sui_sql_macro/lexer.rs
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
// Copyright (c) Mysten Labs, Inc.
// SPDX-License-Identifier: Apache-2.0
use std::fmt;
/// Lexer for SQL format strings. Format string can contain regular text, or binders surrounded by
/// curly braces. Curly braces are escaped by doubling them up.
pub(crate) struct Lexer<'s> {
src: &'s str,
off: usize,
}
/// A lexeme is a token along with its offset in the source string, and the slice of source string
/// that it originated from.
#[derive(Clone, Copy, Debug, PartialEq, Eq)]
pub(crate) struct Lexeme<'s>(pub Token, pub usize, pub &'s str);
#[derive(Clone, Copy, Debug, PartialEq, Eq)]
pub(crate) enum Token {
/// '{'
LCurl,
/// '}'
RCurl,
/// Any other text
Text,
}
impl<'s> Lexer<'s> {
pub(crate) fn new(src: &'s str) -> Self {
Self { src, off: 0 }
}
}
impl<'s> Iterator for Lexer<'s> {
type Item = Lexeme<'s>;
fn next(&mut self) -> Option<Self::Item> {
let off = self.off;
let bytes = self.src.as_bytes();
let fst = bytes.first()?;
Some(match fst {
b'{' => {
let span = &self.src[..1];
self.src = &self.src[1..];
self.off += 1;
Lexeme(Token::LCurl, off, span)
}
b'}' => {
let span = &self.src[..1];
self.src = &self.src[1..];
self.off += 1;
Lexeme(Token::RCurl, off, span)
}
_ => {
let end = self.src.find(['{', '}']).unwrap_or(self.src.len());
let span = &self.src[..end];
self.src = &self.src[end..];
self.off += end;
Lexeme(Token::Text, off, span)
}
})
}
}
impl fmt::Display for Token {
fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result {
use Token as T;
match self {
T::LCurl => write!(f, "'{{'"),
T::RCurl => write!(f, "'}}'"),
T::Text => write!(f, "text"),
}
}
}
#[cfg(test)]
mod tests {
use super::*;
use Lexeme as L;
use Token as T;
/// Lexing source material that only contains text and no curly braces.
#[test]
fn test_all_text() {
let lexer = Lexer::new("foo bar");
let lexemes: Vec<_> = lexer.collect();
assert_eq!(lexemes, vec![L(T::Text, 0, "foo bar")]);
}
/// When the lexer encounters curly braces in the source material it breaks up the text with
/// curly brace tokens.
#[test]
fn test_curlies() {
let lexer = Lexer::new("foo {bar} baz");
let lexemes: Vec<_> = lexer.collect();
assert_eq!(
lexemes,
vec![
L(T::Text, 0, "foo "),
L(T::LCurl, 4, "{"),
L(T::Text, 5, "bar"),
L(T::RCurl, 8, "}"),
L(T::Text, 9, " baz"),
],
);
}
/// Repeated curly braces next to each other are used to escape those braces.
#[test]
fn test_escape_curlies() {
let lexer = Lexer::new("foo {{bar}} baz");
let lexemes: Vec<_> = lexer.collect();
assert_eq!(
lexemes,
vec![
L(T::Text, 0, "foo "),
L(T::LCurl, 4, "{"),
L(T::LCurl, 5, "{"),
L(T::Text, 6, "bar"),
L(T::RCurl, 9, "}"),
L(T::RCurl, 10, "}"),
L(T::Text, 11, " baz"),
],
);
}
/// Each curly brace is given its own token so that the parser can parse this as an escaped
/// opening curly followed by an empty binder, followed by a literal closing curly. If the
/// lexer was responsible for detecting escaped curlies, it would eagerly detect the escaped
/// closing curly and then the closing curly for the binder.
#[test]
fn test_combination_curlies() {
let lexer = Lexer::new("{{{}}}");
let lexemes: Vec<_> = lexer.collect();
assert_eq!(
lexemes,
vec![
L(T::LCurl, 0, "{"),
L(T::LCurl, 1, "{"),
L(T::LCurl, 2, "{"),
L(T::RCurl, 3, "}"),
L(T::RCurl, 4, "}"),
L(T::RCurl, 5, "}"),
],
);
}
}