bjoernager/rust
bjoernager/rust
1
Fork 0
Code Activity

shift no characters when using raw string literals

Browse source 
remove `find_skips`

remove unnecessary variables
This commit is contained in:
Takayuki Maeda 2022-11-09 22:40:51 +09:00
parent 0aaad9e757
commit 9857de218f
5 changed files with 115 additions and 88 deletions
Download patch file Download diff file Expand all files Collapse all files

174
compiler/rustc_parse_format/src/lib.rs
View file

@ -818,96 +818,94 @@ fn find_skips_from_snippet(
_ => return (vec![], false), _ => return (vec![], false),
}; };
fn find_skips(snippet: &str, is_raw: bool) -> Vec<usize> { if str_style.is_some() {
let mut s = snippet.char_indices(); return (vec![], true);
let mut skips = vec![];
while let Some((pos, c)) = s.next() {
match (c, s.clone().next()) {
// skip whitespace and empty lines ending in '\\'
('\\', Some((next_pos, '\n'))) if !is_raw => {
skips.push(pos);
skips.push(next_pos);
let _ = s.next();
while let Some((pos, c)) = s.clone().next() {
if matches!(c, ' ' | '\n' | '\t') {
skips.push(pos);
let _ = s.next();
} else {
break;
}
}
}
('\\', Some((next_pos, 'n' | 't' | 'r' | '0' | '\\' | '\'' | '\"'))) => {
skips.push(next_pos);
let _ = s.next();
}
('\\', Some((_, 'x'))) if !is_raw => {
for _ in 0..3 {
// consume `\xAB` literal
if let Some((pos, _)) = s.next() {
skips.push(pos);
} else {
break;
}
}
}
('\\', Some((_, 'u'))) if !is_raw => {
if let Some((pos, _)) = s.next() {
skips.push(pos);
}
if let Some((next_pos, next_c)) = s.next() {
if next_c == '{' {
// consume up to 6 hexanumeric chars
let digits_len =
s.clone().take(6).take_while(|(_, c)| c.is_digit(16)).count();
let len_utf8 = s
.as_str()
.get(..digits_len)
.and_then(|digits| u32::from_str_radix(digits, 16).ok())
.and_then(char::from_u32)
.map_or(1, char::len_utf8);
// Skip the digits, for chars that encode to more than 1 utf-8 byte
// exclude as many digits as it is greater than 1 byte
//
// So for a 3 byte character, exclude 2 digits
let required_skips =
digits_len.saturating_sub(len_utf8.saturating_sub(1));
// skip '{' and '}' also
for pos in (next_pos..).take(required_skips + 2) {
skips.push(pos)
}
s.nth(digits_len);
} else if next_c.is_digit(16) {
skips.push(next_pos);
// We suggest adding `{` and `}` when appropriate, accept it here as if
// it were correct
let mut i = 0; // consume up to 6 hexanumeric chars
while let (Some((next_pos, c)), _) = (s.next(), i < 6) {
if c.is_digit(16) {
skips.push(next_pos);
} else {
break;
}
i += 1;
}
}
}
}
_ => {}
}
}
skips
} }
let r_start = str_style.map_or(0, |r| r + 1); let snippet = &snippet[1..snippet.len() - 1];
let r_end = str_style.unwrap_or(0);
let s = &snippet[r_start + 1..snippet.len() - r_end - 1]; let mut s = snippet.char_indices();
(find_skips(s, str_style.is_some()), true) let mut skips = vec![];
while let Some((pos, c)) = s.next() {
match (c, s.clone().next()) {
// skip whitespace and empty lines ending in '\\'
('\\', Some((next_pos, '\n'))) => {
skips.push(pos);
skips.push(next_pos);
let _ = s.next();
while let Some((pos, c)) = s.clone().next() {
if matches!(c, ' ' | '\n' | '\t') {
skips.push(pos);
let _ = s.next();
} else {
break;
}
}
}
('\\', Some((next_pos, 'n' | 't' | 'r' | '0' | '\\' | '\'' | '\"'))) => {
skips.push(next_pos);
let _ = s.next();
}
('\\', Some((_, 'x'))) => {
for _ in 0..3 {
// consume `\xAB` literal
if let Some((pos, _)) = s.next() {
skips.push(pos);
} else {
break;
}
}
}
('\\', Some((_, 'u'))) => {
if let Some((pos, _)) = s.next() {
skips.push(pos);
}
if let Some((next_pos, next_c)) = s.next() {
if next_c == '{' {
// consume up to 6 hexanumeric chars
let digits_len =
s.clone().take(6).take_while(|(_, c)| c.is_digit(16)).count();
let len_utf8 = s
.as_str()
.get(..digits_len)
.and_then(|digits| u32::from_str_radix(digits, 16).ok())
.and_then(char::from_u32)
.map_or(1, char::len_utf8);
// Skip the digits, for chars that encode to more than 1 utf-8 byte
// exclude as many digits as it is greater than 1 byte
//
// So for a 3 byte character, exclude 2 digits
let required_skips = digits_len.saturating_sub(len_utf8.saturating_sub(1));
// skip '{' and '}' also
for pos in (next_pos..).take(required_skips + 2) {
skips.push(pos)
}
s.nth(digits_len);
} else if next_c.is_digit(16) {
skips.push(next_pos);
// We suggest adding `{` and `}` when appropriate, accept it here as if
// it were correct
let mut i = 0; // consume up to 6 hexanumeric chars
while let (Some((next_pos, c)), _) = (s.next(), i < 6) {
if c.is_digit(16) {
skips.push(next_pos);
} else {
break;
}
i += 1;
}
}
}
}
_ => {}
}
}
(skips, true)
} }
#[cfg(test)] #[cfg(test)]

3
src/test/ui/fmt/format-raw-string-error.rs Normal file
View file

@ -0,0 +1,3 @@
fn main() {
println!(r#"\'\'\'\'\'\'\'\'\'\'\'\'\'\'}"#); //~ ERROR invalid format string: unmatched `}` found
}

10
src/test/ui/fmt/format-raw-string-error.stderr Normal file
View file

@ -0,0 +1,10 @@
error: invalid format string: unmatched `}` found
--> $DIR/format-raw-string-error.rs:2:45
|
LL | println!(r#"\'\'\'\'\'\'\'\'\'\'\'\'\'\'}"#);
| ^ unmatched `}` in format string
|
= note: if you intended to print `}`, you can escape it using `}}`
error: aborting due to previous error

6
src/test/ui/fmt/issue-104142.rs Normal file
View file

@ -0,0 +1,6 @@
fn main() {
println!(
r#"
\"\'}､"# //~ ERROR invalid format string: unmatched `}` found
);
}

10
src/test/ui/fmt/issue-104142.stderr Normal file
View file

@ -0,0 +1,10 @@
error: invalid format string: unmatched `}` found
--> $DIR/issue-104142.rs:4:9
|
LL | \"\'}､"#
| ^ unmatched `}` in format string
|
= note: if you intended to print `}`, you can escape it using `}}`
error: aborting due to previous error