:
Diffstat (limited to 'third_party/rust/litrs/src/char')
-rw-r--r--third_party/rust/litrs/src/char/mod.rs105
-rw-r--r--third_party/rust/litrs/src/char/tests.rs227
2 files changed, 332 insertions, 0 deletions
diff --git a/third_party/rust/litrs/src/char/mod.rs b/third_party/rust/litrs/src/char/mod.rs
new file mode 100644
index 0000000000..54f6f1137f
--- /dev/null
+++ b/third_party/rust/litrs/src/char/mod.rs
@@ -0,0 +1,105 @@
+use std::fmt;
+
+use crate::{
+ Buffer, ParseError,
+ err::{perr, ParseErrorKind::*},
+ escape::unescape,
+ parse::{first_byte_or_empty, check_suffix},
+};
+
+
+/// A character literal, e.g. `'g'` or `'🦊'`.
+///
+/// See [the reference][ref] for more information.
+///
+/// [ref]: https://doc.rust-lang.org/reference/tokens.html#character-literals
+#[derive(Debug, Clone, Copy, PartialEq, Eq)]
+pub struct CharLit<B: Buffer> {
+ raw: B,
+ /// Start index of the suffix or `raw.len()` if there is no suffix.
+ start_suffix: usize,
+ value: char,
+}
+
+impl<B: Buffer> CharLit<B> {
+ /// Parses the input as a character literal. Returns an error if the input
+ /// is invalid or represents a different kind of literal.
+ pub fn parse(input: B) -> Result<Self, ParseError> {
+ match first_byte_or_empty(&input)? {
+ b'\'' => {
+ let (value, start_suffix) = parse_impl(&input)?;
+ Ok(Self { raw: input, value, start_suffix })
+ },
+ _ => Err(perr(0, DoesNotStartWithQuote)),
+ }
+ }
+
+ /// Returns the character value that this literal represents.
+ pub fn value(&self) -> char {
+ self.value
+ }
+
+ /// The optional suffix. Returns `""` if the suffix is empty/does not exist.
+ pub fn suffix(&self) -> &str {
+ &(*self.raw)[self.start_suffix..]
+ }
+
+ /// Returns the raw input that was passed to `parse`.
+ pub fn raw_input(&self) -> &str {
+ &self.raw
+ }
+
+ /// Returns the raw input that was passed to `parse`, potentially owned.
+ pub fn into_raw_input(self) -> B {
+ self.raw
+ }
+
+}
+
+impl CharLit<&str> {
+ /// Makes a copy of the underlying buffer and returns the owned version of
+ /// `Self`.
+ pub fn to_owned(&self) -> CharLit<String> {
+ CharLit {
+ raw: self.raw.to_owned(),
+ start_suffix: self.start_suffix,
+ value: self.value,
+ }
+ }
+}
+
+impl<B: Buffer> fmt::Display for CharLit<B> {
+ fn fmt(&self, f: &mut fmt::Formatter<'_>) -> fmt::Result {
+ f.pad(&self.raw)
+ }
+}
+
+/// Precondition: first character in input must be `'`.
+#[inline(never)]
+pub(crate) fn parse_impl(input: &str) -> Result<(char, usize), ParseError> {
+ let first = input.chars().nth(1).ok_or(perr(None, UnterminatedCharLiteral))?;
+ let (c, len) = match first {
+ '\'' if input.chars().nth(2) == Some('\'') => return Err(perr(1, UnescapedSingleQuote)),
+ '\'' => return Err(perr(None, EmptyCharLiteral)),
+ '\n' | '\t' | '\r'
+ => return Err(perr(1, UnescapedSpecialWhitespace)),
+
+ '\\' => unescape::<char>(&input[1..], 1)?,
+ other => (other, other.len_utf8()),
+ };
+
+ match input[1 + len..].find('\'') {
+ Some(0) => {}
+ Some(_) => return Err(perr(None, OverlongCharLiteral)),
+ None => return Err(perr(None, UnterminatedCharLiteral)),
+ }
+
+ let start_suffix = 1 + len + 1;
+ let suffix = &input[start_suffix..];
+ check_suffix(suffix).map_err(|kind| perr(start_suffix, kind))?;
+
+ Ok((c, start_suffix))
+}
+
+#[cfg(test)]
+mod tests;
diff --git a/third_party/rust/litrs/src/char/tests.rs b/third_party/rust/litrs/src/char/tests.rs
new file mode 100644
index 0000000000..19219db73b
--- /dev/null
+++ b/third_party/rust/litrs/src/char/tests.rs
@@ -0,0 +1,227 @@
+use crate::{Literal, test_util::{assert_parse_ok_eq, assert_roundtrip}};
+use super::CharLit;
+
+// ===== Utility functions =======================================================================
+
+macro_rules! check {
+ ($lit:literal) => { check!($lit, stringify!($lit), "") };
+ ($lit:literal, $input:expr, $suffix:literal) => {
+ let input = $input;
+ let expected = CharLit {
+ raw: input,
+ start_suffix: input.len() - $suffix.len(),
+ value: $lit,
+ };
+
+ assert_parse_ok_eq(input, CharLit::parse(input), expected.clone(), "CharLit::parse");
+ assert_parse_ok_eq(input, Literal::parse(input), Literal::Char(expected), "Literal::parse");
+ let lit = CharLit::parse(input).unwrap();
+ assert_eq!(lit.value(), $lit);
+ assert_eq!(lit.suffix(), $suffix);
+ assert_roundtrip(expected.to_owned(), input);
+ };
+}
+
+
+// ===== Actual tests ============================================================================
+
+#[test]
+fn alphanumeric() {
+ check!('a');
+ check!('b');
+ check!('y');
+ check!('z');
+ check!('A');
+ check!('B');
+ check!('Y');
+ check!('Z');
+
+ check!('0');
+ check!('1');
+ check!('8');
+ check!('9');
+}
+
+#[test]
+fn special_chars() {
+ check!(' ');
+ check!('!');
+ check!('"');
+ check!('#');
+ check!('$');
+ check!('%');
+ check!('&');
+ check!('(');
+ check!(')');
+ check!('*');
+ check!('+');
+ check!(',');
+ check!('-');
+ check!('.');
+ check!('/');
+ check!(':');
+ check!(';');
+ check!('<');
+ check!('=');
+ check!('>');
+ check!('?');
+ check!('@');
+ check!('[');
+ check!(']');
+ check!('^');
+ check!('_');
+ check!('`');
+ check!('{');
+ check!('|');
+ check!('}');
+ check!('~');
+}
+
+#[test]
+fn unicode() {
+ check!('à°¨');
+ check!('犬');
+ check!('🦊');
+}
+
+#[test]
+fn quote_escapes() {
+ check!('\'');
+ check!('\"');
+}
+
+#[test]
+fn ascii_escapes() {
+ check!('\n');
+ check!('\r');
+ check!('\t');
+ check!('\\');
+ check!('\0');
+
+ check!('\x00');
+ check!('\x01');
+ check!('\x0c');
+ check!('\x0D');
+ check!('\x13');
+ check!('\x30');
+ check!('\x30');
+ check!('\x4B');
+ check!('\x6b');
+ check!('\x7F');
+ check!('\x7f');
+}
+
+#[test]
+fn unicode_escapes() {
+ check!('\u{0}');
+ check!('\u{00}');
+ check!('\u{b}');
+ check!('\u{B}');
+ check!('\u{7e}');
+ check!('\u{E4}');
+ check!('\u{e4}');
+ check!('\u{fc}');
+ check!('\u{Fc}');
+ check!('\u{fC}');
+ check!('\u{FC}');
+ check!('\u{b10}');
+ check!('\u{B10}');
+ check!('\u{0b10}');
+ check!('\u{2764}');
+ check!('\u{1f602}');
+ check!('\u{1F602}');
+
+ check!('\u{0}');
+ check!('\u{0__}');
+ check!('\u{3_b}');
+ check!('\u{1_F_6_0_2}');
+ check!('\u{1_F6_02_____}');
+}
+
+#[test]
+fn suffixes() {
+ check!('a', r##"'a'peter"##, "peter");
+ check!('#', r##"'#'peter"##, "peter");
+ check!('\n', r##"'\n'peter"##, "peter");
+ check!('\'', r##"'\''peter"##, "peter");
+ check!('\"', r##"'\"'peter"##, "peter");
+}
+
+#[test]
+fn invald_ascii_escapes() {
+ assert_err!(CharLit, r"'\x80'", NonAsciiXEscape, 1..5);
+ assert_err!(CharLit, r"'\x81'", NonAsciiXEscape, 1..5);
+ assert_err!(CharLit, r"'\x8a'", NonAsciiXEscape, 1..5);
+ assert_err!(CharLit, r"'\x8F'", NonAsciiXEscape, 1..5);
+ assert_err!(CharLit, r"'\xa0'", NonAsciiXEscape, 1..5);
+ assert_err!(CharLit, r"'\xB0'", NonAsciiXEscape, 1..5);
+ assert_err!(CharLit, r"'\xc3'", NonAsciiXEscape, 1..5);
+ assert_err!(CharLit, r"'\xDf'", NonAsciiXEscape, 1..5);
+ assert_err!(CharLit, r"'\xff'", NonAsciiXEscape, 1..5);
+ assert_err!(CharLit, r"'\xfF'", NonAsciiXEscape, 1..5);
+ assert_err!(CharLit, r"'\xFf'", NonAsciiXEscape, 1..5);
+ assert_err!(CharLit, r"'\xFF'", NonAsciiXEscape, 1..5);
+}
+
+#[test]
+fn invalid_escapes() {
+ assert_err!(CharLit, r"'\a'", UnknownEscape, 1..3);
+ assert_err!(CharLit, r"'\y'", UnknownEscape, 1..3);
+ assert_err!(CharLit, r"'\", UnterminatedEscape, 1);
+ assert_err!(CharLit, r"'\x'", UnterminatedEscape, 1..4);
+ assert_err!(CharLit, r"'\x1'", InvalidXEscape, 1..5);
+ assert_err!(CharLit, r"'\xaj'", InvalidXEscape, 1..5);
+ assert_err!(CharLit, r"'\xjb'", InvalidXEscape, 1..5);
+}
+
+#[test]
+fn invalid_unicode_escapes() {
+ assert_err!(CharLit, r"'\u'", UnicodeEscapeWithoutBrace, 1..3);
+ assert_err!(CharLit, r"'\u '", UnicodeEscapeWithoutBrace, 1..3);
+ assert_err!(CharLit, r"'\u3'", UnicodeEscapeWithoutBrace, 1..3);
+
+ assert_err!(CharLit, r"'\u{'", UnterminatedUnicodeEscape, 1..5);
+ assert_err!(CharLit, r"'\u{12'", UnterminatedUnicodeEscape, 1..7);
+ assert_err!(CharLit, r"'\u{a0b'", UnterminatedUnicodeEscape, 1..8);
+ assert_err!(CharLit, r"'\u{a0_b '", UnterminatedUnicodeEscape, 1..11);
+
+ assert_err!(CharLit, r"'\u{_}'", InvalidStartOfUnicodeEscape, 4);
+ assert_err!(CharLit, r"'\u{_5f}'", InvalidStartOfUnicodeEscape, 4);
+
+ assert_err!(CharLit, r"'\u{x}'", NonHexDigitInUnicodeEscape, 4);
+ assert_err!(CharLit, r"'\u{0x}'", NonHexDigitInUnicodeEscape, 5);
+ assert_err!(CharLit, r"'\u{3bx}'", NonHexDigitInUnicodeEscape, 6);
+ assert_err!(CharLit, r"'\u{3b_x}'", NonHexDigitInUnicodeEscape, 7);
+ assert_err!(CharLit, r"'\u{4x_}'", NonHexDigitInUnicodeEscape, 5);
+
+ assert_err!(CharLit, r"'\u{1234567}'", TooManyDigitInUnicodeEscape, 10);
+ assert_err!(CharLit, r"'\u{1234567}'", TooManyDigitInUnicodeEscape, 10);
+ assert_err!(CharLit, r"'\u{1_23_4_56_7}'", TooManyDigitInUnicodeEscape, 14);
+ assert_err!(CharLit, r"'\u{abcdef123}'", TooManyDigitInUnicodeEscape, 10);
+
+ assert_err!(CharLit, r"'\u{110000}'", InvalidUnicodeEscapeChar, 1..10);
+}
+
+#[test]
+fn parse_err() {
+ assert_err!(CharLit, r"''", EmptyCharLiteral, None);
+ assert_err!(CharLit, r"' ''", UnexpectedChar, 3);
+
+ assert_err!(CharLit, r"'", UnterminatedCharLiteral, None);
+ assert_err!(CharLit, r"'a", UnterminatedCharLiteral, None);
+ assert_err!(CharLit, r"'\n", UnterminatedCharLiteral, None);
+ assert_err!(CharLit, r"'\x35", UnterminatedCharLiteral, None);
+
+ assert_err!(CharLit, r"'ab'", OverlongCharLiteral, None);
+ assert_err!(CharLit, r"'a _'", OverlongCharLiteral, None);
+ assert_err!(CharLit, r"'\n3'", OverlongCharLiteral, None);
+
+ assert_err!(CharLit, r"", Empty, None);
+
+ assert_err!(CharLit, r"'''", UnescapedSingleQuote, 1);
+ assert_err!(CharLit, r"''''", UnescapedSingleQuote, 1);
+
+ assert_err!(CharLit, "'\n'", UnescapedSpecialWhitespace, 1);
+ assert_err!(CharLit, "'\t'", UnescapedSpecialWhitespace, 1);
+ assert_err!(CharLit, "'\r'", UnescapedSpecialWhitespace, 1);
+}
generated by cgit v1.2.3 (git 2.39.1) at 2024-08-15 17:22:28 +0000