Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

feat: Implement raw string literals #3556

Merged
merged 10 commits into from
Nov 28, 2023
Merged
Show file tree
Hide file tree
Changes from 3 commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
10 changes: 10 additions & 0 deletions compiler/noirc_frontend/src/ast/expression.rs
Original file line number Diff line number Diff line change
Expand Up @@ -76,6 +76,10 @@ impl ExpressionKind {
ExpressionKind::Literal(Literal::Str(contents))
}

pub fn raw_string(contents: String, hashes: u8) -> ExpressionKind {
ExpressionKind::Literal(Literal::RawStr(contents, hashes))
}

pub fn format_string(contents: String) -> ExpressionKind {
ExpressionKind::Literal(Literal::FmtStr(contents))
}
Expand Down Expand Up @@ -312,6 +316,7 @@ pub enum Literal {
Bool(bool),
Integer(FieldElement),
Str(String),
RawStr(String, u8),
FmtStr(String),
Unit,
}
Expand Down Expand Up @@ -507,6 +512,11 @@ impl Display for Literal {
Literal::Bool(boolean) => write!(f, "{}", if *boolean { "true" } else { "false" }),
Literal::Integer(integer) => write!(f, "{}", integer.to_u128()),
Literal::Str(string) => write!(f, "\"{string}\""),
Literal::RawStr(string, num_hashes) => {
let hashes: String =
std::iter::once('#').cycle().take(*num_hashes as usize).collect();
write!(f, "r{hashes}\"{string}\"{hashes}")
}
Literal::FmtStr(string) => write!(f, "f\"{string}\""),
Literal::Unit => write!(f, "()"),
}
Expand Down
1 change: 1 addition & 0 deletions compiler/noirc_frontend/src/hir/resolution/resolver.rs
Original file line number Diff line number Diff line change
Expand Up @@ -95,7 +95,7 @@

/// True if the current module is a contract.
/// This is usually determined by self.path_resolver.module_id(), but it can
/// be overriden for impls. Impls are an odd case since the methods within resolve

Check warning on line 98 in compiler/noirc_frontend/src/hir/resolution/resolver.rs

View workflow job for this annotation

GitHub Actions / Spellcheck / Spellcheck

Unknown word (overriden)
/// as if they're in the parent module, but should be placed in a child module.
/// Since they should be within a child module, in_contract is manually set to false
/// for these so we can still resolve them in the parent module without them being in a contract.
Expand Down Expand Up @@ -1203,6 +1203,7 @@
}
Literal::Integer(integer) => HirLiteral::Integer(integer),
Literal::Str(str) => HirLiteral::Str(str),
Literal::RawStr(str, _) => HirLiteral::Str(str),
Literal::FmtStr(str) => self.resolve_fmt_str_literal(str, expr.span),
Literal::Unit => HirLiteral::Unit,
}),
Expand Down
75 changes: 75 additions & 0 deletions compiler/noirc_frontend/src/lexer/lexer.rs
Original file line number Diff line number Diff line change
Expand Up @@ -16,7 +16,7 @@
position: Position,
done: bool,
skip_comments: bool,
skip_whitespaces: bool,

Check warning on line 19 in compiler/noirc_frontend/src/lexer/lexer.rs

View workflow job for this annotation

GitHub Actions / Spellcheck / Spellcheck

Unknown word (whitespaces)
}

pub type SpannedTokenResult = Result<SpannedToken, LexerErrorKind>;
Expand All @@ -43,7 +43,7 @@
position: 0,
done: false,
skip_comments: true,
skip_whitespaces: true,

Check warning on line 46 in compiler/noirc_frontend/src/lexer/lexer.rs

View workflow job for this annotation

GitHub Actions / Spellcheck / Spellcheck

Unknown word (whitespaces)
}
}

Expand All @@ -52,8 +52,8 @@
self
}

pub fn skip_whitespaces(mut self, flag: bool) -> Self {

Check warning on line 55 in compiler/noirc_frontend/src/lexer/lexer.rs

View workflow job for this annotation

GitHub Actions / Spellcheck / Spellcheck

Unknown word (whitespaces)
self.skip_whitespaces = flag;

Check warning on line 56 in compiler/noirc_frontend/src/lexer/lexer.rs

View workflow job for this annotation

GitHub Actions / Spellcheck / Spellcheck

Unknown word (whitespaces)
self
}

Expand Down Expand Up @@ -96,7 +96,7 @@
match self.next_char() {
Some(x) if x.is_whitespace() => {
let spanned = self.eat_whitespace(x);
if self.skip_whitespaces {

Check warning on line 99 in compiler/noirc_frontend/src/lexer/lexer.rs

View workflow job for this annotation

GitHub Actions / Spellcheck / Spellcheck

Unknown word (whitespaces)
self.next_token()
} else {
Ok(spanned)
Expand Down Expand Up @@ -126,6 +126,7 @@
Some(']') => self.single_char_token(Token::RightBracket),
Some('"') => self.eat_string_literal(),
Some('f') => self.eat_format_string_or_alpha_numeric(),
Some('r') => self.eat_raw_string_or_alpha_numeric(),
Some('#') => self.eat_attribute(),
Some(ch) if ch.is_ascii_alphanumeric() || ch == '_' => self.eat_alpha_numeric(ch),
Some(ch) => {
Expand Down Expand Up @@ -400,6 +401,80 @@
}
}

fn eat_raw_string(&mut self) -> SpannedTokenResult {
let start = self.position;

let beginning_hashes = self.eat_while(None, |ch| ch == '#');
let beginning_hashes_count = beginning_hashes.chars().count();
if beginning_hashes_count > 255 {
// too many hashes (unlikely in practice)
// also, Rust disallows 256+ hashes as well
return Err(LexerErrorKind::UnexpectedCharacter {
span: Span::single_char(self.position + 255),
found: Some('#'),
expected: "\"".to_owned(),
});
}

if !self.peek_char_is('"') {
return Err(LexerErrorKind::UnexpectedCharacter {
span: Span::single_char(self.position),
found: self.next_char(),
expected: "\"".to_owned(),
});
}
self.next_char();

let mut str_literal = String::new();
loop {
let chars = self.eat_while(None, |ch| ch != '"');
str_literal.push_str(&chars[..]);
if !self.peek_char_is('"') {
return Err(LexerErrorKind::UnexpectedCharacter {
span: Span::single_char(self.position),
found: self.next_char(),
expected: "\"".to_owned(),
});
}
self.next_char();
let mut ending_hashes_count = 0;
while let Some('#') = self.peek_char() {
if ending_hashes_count == beginning_hashes_count {
break;
}
self.next_char();
ending_hashes_count += 1;
}
if ending_hashes_count == beginning_hashes_count {
break;
} else {
str_literal.push('"');
for _ in 0..ending_hashes_count {
str_literal.push('#');
}
}
}

let str_literal_token = Token::RawStr(str_literal, beginning_hashes_count as u8);

let end = self.position;
Ok(str_literal_token.into_span(start, end))
}

fn eat_raw_string_or_alpha_numeric(&mut self) -> SpannedTokenResult {
// Problem: we commit to eating raw strings once we see one or two characters.
// This is unclean, but likely ok in all practical cases, and works with existing
// `Lexer` methods.
let peek2 = self.peek2_char();
if (self.peek_char_is('#') && (peek2 == Some('#') || peek2 == Some('"')))
jfecher marked this conversation as resolved.
Show resolved Hide resolved
|| self.peek_char_is('"')
{
self.eat_raw_string()
} else {
self.eat_alpha_numeric('r')
}
}

fn parse_comment(&mut self, start: u32) -> SpannedTokenResult {
let doc_style = match self.peek_char() {
Some('!') => {
Expand Down
11 changes: 10 additions & 1 deletion compiler/noirc_frontend/src/lexer/token.rs
Original file line number Diff line number Diff line change
Expand Up @@ -15,6 +15,7 @@ pub enum Token {
Int(FieldElement),
Bool(bool),
Str(String),
RawStr(String, u8),
FmtStr(String),
Keyword(Keyword),
IntType(IntType),
Expand Down Expand Up @@ -157,6 +158,10 @@ impl fmt::Display for Token {
Token::Bool(b) => write!(f, "{b}"),
Token::Str(ref b) => write!(f, "{b}"),
Token::FmtStr(ref b) => write!(f, "f{b}"),
Token::RawStr(ref b, hashes) => {
let h: String = std::iter::once('#').cycle().take(hashes as usize).collect();
write!(f, "r{h}\"{b}\"{h}")
}
Token::Keyword(k) => write!(f, "{k}"),
Token::Attribute(ref a) => write!(f, "{a}"),
Token::LineComment(ref s, _style) => write!(f, "//{s}"),
Expand Down Expand Up @@ -227,7 +232,11 @@ impl Token {
pub fn kind(&self) -> TokenKind {
match *self {
Token::Ident(_) => TokenKind::Ident,
Token::Int(_) | Token::Bool(_) | Token::Str(_) | Token::FmtStr(_) => TokenKind::Literal,
Token::Int(_)
| Token::Bool(_)
| Token::Str(_)
| Token::RawStr(..)
| Token::FmtStr(_) => TokenKind::Literal,
Token::Keyword(_) => TokenKind::Keyword,
Token::Attribute(_) => TokenKind::Attribute,
ref tok => TokenKind::Token(tok.clone()),
Expand Down
20 changes: 20 additions & 0 deletions compiler/noirc_frontend/src/parser/parser.rs
Original file line number Diff line number Diff line change
Expand Up @@ -1657,6 +1657,7 @@ fn literal() -> impl NoirParser<ExpressionKind> {
Token::Int(x) => ExpressionKind::integer(x),
Token::Bool(b) => ExpressionKind::boolean(b),
Token::Str(s) => ExpressionKind::string(s),
Token::RawStr(s, hashes) => ExpressionKind::raw_string(s, hashes),
Token::FmtStr(s) => ExpressionKind::format_string(s),
unexpected => unreachable!("Non-literal {} parsed as a literal", unexpected),
})
Expand Down Expand Up @@ -2549,4 +2550,23 @@ mod test {

check_cases_with_errors(&cases[..], block(fresh_statement()));
}

#[test]
fn parse_raw_string() {
let cases = vec![
Case { source: r##" r"foo" "##, expect: r##"r"foo""##, errors: 0 },
Case { source: r##" r#"foo"# "##, expect: r##"r#"foo"#"##, errors: 0 },
// backslash
Case { source: r##" r"\\" "##, expect: r##"r"\\""##, errors: 0 },
Case { source: r##" r#"\"# "##, expect: r##"r#"\"#"##, errors: 0 },
Case { source: r##" r#"\\"# "##, expect: r##"r#"\\"#"##, errors: 0 },
Case { source: r##" r#"\\\"# "##, expect: r##"r#"\\\"#"##, errors: 0 },
// mismatch - errors:
Case { source: r###" r#"foo"## "###, expect: r###"r#"foo"#"###, errors: 1 },
Case { source: r###" r##"foo"# "###, expect: "(none)", errors: 2 },
];

check_cases_with_errors(&cases[..], expression());
check_cases_with_errors(&cases[..], literal());
}
}
13 changes: 13 additions & 0 deletions docs/docs/language_concepts/data_types/03_strings.md
Original file line number Diff line number Diff line change
Expand Up @@ -16,13 +16,13 @@
The string type is a fixed length value defined with `str<N>`.

You can use strings in `assert()` functions or print them with
`std::println()`. See more about [Logging](../../standard_library/logging).

Check warning on line 19 in docs/docs/language_concepts/data_types/03_strings.md

View workflow job for this annotation

GitHub Actions / Spellcheck / Spellcheck

Unknown word (println)

```rust
use dep::std;

fn main(message : pub str<11>, hex_as_string : str<4>) {
std::println(message);

Check warning on line 25 in docs/docs/language_concepts/data_types/03_strings.md

View workflow job for this annotation

GitHub Actions / Spellcheck / Spellcheck

Unknown word (println)
assert(message == "hello world");
assert(hex_as_string == "0x41");
}
Expand Down Expand Up @@ -61,3 +61,16 @@
let s = "Hello \"world" // prints "Hello "world"
let s = "hey \tyou"; // prints "hey you"
```

## Raw strings

A raw string begins with the letter `r` and is optionally delimited by a number of hashes `#`.

Escape characters are *not* processed within raw strings. All contents are interpreted literally.

Example:

```rust
let s = r"Hello world";
let s = r#"Simon says "hello world""#;
jfecher marked this conversation as resolved.
Show resolved Hide resolved
```
Original file line number Diff line number Diff line change
@@ -0,0 +1,5 @@
[package]
name = "raw_string_huge"
type = "bin"
authors = [""]
[dependencies]
TomAFrench marked this conversation as resolved.
Show resolved Hide resolved
Original file line number Diff line number Diff line change
@@ -0,0 +1,3 @@
fn main() {
let _a = r##############################################################################################################################################################################################################################################################################"hello"##############################################################################################################################################################################################################################################################################;
}
Original file line number Diff line number Diff line change
@@ -0,0 +1,6 @@
[package]
name = "unit"
TomAFrench marked this conversation as resolved.
Show resolved Hide resolved
type = "bin"
authors = [""]

[dependencies]
Original file line number Diff line number Diff line change
@@ -0,0 +1,10 @@
global D = r#####"Hello "world""#####;

fn main() {
let a = "Hello \"world\"";
let b = r#"Hello "world""#;
let c = r##"Hello "world""##;
assert(a == b);
jfecher marked this conversation as resolved.
Show resolved Hide resolved
assert(b == c);
assert(c == D);
}
Loading