Auto merge of #7039 - phansch:melt-ice, r=flip1995

tabs_in_doc_comments: Fix ICE due to char indexing

This is a quick-fix for an ICE in `tabs_in_doc_comments`. The problem
was that we we're indexing into possibly multi-byte characters, such as '位'.

More specifically `get_chunks_of_tabs` was returning indices into
multi-byte characters. Those were passed on to a `Span` creation that
then caused the ICE.

This fix makes sure that we don't return indices that point inside a
multi-byte character. *However*, we are still iterating over unicode
codepoints, not grapheme clusters. So a seemingly single character like y̆ ,
which actually consists of two codepoints, will probably still cause
incorrect spans in the output. But I don't think we handle those cases
anywhere in Clippy currently?

Fixes #5835

changelog: Fix ICE in `tabs_in_doc_comments`
This commit is contained in:
bors 2021-04-14 04:53:24 +00:00
commit 24921df7bd
3 changed files with 39 additions and 11 deletions

View file

@ -104,30 +104,32 @@ fn get_chunks_of_tabs(the_str: &str) -> Vec<(u32, u32)> {
// tracker to decide if the last group of tabs is not closed by a non-tab character
let mut is_active = false;
let chars_array: Vec<_> = the_str.chars().collect();
// Note that we specifically need the char _byte_ indices here, not the positional indexes
// within the char array to deal with multi-byte characters properly. `char_indices` does
// exactly that. It provides an iterator over tuples of the form `(byte position, char)`.
let char_indices: Vec<_> = the_str.char_indices().collect();
if chars_array == vec!['\t'] {
if let [(_, '\t')] = char_indices.as_slice() {
return vec![(0, 1)];
}
for (index, arr) in chars_array.windows(2).enumerate() {
let index = u32::try_from(index).expect(line_length_way_to_long);
match arr {
['\t', '\t'] => {
for entry in char_indices.windows(2) {
match entry {
[(_, '\t'), (_, '\t')] => {
// either string starts with double tab, then we have to set it active,
// otherwise is_active is true anyway
is_active = true;
},
[_, '\t'] => {
[(_, _), (index_b, '\t')] => {
// as ['\t', '\t'] is excluded, this has to be a start of a tab group,
// set indices accordingly
is_active = true;
current_start = index + 1;
current_start = u32::try_from(*index_b).unwrap();
},
['\t', _] => {
[(_, '\t'), (index_b, _)] => {
// this now has to be an end of the group, hence we have to push a new tuple
is_active = false;
spans.push((current_start, index + 1));
spans.push((current_start, u32::try_from(*index_b).unwrap()));
},
_ => {},
}
@ -137,7 +139,7 @@ fn get_chunks_of_tabs(the_str: &str) -> Vec<(u32, u32)> {
if is_active {
spans.push((
current_start,
u32::try_from(the_str.chars().count()).expect(line_length_way_to_long),
u32::try_from(char_indices.last().unwrap().0 + 1).expect(line_length_way_to_long),
));
}
@ -148,6 +150,13 @@ fn get_chunks_of_tabs(the_str: &str) -> Vec<(u32, u32)> {
mod tests_for_get_chunks_of_tabs {
use super::get_chunks_of_tabs;
#[test]
fn test_unicode_han_string() {
let res = get_chunks_of_tabs(" \u{4f4d}\t");
assert_eq!(res, vec![(4, 5)]);
}
#[test]
fn test_empty_string() {
let res = get_chunks_of_tabs("");

View file

@ -0,0 +1,9 @@
#[rustfmt::skip]
pub struct Foo {
/// 位
/// ^ Do not remove this tab character.
/// It was required to trigger the ICE.
pub bar: u8,
}
fn main() {}

View file

@ -0,0 +1,10 @@
error: using tabs in doc comments is not recommended
--> $DIR/ice-5835.rs:3:10
|
LL | /// 位
| ^^^^ help: consider using four spaces per tab
|
= note: `-D clippy::tabs-in-doc-comments` implied by `-D warnings`
error: aborting due to previous error