2019-10-26 11:12:58 -05:00
|
|
|
//! Low-level Rust lexer.
|
|
|
|
//!
|
2020-04-25 13:19:54 -05:00
|
|
|
//! The idea with `librustc_lexer` is to make a reusable library,
|
|
|
|
//! by separating out pure lexing and rustc-specific concerns, like spans,
|
|
|
|
//! error reporting an interning. So, rustc_lexer operates directly on `&str`,
|
|
|
|
//! produces simple tokens which are a pair of type-tag and a bit of original text,
|
|
|
|
//! and does not report errors, instead storing them as flags on the token.
|
|
|
|
//!
|
2019-10-26 11:12:58 -05:00
|
|
|
//! Tokens produced by this lexer are not yet ready for parsing the Rust syntax,
|
2019-10-15 15:48:13 -05:00
|
|
|
//! for that see `librustc_parse::lexer`, which converts this basic token stream
|
2019-10-26 11:12:58 -05:00
|
|
|
//! into wide tokens used by actual parser.
|
|
|
|
//!
|
|
|
|
//! The purpose of this crate is to convert raw sources into a labeled sequence
|
|
|
|
//! of well-known token types, so building an actual Rust token stream will
|
|
|
|
//! be easier.
|
|
|
|
//!
|
|
|
|
//! Main entity of this crate is [`TokenKind`] enum which represents common
|
|
|
|
//! lexeme types.
|
|
|
|
|
2019-07-21 06:50:39 -05:00
|
|
|
// We want to be able to build this crate with a stable compiler, so no
|
|
|
|
// `#![feature]` attributes should be added.
|
2019-05-06 03:53:40 -05:00
|
|
|
|
|
|
|
mod cursor;
|
2019-07-21 08:46:11 -05:00
|
|
|
pub mod unescape;
|
2019-05-06 03:53:40 -05:00
|
|
|
|
2020-03-28 00:46:20 -05:00
|
|
|
#[cfg(test)]
|
|
|
|
mod tests;
|
|
|
|
|
2019-11-03 03:57:12 -06:00
|
|
|
use self::LiteralKind::*;
|
2019-12-22 16:42:04 -06:00
|
|
|
use self::TokenKind::*;
|
|
|
|
use crate::cursor::{Cursor, EOF_CHAR};
|
2020-05-29 10:37:16 -05:00
|
|
|
use std::convert::TryFrom;
|
2019-05-06 03:53:40 -05:00
|
|
|
|
2019-10-26 11:12:58 -05:00
|
|
|
/// Parsed token.
|
|
|
|
/// It doesn't contain information about data that has been parsed,
|
|
|
|
/// only the type of the token and its size.
|
2019-05-06 03:53:40 -05:00
|
|
|
pub struct Token {
|
|
|
|
pub kind: TokenKind,
|
|
|
|
pub len: usize,
|
|
|
|
}
|
|
|
|
|
2019-10-26 11:12:58 -05:00
|
|
|
impl Token {
|
|
|
|
fn new(kind: TokenKind, len: usize) -> Token {
|
|
|
|
Token { kind, len }
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2020-02-13 10:14:21 -06:00
|
|
|
/// Enum representing common lexeme types.
|
2019-05-06 03:53:40 -05:00
|
|
|
#[derive(Clone, Copy, Debug, PartialEq, Eq, PartialOrd, Ord)]
|
|
|
|
pub enum TokenKind {
|
2019-10-26 11:12:58 -05:00
|
|
|
// Multi-char tokens:
|
|
|
|
/// "// comment"
|
2019-05-06 03:53:40 -05:00
|
|
|
LineComment,
|
2019-10-26 11:12:58 -05:00
|
|
|
/// "/* block comment */"
|
|
|
|
/// Block comments can be recursive, so the sequence like "/* /* */"
|
|
|
|
/// will not be considered terminated and will result in a parsing error.
|
2019-05-06 03:53:40 -05:00
|
|
|
BlockComment { terminated: bool },
|
2019-10-26 11:12:58 -05:00
|
|
|
/// Any whitespace characters sequence.
|
2019-05-06 03:53:40 -05:00
|
|
|
Whitespace,
|
2019-10-26 11:12:58 -05:00
|
|
|
/// "ident" or "continue"
|
|
|
|
/// At this step keywords are also considered identifiers.
|
2019-05-06 03:53:40 -05:00
|
|
|
Ident,
|
2019-10-26 11:12:58 -05:00
|
|
|
/// "r#ident"
|
2019-05-06 03:53:40 -05:00
|
|
|
RawIdent,
|
2019-10-26 11:12:58 -05:00
|
|
|
/// "12_u8", "1.0e-40", "b"123"". See `LiteralKind` for more details.
|
2019-05-06 03:53:40 -05:00
|
|
|
Literal { kind: LiteralKind, suffix_start: usize },
|
2019-10-26 11:12:58 -05:00
|
|
|
/// "'a"
|
2019-05-06 03:53:40 -05:00
|
|
|
Lifetime { starts_with_number: bool },
|
2019-10-26 11:12:58 -05:00
|
|
|
|
|
|
|
// One-char tokens:
|
|
|
|
/// ";"
|
2019-05-06 03:53:40 -05:00
|
|
|
Semi,
|
2019-10-26 11:12:58 -05:00
|
|
|
/// ","
|
2019-05-06 03:53:40 -05:00
|
|
|
Comma,
|
2019-10-26 11:12:58 -05:00
|
|
|
/// "."
|
2019-05-06 03:53:40 -05:00
|
|
|
Dot,
|
2019-10-26 11:12:58 -05:00
|
|
|
/// "("
|
2019-05-06 03:53:40 -05:00
|
|
|
OpenParen,
|
2019-10-26 11:12:58 -05:00
|
|
|
/// ")"
|
2019-05-06 03:53:40 -05:00
|
|
|
CloseParen,
|
2019-10-26 11:12:58 -05:00
|
|
|
/// "{"
|
2019-05-06 03:53:40 -05:00
|
|
|
OpenBrace,
|
2019-10-26 11:12:58 -05:00
|
|
|
/// "}"
|
2019-05-06 03:53:40 -05:00
|
|
|
CloseBrace,
|
2019-10-26 11:12:58 -05:00
|
|
|
/// "["
|
2019-05-06 03:53:40 -05:00
|
|
|
OpenBracket,
|
2019-10-26 11:12:58 -05:00
|
|
|
/// "]"
|
2019-05-06 03:53:40 -05:00
|
|
|
CloseBracket,
|
2019-10-26 11:12:58 -05:00
|
|
|
/// "@"
|
2019-05-06 03:53:40 -05:00
|
|
|
At,
|
2019-10-26 11:12:58 -05:00
|
|
|
/// "#"
|
2019-05-06 03:53:40 -05:00
|
|
|
Pound,
|
2019-10-26 11:12:58 -05:00
|
|
|
/// "~"
|
2019-05-06 03:53:40 -05:00
|
|
|
Tilde,
|
2019-10-26 11:12:58 -05:00
|
|
|
/// "?"
|
2019-05-06 03:53:40 -05:00
|
|
|
Question,
|
2019-10-26 11:12:58 -05:00
|
|
|
/// ":"
|
2019-05-06 03:53:40 -05:00
|
|
|
Colon,
|
2019-10-26 11:12:58 -05:00
|
|
|
/// "$"
|
2019-05-06 03:53:40 -05:00
|
|
|
Dollar,
|
2019-10-26 11:12:58 -05:00
|
|
|
/// "="
|
2019-05-06 03:53:40 -05:00
|
|
|
Eq,
|
2019-10-26 11:12:58 -05:00
|
|
|
/// "!"
|
2019-05-06 03:53:40 -05:00
|
|
|
Not,
|
2019-10-26 11:12:58 -05:00
|
|
|
/// "<"
|
2019-05-06 03:53:40 -05:00
|
|
|
Lt,
|
2019-10-26 11:12:58 -05:00
|
|
|
/// ">"
|
2019-05-06 03:53:40 -05:00
|
|
|
Gt,
|
2019-10-26 11:12:58 -05:00
|
|
|
/// "-"
|
2019-05-06 03:53:40 -05:00
|
|
|
Minus,
|
2019-10-26 11:12:58 -05:00
|
|
|
/// "&"
|
2019-05-06 03:53:40 -05:00
|
|
|
And,
|
2019-10-26 11:12:58 -05:00
|
|
|
/// "|"
|
2019-05-06 03:53:40 -05:00
|
|
|
Or,
|
2019-10-26 11:12:58 -05:00
|
|
|
/// "+"
|
2019-05-06 03:53:40 -05:00
|
|
|
Plus,
|
2019-10-26 11:12:58 -05:00
|
|
|
/// "*"
|
2019-05-06 03:53:40 -05:00
|
|
|
Star,
|
2019-10-26 11:12:58 -05:00
|
|
|
/// "/"
|
2019-05-06 03:53:40 -05:00
|
|
|
Slash,
|
2019-10-26 11:12:58 -05:00
|
|
|
/// "^"
|
2019-05-06 03:53:40 -05:00
|
|
|
Caret,
|
2019-10-26 11:12:58 -05:00
|
|
|
/// "%"
|
2019-05-06 03:53:40 -05:00
|
|
|
Percent,
|
2019-10-26 11:12:58 -05:00
|
|
|
|
|
|
|
/// Unknown token, not expected by the lexer, e.g. "№"
|
2019-05-06 03:53:40 -05:00
|
|
|
Unknown,
|
|
|
|
}
|
|
|
|
|
|
|
|
#[derive(Clone, Copy, Debug, PartialEq, Eq, PartialOrd, Ord)]
|
|
|
|
pub enum LiteralKind {
|
2019-10-26 11:12:58 -05:00
|
|
|
/// "12_u8", "0o100", "0b120i99"
|
2019-05-06 03:53:40 -05:00
|
|
|
Int { base: Base, empty_int: bool },
|
2019-10-26 11:12:58 -05:00
|
|
|
/// "12.34f32", "0b100.100"
|
2019-05-06 03:53:40 -05:00
|
|
|
Float { base: Base, empty_exponent: bool },
|
2019-10-26 11:12:58 -05:00
|
|
|
/// "'a'", "'\\'", "'''", "';"
|
2019-05-06 03:53:40 -05:00
|
|
|
Char { terminated: bool },
|
2019-10-26 11:12:58 -05:00
|
|
|
/// "b'a'", "b'\\'", "b'''", "b';"
|
2019-05-06 03:53:40 -05:00
|
|
|
Byte { terminated: bool },
|
2019-10-26 11:12:58 -05:00
|
|
|
/// ""abc"", ""abc"
|
2019-05-06 03:53:40 -05:00
|
|
|
Str { terminated: bool },
|
2019-10-26 11:12:58 -05:00
|
|
|
/// "b"abc"", "b"abc"
|
2019-05-06 03:53:40 -05:00
|
|
|
ByteStr { terminated: bool },
|
2019-10-26 11:12:58 -05:00
|
|
|
/// "r"abc"", "r#"abc"#", "r####"ab"###"c"####", "r#"a"
|
2020-05-29 10:37:16 -05:00
|
|
|
RawStr { n_hashes: u16, err: Option<RawStrError> },
|
2019-10-26 11:12:58 -05:00
|
|
|
/// "br"abc"", "br#"abc"#", "br####"ab"###"c"####", "br#"a"
|
2020-05-29 10:37:16 -05:00
|
|
|
RawByteStr { n_hashes: u16, err: Option<RawStrError> },
|
2020-03-28 00:46:20 -05:00
|
|
|
}
|
|
|
|
|
2020-03-29 10:12:48 -05:00
|
|
|
/// Error produced validating a raw string. Represents cases like:
|
2020-05-29 10:37:16 -05:00
|
|
|
/// - `r##~"abcde"##`: `InvalidStarter`
|
|
|
|
/// - `r###"abcde"##`: `NoTerminator { expected: 3, found: 2, possible_terminator_offset: Some(11)`
|
|
|
|
/// - Too many `#`s (>65535): `TooManyDelimiters`
|
2020-03-28 00:46:20 -05:00
|
|
|
#[derive(Clone, Copy, Debug, PartialEq, Eq, PartialOrd, Ord)]
|
2020-05-29 10:37:16 -05:00
|
|
|
pub enum RawStrError {
|
2020-03-29 10:12:48 -05:00
|
|
|
/// Non `#` characters exist between `r` and `"` eg. `r#~"..`
|
2020-05-29 10:37:16 -05:00
|
|
|
InvalidStarter { bad_char: char },
|
2020-03-29 10:12:48 -05:00
|
|
|
/// The string was never terminated. `possible_terminator_offset` is the number of characters after `r` or `br` where they
|
2020-03-28 00:46:20 -05:00
|
|
|
/// may have intended to terminate it.
|
|
|
|
NoTerminator { expected: usize, found: usize, possible_terminator_offset: Option<usize> },
|
2020-05-29 10:37:16 -05:00
|
|
|
/// More than 65535 `#`s exist.
|
|
|
|
TooManyDelimiters { found: usize },
|
2019-05-06 03:53:40 -05:00
|
|
|
}
|
|
|
|
|
2019-10-26 11:12:58 -05:00
|
|
|
/// Base of numeric literal encoding according to its prefix.
|
2019-05-06 03:53:40 -05:00
|
|
|
#[derive(Clone, Copy, Debug, PartialEq, Eq, PartialOrd, Ord)]
|
|
|
|
pub enum Base {
|
2019-10-26 11:12:58 -05:00
|
|
|
/// Literal starts with "0b".
|
2019-05-06 03:53:40 -05:00
|
|
|
Binary,
|
2019-10-26 11:12:58 -05:00
|
|
|
/// Literal starts with "0o".
|
2019-05-06 03:53:40 -05:00
|
|
|
Octal,
|
2019-10-26 11:12:58 -05:00
|
|
|
/// Literal starts with "0x".
|
2019-05-06 03:53:40 -05:00
|
|
|
Hexadecimal,
|
2019-10-26 11:12:58 -05:00
|
|
|
/// Literal doesn't contain a prefix.
|
2019-05-06 03:53:40 -05:00
|
|
|
Decimal,
|
|
|
|
}
|
|
|
|
|
2019-10-26 11:12:58 -05:00
|
|
|
/// `rustc` allows files to have a shebang, e.g. "#!/usr/bin/rustrun",
|
Fix bug in shebang handling
Shebang handling was too agressive in stripping out the first line in cases where it is actually _not_ a shebang, but instead, valid rust (#70528). This is a second attempt at resolving this issue (the first attempt was flawed, for, among other reasons, causing an ICE in certain cases (#71372, #71471).
The behavior is now codified by a number of UI tests, but simply:
For the first line to be a shebang, the following must all be true:
1. The line must start with `#!`
2. The line must contain a non whitespace character after `#!`
3. The next character in the file, ignoring comments & whitespace must not be `[`
I believe this is a strict superset of what we used to allow, so perhaps a crater run is unnecessary, but probably not a terrible idea.
2020-04-23 14:51:12 -05:00
|
|
|
/// but shebang isn't a part of rust syntax.
|
2019-05-06 03:53:40 -05:00
|
|
|
pub fn strip_shebang(input: &str) -> Option<usize> {
|
2020-05-29 13:51:46 -05:00
|
|
|
// Shebang must start with `#!` literally, without any preceding whitespace.
|
2020-06-21 16:40:11 -05:00
|
|
|
// For simplicity we consider any line starting with `#!` a shebang,
|
|
|
|
// regardless of restrictions put on shebangs by specific platforms.
|
|
|
|
if let Some(input_tail) = input.strip_prefix("#!") {
|
|
|
|
// Ok, this is a shebang but if the next non-whitespace token is `[` or maybe
|
|
|
|
// a doc comment (due to `TokenKind::(Line,Block)Comment` ambiguity at lexer level),
|
|
|
|
// then it may be valid Rust code, so consider it Rust code.
|
|
|
|
let next_non_whitespace_token = tokenize(input_tail).map(|tok| tok.kind).find(|tok|
|
|
|
|
!matches!(tok, TokenKind::Whitespace | TokenKind::LineComment | TokenKind::BlockComment { .. })
|
|
|
|
);
|
|
|
|
if next_non_whitespace_token != Some(TokenKind::OpenBracket) {
|
|
|
|
// No other choice than to consider this a shebang.
|
|
|
|
return Some(2 + input_tail.lines().next().unwrap_or_default().len());
|
2020-05-29 13:51:46 -05:00
|
|
|
}
|
Fix bug in shebang handling
Shebang handling was too agressive in stripping out the first line in cases where it is actually _not_ a shebang, but instead, valid rust (#70528). This is a second attempt at resolving this issue (the first attempt was flawed, for, among other reasons, causing an ICE in certain cases (#71372, #71471).
The behavior is now codified by a number of UI tests, but simply:
For the first line to be a shebang, the following must all be true:
1. The line must start with `#!`
2. The line must contain a non whitespace character after `#!`
3. The next character in the file, ignoring comments & whitespace must not be `[`
I believe this is a strict superset of what we used to allow, so perhaps a crater run is unnecessary, but probably not a terrible idea.
2020-04-23 14:51:12 -05:00
|
|
|
}
|
2020-05-29 13:51:46 -05:00
|
|
|
None
|
2019-05-06 03:53:40 -05:00
|
|
|
}
|
|
|
|
|
2019-10-26 11:12:58 -05:00
|
|
|
/// Parses the first token from the provided input string.
|
2019-05-06 03:53:40 -05:00
|
|
|
pub fn first_token(input: &str) -> Token {
|
|
|
|
debug_assert!(!input.is_empty());
|
|
|
|
Cursor::new(input).advance_token()
|
|
|
|
}
|
|
|
|
|
2019-10-26 11:12:58 -05:00
|
|
|
/// Creates an iterator that produces tokens from the input string.
|
2019-05-06 03:53:40 -05:00
|
|
|
pub fn tokenize(mut input: &str) -> impl Iterator<Item = Token> + '_ {
|
|
|
|
std::iter::from_fn(move || {
|
|
|
|
if input.is_empty() {
|
|
|
|
return None;
|
|
|
|
}
|
|
|
|
let token = first_token(input);
|
|
|
|
input = &input[token.len..];
|
|
|
|
Some(token)
|
|
|
|
})
|
|
|
|
}
|
|
|
|
|
2019-09-04 05:16:36 -05:00
|
|
|
/// True if `c` is considered a whitespace according to Rust language definition.
|
2019-10-26 11:12:58 -05:00
|
|
|
/// See [Rust language reference](https://doc.rust-lang.org/reference/whitespace.html)
|
|
|
|
/// for definitions of these classes.
|
2019-09-04 05:16:36 -05:00
|
|
|
pub fn is_whitespace(c: char) -> bool {
|
|
|
|
// This is Pattern_White_Space.
|
|
|
|
//
|
|
|
|
// Note that this set is stable (ie, it doesn't change with different
|
|
|
|
// Unicode versions), so it's ok to just hard-code the values.
|
|
|
|
|
|
|
|
match c {
|
|
|
|
// Usual ASCII suspects
|
|
|
|
| '\u{0009}' // \t
|
|
|
|
| '\u{000A}' // \n
|
|
|
|
| '\u{000B}' // vertical tab
|
|
|
|
| '\u{000C}' // form feed
|
|
|
|
| '\u{000D}' // \r
|
|
|
|
| '\u{0020}' // space
|
|
|
|
|
|
|
|
// NEXT LINE from latin1
|
|
|
|
| '\u{0085}'
|
|
|
|
|
|
|
|
// Bidi markers
|
|
|
|
| '\u{200E}' // LEFT-TO-RIGHT MARK
|
|
|
|
| '\u{200F}' // RIGHT-TO-LEFT MARK
|
|
|
|
|
|
|
|
// Dedicated whitespace characters from Unicode
|
|
|
|
| '\u{2028}' // LINE SEPARATOR
|
|
|
|
| '\u{2029}' // PARAGRAPH SEPARATOR
|
2020-03-28 00:46:20 -05:00
|
|
|
=> true,
|
2019-09-04 05:16:36 -05:00
|
|
|
_ => false,
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
/// True if `c` is valid as a first character of an identifier.
|
2019-10-26 11:12:58 -05:00
|
|
|
/// See [Rust language reference](https://doc.rust-lang.org/reference/identifiers.html) for
|
|
|
|
/// a formal definition of valid identifier name.
|
2019-09-04 05:16:36 -05:00
|
|
|
pub fn is_id_start(c: char) -> bool {
|
|
|
|
// This is XID_Start OR '_' (which formally is not a XID_Start).
|
|
|
|
// We also add fast-path for ascii idents
|
|
|
|
('a' <= c && c <= 'z')
|
|
|
|
|| ('A' <= c && c <= 'Z')
|
|
|
|
|| c == '_'
|
|
|
|
|| (c > '\x7f' && unicode_xid::UnicodeXID::is_xid_start(c))
|
|
|
|
}
|
|
|
|
|
|
|
|
/// True if `c` is valid as a non-first character of an identifier.
|
2019-10-26 11:12:58 -05:00
|
|
|
/// See [Rust language reference](https://doc.rust-lang.org/reference/identifiers.html) for
|
|
|
|
/// a formal definition of valid identifier name.
|
2019-09-04 05:16:36 -05:00
|
|
|
pub fn is_id_continue(c: char) -> bool {
|
|
|
|
// This is exactly XID_Continue.
|
|
|
|
// We also add fast-path for ascii idents
|
|
|
|
('a' <= c && c <= 'z')
|
|
|
|
|| ('A' <= c && c <= 'Z')
|
|
|
|
|| ('0' <= c && c <= '9')
|
|
|
|
|| c == '_'
|
|
|
|
|| (c > '\x7f' && unicode_xid::UnicodeXID::is_xid_continue(c))
|
|
|
|
}
|
|
|
|
|
2019-05-06 03:53:40 -05:00
|
|
|
impl Cursor<'_> {
|
2019-10-26 11:12:58 -05:00
|
|
|
/// Parses a token from the input string.
|
2019-05-06 03:53:40 -05:00
|
|
|
fn advance_token(&mut self) -> Token {
|
|
|
|
let first_char = self.bump().unwrap();
|
|
|
|
let token_kind = match first_char {
|
2019-10-26 11:12:58 -05:00
|
|
|
// Slash, comment or block comment.
|
2019-11-03 02:39:39 -06:00
|
|
|
'/' => match self.first() {
|
2019-05-06 03:53:40 -05:00
|
|
|
'/' => self.line_comment(),
|
|
|
|
'*' => self.block_comment(),
|
2019-08-19 11:00:24 -05:00
|
|
|
_ => Slash,
|
2019-05-06 03:53:40 -05:00
|
|
|
},
|
2019-10-26 11:12:58 -05:00
|
|
|
|
|
|
|
// Whitespace sequence.
|
2019-09-04 05:16:36 -05:00
|
|
|
c if is_whitespace(c) => self.whitespace(),
|
2019-10-26 11:12:58 -05:00
|
|
|
|
2019-11-03 02:39:39 -06:00
|
|
|
// Raw identifier, raw string literal or identifier.
|
|
|
|
'r' => match (self.first(), self.second()) {
|
2019-09-04 05:16:36 -05:00
|
|
|
('#', c1) if is_id_start(c1) => self.raw_ident(),
|
2019-05-06 03:53:40 -05:00
|
|
|
('#', _) | ('"', _) => {
|
2020-05-29 10:37:16 -05:00
|
|
|
let (n_hashes, err) = self.raw_double_quoted_string(1);
|
2019-05-06 03:53:40 -05:00
|
|
|
let suffix_start = self.len_consumed();
|
2020-05-29 10:37:16 -05:00
|
|
|
if err.is_none() {
|
2019-05-06 03:53:40 -05:00
|
|
|
self.eat_literal_suffix();
|
|
|
|
}
|
2020-05-29 10:37:16 -05:00
|
|
|
let kind = RawStr { n_hashes, err };
|
2019-05-06 03:53:40 -05:00
|
|
|
Literal { kind, suffix_start }
|
|
|
|
}
|
|
|
|
_ => self.ident(),
|
|
|
|
},
|
2019-10-26 11:12:58 -05:00
|
|
|
|
|
|
|
// Byte literal, byte string literal, raw byte string literal or identifier.
|
2019-11-03 02:39:39 -06:00
|
|
|
'b' => match (self.first(), self.second()) {
|
2019-05-06 03:53:40 -05:00
|
|
|
('\'', _) => {
|
|
|
|
self.bump();
|
|
|
|
let terminated = self.single_quoted_string();
|
|
|
|
let suffix_start = self.len_consumed();
|
|
|
|
if terminated {
|
|
|
|
self.eat_literal_suffix();
|
|
|
|
}
|
|
|
|
let kind = Byte { terminated };
|
|
|
|
Literal { kind, suffix_start }
|
|
|
|
}
|
|
|
|
('"', _) => {
|
|
|
|
self.bump();
|
|
|
|
let terminated = self.double_quoted_string();
|
|
|
|
let suffix_start = self.len_consumed();
|
|
|
|
if terminated {
|
|
|
|
self.eat_literal_suffix();
|
|
|
|
}
|
|
|
|
let kind = ByteStr { terminated };
|
|
|
|
Literal { kind, suffix_start }
|
|
|
|
}
|
|
|
|
('r', '"') | ('r', '#') => {
|
|
|
|
self.bump();
|
2020-05-29 10:37:16 -05:00
|
|
|
let (n_hashes, err) = self.raw_double_quoted_string(2);
|
2019-05-06 03:53:40 -05:00
|
|
|
let suffix_start = self.len_consumed();
|
2020-05-29 10:37:16 -05:00
|
|
|
if err.is_none() {
|
2019-05-06 03:53:40 -05:00
|
|
|
self.eat_literal_suffix();
|
|
|
|
}
|
2020-05-29 10:37:16 -05:00
|
|
|
let kind = RawByteStr { n_hashes, err };
|
2019-05-06 03:53:40 -05:00
|
|
|
Literal { kind, suffix_start }
|
|
|
|
}
|
|
|
|
_ => self.ident(),
|
|
|
|
},
|
2019-10-26 11:12:58 -05:00
|
|
|
|
|
|
|
// Identifier (this should be checked after other variant that can
|
|
|
|
// start as identifier).
|
2019-09-04 05:16:36 -05:00
|
|
|
c if is_id_start(c) => self.ident(),
|
2019-10-26 11:12:58 -05:00
|
|
|
|
|
|
|
// Numeric literal.
|
2019-05-06 03:53:40 -05:00
|
|
|
c @ '0'..='9' => {
|
|
|
|
let literal_kind = self.number(c);
|
|
|
|
let suffix_start = self.len_consumed();
|
|
|
|
self.eat_literal_suffix();
|
|
|
|
TokenKind::Literal { kind: literal_kind, suffix_start }
|
|
|
|
}
|
2019-10-26 11:12:58 -05:00
|
|
|
|
|
|
|
// One-symbol tokens.
|
2019-05-06 03:53:40 -05:00
|
|
|
';' => Semi,
|
|
|
|
',' => Comma,
|
2019-08-19 11:00:24 -05:00
|
|
|
'.' => Dot,
|
2019-05-06 03:53:40 -05:00
|
|
|
'(' => OpenParen,
|
|
|
|
')' => CloseParen,
|
|
|
|
'{' => OpenBrace,
|
|
|
|
'}' => CloseBrace,
|
|
|
|
'[' => OpenBracket,
|
|
|
|
']' => CloseBracket,
|
|
|
|
'@' => At,
|
|
|
|
'#' => Pound,
|
|
|
|
'~' => Tilde,
|
|
|
|
'?' => Question,
|
2019-08-19 11:00:24 -05:00
|
|
|
':' => Colon,
|
2019-05-06 03:53:40 -05:00
|
|
|
'$' => Dollar,
|
2019-08-19 11:00:24 -05:00
|
|
|
'=' => Eq,
|
|
|
|
'!' => Not,
|
|
|
|
'<' => Lt,
|
|
|
|
'>' => Gt,
|
|
|
|
'-' => Minus,
|
|
|
|
'&' => And,
|
|
|
|
'|' => Or,
|
|
|
|
'+' => Plus,
|
|
|
|
'*' => Star,
|
|
|
|
'^' => Caret,
|
|
|
|
'%' => Percent,
|
2019-10-26 11:12:58 -05:00
|
|
|
|
|
|
|
// Lifetime or character literal.
|
2019-05-06 03:53:40 -05:00
|
|
|
'\'' => self.lifetime_or_char(),
|
2019-10-26 11:12:58 -05:00
|
|
|
|
|
|
|
// String literal.
|
2019-05-06 03:53:40 -05:00
|
|
|
'"' => {
|
|
|
|
let terminated = self.double_quoted_string();
|
|
|
|
let suffix_start = self.len_consumed();
|
|
|
|
if terminated {
|
|
|
|
self.eat_literal_suffix();
|
|
|
|
}
|
|
|
|
let kind = Str { terminated };
|
|
|
|
Literal { kind, suffix_start }
|
|
|
|
}
|
|
|
|
_ => Unknown,
|
|
|
|
};
|
|
|
|
Token::new(token_kind, self.len_consumed())
|
|
|
|
}
|
|
|
|
|
|
|
|
fn line_comment(&mut self) -> TokenKind {
|
2019-11-03 02:39:39 -06:00
|
|
|
debug_assert!(self.prev() == '/' && self.first() == '/');
|
2019-05-06 03:53:40 -05:00
|
|
|
self.bump();
|
2019-11-03 02:42:08 -06:00
|
|
|
self.eat_while(|c| c != '\n');
|
2019-05-06 03:53:40 -05:00
|
|
|
LineComment
|
|
|
|
}
|
|
|
|
|
|
|
|
fn block_comment(&mut self) -> TokenKind {
|
2019-11-03 02:39:39 -06:00
|
|
|
debug_assert!(self.prev() == '/' && self.first() == '*');
|
2019-05-06 03:53:40 -05:00
|
|
|
self.bump();
|
|
|
|
let mut depth = 1usize;
|
|
|
|
while let Some(c) = self.bump() {
|
|
|
|
match c {
|
2019-11-03 02:39:39 -06:00
|
|
|
'/' if self.first() == '*' => {
|
2019-05-06 03:53:40 -05:00
|
|
|
self.bump();
|
|
|
|
depth += 1;
|
|
|
|
}
|
2019-11-03 02:39:39 -06:00
|
|
|
'*' if self.first() == '/' => {
|
2019-05-06 03:53:40 -05:00
|
|
|
self.bump();
|
|
|
|
depth -= 1;
|
|
|
|
if depth == 0 {
|
2019-10-26 11:12:58 -05:00
|
|
|
// This block comment is closed, so for a construction like "/* */ */"
|
|
|
|
// there will be a successfully parsed block comment "/* */"
|
|
|
|
// and " */" will be processed separately.
|
2019-05-06 03:53:40 -05:00
|
|
|
break;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
_ => (),
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
BlockComment { terminated: depth == 0 }
|
|
|
|
}
|
|
|
|
|
|
|
|
fn whitespace(&mut self) -> TokenKind {
|
2019-09-04 05:16:36 -05:00
|
|
|
debug_assert!(is_whitespace(self.prev()));
|
2019-11-03 02:42:08 -06:00
|
|
|
self.eat_while(is_whitespace);
|
2019-05-06 03:53:40 -05:00
|
|
|
Whitespace
|
|
|
|
}
|
|
|
|
|
|
|
|
fn raw_ident(&mut self) -> TokenKind {
|
2019-12-22 16:42:04 -06:00
|
|
|
debug_assert!(self.prev() == 'r' && self.first() == '#' && is_id_start(self.second()));
|
2019-11-03 02:42:08 -06:00
|
|
|
// Eat "#" symbol.
|
2019-05-06 03:53:40 -05:00
|
|
|
self.bump();
|
2019-11-03 02:42:08 -06:00
|
|
|
// Eat the identifier part of RawIdent.
|
|
|
|
self.eat_identifier();
|
2019-05-06 03:53:40 -05:00
|
|
|
RawIdent
|
|
|
|
}
|
|
|
|
|
|
|
|
fn ident(&mut self) -> TokenKind {
|
2019-09-04 05:16:36 -05:00
|
|
|
debug_assert!(is_id_start(self.prev()));
|
2019-11-03 02:42:08 -06:00
|
|
|
// Start is already eaten, eat the rest of identifier.
|
|
|
|
self.eat_while(is_id_continue);
|
2019-05-06 03:53:40 -05:00
|
|
|
Ident
|
|
|
|
}
|
|
|
|
|
|
|
|
fn number(&mut self, first_digit: char) -> LiteralKind {
|
|
|
|
debug_assert!('0' <= self.prev() && self.prev() <= '9');
|
|
|
|
let mut base = Base::Decimal;
|
|
|
|
if first_digit == '0' {
|
2019-10-26 11:12:58 -05:00
|
|
|
// Attempt to parse encoding base.
|
2019-11-03 02:39:39 -06:00
|
|
|
let has_digits = match self.first() {
|
2019-05-06 03:53:40 -05:00
|
|
|
'b' => {
|
|
|
|
base = Base::Binary;
|
|
|
|
self.bump();
|
|
|
|
self.eat_decimal_digits()
|
|
|
|
}
|
|
|
|
'o' => {
|
|
|
|
base = Base::Octal;
|
|
|
|
self.bump();
|
|
|
|
self.eat_decimal_digits()
|
|
|
|
}
|
|
|
|
'x' => {
|
|
|
|
base = Base::Hexadecimal;
|
|
|
|
self.bump();
|
|
|
|
self.eat_hexadecimal_digits()
|
|
|
|
}
|
2019-10-26 11:12:58 -05:00
|
|
|
// Not a base prefix.
|
2019-05-06 03:53:40 -05:00
|
|
|
'0'..='9' | '_' | '.' | 'e' | 'E' => {
|
|
|
|
self.eat_decimal_digits();
|
|
|
|
true
|
|
|
|
}
|
2019-10-26 11:12:58 -05:00
|
|
|
// Just a 0.
|
2019-05-06 03:53:40 -05:00
|
|
|
_ => return Int { base, empty_int: false },
|
|
|
|
};
|
2019-10-26 11:12:58 -05:00
|
|
|
// Base prefix was provided, but there were no digits
|
|
|
|
// after it, e.g. "0x".
|
2019-05-06 03:53:40 -05:00
|
|
|
if !has_digits {
|
|
|
|
return Int { base, empty_int: true };
|
|
|
|
}
|
|
|
|
} else {
|
2019-10-26 11:12:58 -05:00
|
|
|
// No base prefix, parse number in the usual way.
|
2019-05-06 03:53:40 -05:00
|
|
|
self.eat_decimal_digits();
|
|
|
|
};
|
|
|
|
|
2019-11-03 02:39:39 -06:00
|
|
|
match self.first() {
|
2019-05-06 03:53:40 -05:00
|
|
|
// Don't be greedy if this is actually an
|
|
|
|
// integer literal followed by field/method access or a range pattern
|
|
|
|
// (`0..2` and `12.foo()`)
|
2019-12-22 16:42:04 -06:00
|
|
|
'.' if self.second() != '.' && !is_id_start(self.second()) => {
|
2019-05-06 03:53:40 -05:00
|
|
|
// might have stuff after the ., and if it does, it needs to start
|
|
|
|
// with a number
|
|
|
|
self.bump();
|
|
|
|
let mut empty_exponent = false;
|
2019-11-03 02:39:39 -06:00
|
|
|
if self.first().is_digit(10) {
|
2019-05-06 03:53:40 -05:00
|
|
|
self.eat_decimal_digits();
|
2019-11-03 02:39:39 -06:00
|
|
|
match self.first() {
|
2019-05-06 03:53:40 -05:00
|
|
|
'e' | 'E' => {
|
|
|
|
self.bump();
|
2019-11-03 02:43:47 -06:00
|
|
|
empty_exponent = !self.eat_float_exponent();
|
2019-05-06 03:53:40 -05:00
|
|
|
}
|
|
|
|
_ => (),
|
|
|
|
}
|
|
|
|
}
|
|
|
|
Float { base, empty_exponent }
|
|
|
|
}
|
|
|
|
'e' | 'E' => {
|
|
|
|
self.bump();
|
2019-11-03 02:43:47 -06:00
|
|
|
let empty_exponent = !self.eat_float_exponent();
|
2019-05-06 03:53:40 -05:00
|
|
|
Float { base, empty_exponent }
|
|
|
|
}
|
|
|
|
_ => Int { base, empty_int: false },
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
fn lifetime_or_char(&mut self) -> TokenKind {
|
|
|
|
debug_assert!(self.prev() == '\'');
|
2019-10-26 11:12:58 -05:00
|
|
|
|
2019-11-03 03:56:49 -06:00
|
|
|
let can_be_a_lifetime = if self.second() == '\'' {
|
|
|
|
// It's surely not a lifetime.
|
|
|
|
false
|
|
|
|
} else {
|
|
|
|
// If the first symbol is valid for identifier, it can be a lifetime.
|
|
|
|
// Also check if it's a number for a better error reporting (so '0 will
|
|
|
|
// be reported as invalid lifetime and not as unterminated char literal).
|
|
|
|
is_id_start(self.first()) || self.first().is_digit(10)
|
|
|
|
};
|
2019-05-06 03:53:40 -05:00
|
|
|
|
2019-11-03 03:56:49 -06:00
|
|
|
if !can_be_a_lifetime {
|
|
|
|
let terminated = self.single_quoted_string();
|
|
|
|
let suffix_start = self.len_consumed();
|
|
|
|
if terminated {
|
|
|
|
self.eat_literal_suffix();
|
|
|
|
}
|
|
|
|
let kind = Char { terminated };
|
|
|
|
return Literal { kind, suffix_start };
|
2019-05-06 03:53:40 -05:00
|
|
|
}
|
2019-10-26 11:12:58 -05:00
|
|
|
|
2019-11-03 03:56:49 -06:00
|
|
|
// Either a lifetime or a character literal with
|
|
|
|
// length greater than 1.
|
|
|
|
|
|
|
|
let starts_with_number = self.first().is_digit(10);
|
|
|
|
|
|
|
|
// Skip the literal contents.
|
|
|
|
// First symbol can be a number (which isn't a valid identifier start),
|
|
|
|
// so skip it without any checks.
|
|
|
|
self.bump();
|
|
|
|
self.eat_while(is_id_continue);
|
|
|
|
|
|
|
|
// Check if after skipping literal contents we've met a closing
|
|
|
|
// single quote (which means that user attempted to create a
|
|
|
|
// string with single quotes).
|
|
|
|
if self.first() == '\'' {
|
|
|
|
self.bump();
|
|
|
|
let kind = Char { terminated: true };
|
2020-03-20 09:03:11 -05:00
|
|
|
Literal { kind, suffix_start: self.len_consumed() }
|
|
|
|
} else {
|
|
|
|
Lifetime { starts_with_number }
|
2019-05-06 03:53:40 -05:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
fn single_quoted_string(&mut self) -> bool {
|
|
|
|
debug_assert!(self.prev() == '\'');
|
2019-11-03 03:54:23 -06:00
|
|
|
// Check if it's a one-symbol literal.
|
|
|
|
if self.second() == '\'' && self.first() != '\\' {
|
2019-05-06 03:53:40 -05:00
|
|
|
self.bump();
|
2019-11-03 03:54:23 -06:00
|
|
|
self.bump();
|
|
|
|
return true;
|
2019-05-06 03:53:40 -05:00
|
|
|
}
|
2019-11-03 03:54:23 -06:00
|
|
|
|
|
|
|
// Literal has more than one symbol.
|
|
|
|
|
2019-10-26 11:12:58 -05:00
|
|
|
// Parse until either quotes are terminated or error is detected.
|
2019-05-06 03:53:40 -05:00
|
|
|
loop {
|
2019-11-03 02:39:39 -06:00
|
|
|
match self.first() {
|
2019-10-26 11:12:58 -05:00
|
|
|
// Quotes are terminated, finish parsing.
|
2019-05-06 03:53:40 -05:00
|
|
|
'\'' => {
|
|
|
|
self.bump();
|
|
|
|
return true;
|
|
|
|
}
|
2019-11-03 03:54:23 -06:00
|
|
|
// Probably beginning of the comment, which we don't want to include
|
|
|
|
// to the error report.
|
|
|
|
'/' => break,
|
|
|
|
// Newline without following '\'' means unclosed quote, stop parsing.
|
|
|
|
'\n' if self.second() != '\'' => break,
|
|
|
|
// End of file, stop parsing.
|
|
|
|
EOF_CHAR if self.is_eof() => break,
|
2019-10-26 11:12:58 -05:00
|
|
|
// Escaped slash is considered one character, so bump twice.
|
2019-05-06 03:53:40 -05:00
|
|
|
'\\' => {
|
|
|
|
self.bump();
|
|
|
|
self.bump();
|
|
|
|
}
|
2019-10-26 11:12:58 -05:00
|
|
|
// Skip the character.
|
2019-05-06 03:53:40 -05:00
|
|
|
_ => {
|
|
|
|
self.bump();
|
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
2019-11-03 03:54:23 -06:00
|
|
|
// String was not terminated.
|
2019-05-06 03:53:40 -05:00
|
|
|
false
|
|
|
|
}
|
|
|
|
|
2019-10-26 11:12:58 -05:00
|
|
|
/// Eats double-quoted string and returns true
|
|
|
|
/// if string is terminated.
|
2019-05-06 03:53:40 -05:00
|
|
|
fn double_quoted_string(&mut self) -> bool {
|
|
|
|
debug_assert!(self.prev() == '"');
|
2019-11-03 03:55:05 -06:00
|
|
|
while let Some(c) = self.bump() {
|
|
|
|
match c {
|
2019-05-06 03:53:40 -05:00
|
|
|
'"' => {
|
|
|
|
return true;
|
|
|
|
}
|
2019-11-03 03:55:05 -06:00
|
|
|
'\\' if self.first() == '\\' || self.first() == '"' => {
|
|
|
|
// Bump again to skip escaped character.
|
2019-05-06 03:53:40 -05:00
|
|
|
self.bump();
|
|
|
|
}
|
|
|
|
_ => (),
|
|
|
|
}
|
|
|
|
}
|
2019-11-03 03:55:05 -06:00
|
|
|
// End of file reached.
|
|
|
|
false
|
2019-05-06 03:53:40 -05:00
|
|
|
}
|
|
|
|
|
2020-05-29 10:37:16 -05:00
|
|
|
/// Eats the double-quoted string and returns `n_hashes` and an error if encountered.
|
|
|
|
fn raw_double_quoted_string(&mut self, prefix_len: usize) -> (u16, Option<RawStrError>) {
|
|
|
|
// Wrap the actual function to handle the error with too many hashes.
|
|
|
|
// This way, it eats the whole raw string.
|
|
|
|
let (n_hashes, err) = self.raw_string_unvalidated(prefix_len);
|
|
|
|
// Only up to 65535 `#`s are allowed in raw strings
|
|
|
|
match u16::try_from(n_hashes) {
|
|
|
|
Ok(num) => (num, err),
|
|
|
|
// We lie about the number of hashes here :P
|
|
|
|
Err(_) => (0, Some(RawStrError::TooManyDelimiters { found: n_hashes })),
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
fn raw_string_unvalidated(&mut self, prefix_len: usize) -> (usize, Option<RawStrError>) {
|
2019-05-06 03:53:40 -05:00
|
|
|
debug_assert!(self.prev() == 'r');
|
2020-03-28 00:46:20 -05:00
|
|
|
let start_pos = self.len_consumed();
|
2020-05-29 10:37:16 -05:00
|
|
|
let mut possible_terminator_offset = None;
|
|
|
|
let mut max_hashes = 0;
|
2019-11-03 03:55:50 -06:00
|
|
|
|
2019-10-26 11:12:58 -05:00
|
|
|
// Count opening '#' symbols.
|
2020-03-28 00:46:20 -05:00
|
|
|
let n_start_hashes = self.eat_while(|c| c == '#');
|
2019-11-03 03:55:50 -06:00
|
|
|
|
|
|
|
// Check that string is started.
|
|
|
|
match self.bump() {
|
2020-05-29 10:37:16 -05:00
|
|
|
Some('"') => (),
|
|
|
|
c => {
|
|
|
|
let c = c.unwrap_or(EOF_CHAR);
|
|
|
|
return (n_start_hashes, Some(RawStrError::InvalidStarter { bad_char: c }));
|
2020-03-28 00:46:20 -05:00
|
|
|
}
|
2019-11-03 03:55:50 -06:00
|
|
|
}
|
|
|
|
|
|
|
|
// Skip the string contents and on each '#' character met, check if this is
|
|
|
|
// a raw string termination.
|
2020-03-28 00:46:20 -05:00
|
|
|
loop {
|
2019-11-03 03:55:50 -06:00
|
|
|
self.eat_while(|c| c != '"');
|
|
|
|
|
|
|
|
if self.is_eof() {
|
2020-05-29 10:37:16 -05:00
|
|
|
return (
|
2020-03-28 00:46:20 -05:00
|
|
|
n_start_hashes,
|
2020-05-29 10:37:16 -05:00
|
|
|
Some(RawStrError::NoTerminator {
|
|
|
|
expected: n_start_hashes,
|
|
|
|
found: max_hashes,
|
|
|
|
possible_terminator_offset,
|
|
|
|
}),
|
|
|
|
);
|
2019-05-06 03:53:40 -05:00
|
|
|
}
|
|
|
|
|
2019-11-03 03:55:50 -06:00
|
|
|
// Eat closing double quote.
|
|
|
|
self.bump();
|
|
|
|
|
|
|
|
// Check that amount of closing '#' symbols
|
|
|
|
// is equal to the amount of opening ones.
|
2020-04-25 13:19:54 -05:00
|
|
|
// Note that this will not consume extra trailing `#` characters:
|
2020-05-29 10:37:16 -05:00
|
|
|
// `r###"abcde"####` is lexed as a `RawStr { n_hashes: 3 }`
|
2020-04-25 13:19:54 -05:00
|
|
|
// followed by a `#` token.
|
2020-03-28 00:46:20 -05:00
|
|
|
let mut hashes_left = n_start_hashes;
|
2019-11-03 03:55:50 -06:00
|
|
|
let is_closing_hash = |c| {
|
|
|
|
if c == '#' && hashes_left != 0 {
|
|
|
|
hashes_left -= 1;
|
|
|
|
true
|
|
|
|
} else {
|
|
|
|
false
|
2019-05-06 03:53:40 -05:00
|
|
|
}
|
2019-11-03 03:55:50 -06:00
|
|
|
};
|
2020-03-28 00:46:20 -05:00
|
|
|
let n_end_hashes = self.eat_while(is_closing_hash);
|
|
|
|
|
|
|
|
if n_end_hashes == n_start_hashes {
|
2020-05-29 10:37:16 -05:00
|
|
|
return (n_start_hashes, None);
|
2020-03-30 11:39:40 -05:00
|
|
|
} else if n_end_hashes > max_hashes {
|
2020-04-25 13:19:54 -05:00
|
|
|
// Keep track of possible terminators to give a hint about
|
|
|
|
// where there might be a missing terminator
|
2020-03-28 00:46:20 -05:00
|
|
|
possible_terminator_offset =
|
|
|
|
Some(self.len_consumed() - start_pos - n_end_hashes + prefix_len);
|
|
|
|
max_hashes = n_end_hashes;
|
|
|
|
}
|
2019-05-06 03:53:40 -05:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
fn eat_decimal_digits(&mut self) -> bool {
|
|
|
|
let mut has_digits = false;
|
|
|
|
loop {
|
2019-11-03 02:39:39 -06:00
|
|
|
match self.first() {
|
2019-05-06 03:53:40 -05:00
|
|
|
'_' => {
|
|
|
|
self.bump();
|
|
|
|
}
|
|
|
|
'0'..='9' => {
|
|
|
|
has_digits = true;
|
|
|
|
self.bump();
|
|
|
|
}
|
|
|
|
_ => break,
|
|
|
|
}
|
|
|
|
}
|
|
|
|
has_digits
|
|
|
|
}
|
|
|
|
|
|
|
|
fn eat_hexadecimal_digits(&mut self) -> bool {
|
|
|
|
let mut has_digits = false;
|
|
|
|
loop {
|
2019-11-03 02:39:39 -06:00
|
|
|
match self.first() {
|
2019-05-06 03:53:40 -05:00
|
|
|
'_' => {
|
|
|
|
self.bump();
|
|
|
|
}
|
|
|
|
'0'..='9' | 'a'..='f' | 'A'..='F' => {
|
|
|
|
has_digits = true;
|
|
|
|
self.bump();
|
|
|
|
}
|
|
|
|
_ => break,
|
|
|
|
}
|
|
|
|
}
|
|
|
|
has_digits
|
|
|
|
}
|
|
|
|
|
2019-11-03 02:43:47 -06:00
|
|
|
/// Eats the float exponent. Returns true if at least one digit was met,
|
|
|
|
/// and returns false otherwise.
|
|
|
|
fn eat_float_exponent(&mut self) -> bool {
|
2019-05-06 03:53:40 -05:00
|
|
|
debug_assert!(self.prev() == 'e' || self.prev() == 'E');
|
2019-11-03 02:39:39 -06:00
|
|
|
if self.first() == '-' || self.first() == '+' {
|
2019-05-06 03:53:40 -05:00
|
|
|
self.bump();
|
|
|
|
}
|
2019-11-03 02:43:47 -06:00
|
|
|
self.eat_decimal_digits()
|
2019-05-06 03:53:40 -05:00
|
|
|
}
|
|
|
|
|
2019-11-03 02:42:08 -06:00
|
|
|
// Eats the suffix of the literal, e.g. "_u8".
|
2019-05-06 03:53:40 -05:00
|
|
|
fn eat_literal_suffix(&mut self) {
|
2019-11-03 02:42:08 -06:00
|
|
|
self.eat_identifier();
|
|
|
|
}
|
|
|
|
|
|
|
|
// Eats the identifier.
|
|
|
|
fn eat_identifier(&mut self) {
|
|
|
|
if !is_id_start(self.first()) {
|
2019-05-06 03:53:40 -05:00
|
|
|
return;
|
|
|
|
}
|
|
|
|
self.bump();
|
|
|
|
|
2019-11-03 02:42:08 -06:00
|
|
|
self.eat_while(is_id_continue);
|
|
|
|
}
|
|
|
|
|
|
|
|
/// Eats symbols while predicate returns true or until the end of file is reached.
|
|
|
|
/// Returns amount of eaten symbols.
|
|
|
|
fn eat_while<F>(&mut self, mut predicate: F) -> usize
|
|
|
|
where
|
2019-12-22 16:42:04 -06:00
|
|
|
F: FnMut(char) -> bool,
|
2019-11-03 02:42:08 -06:00
|
|
|
{
|
|
|
|
let mut eaten: usize = 0;
|
|
|
|
while predicate(self.first()) && !self.is_eof() {
|
|
|
|
eaten += 1;
|
2019-05-06 03:53:40 -05:00
|
|
|
self.bump();
|
|
|
|
}
|
2019-11-03 02:42:08 -06:00
|
|
|
|
|
|
|
eaten
|
2019-05-06 03:53:40 -05:00
|
|
|
}
|
|
|
|
}
|