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.

Sign up for GitHub

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

Jump to bottom

Fix offset conversion for end of file #100

Merged
merged 1 commit into from
Aug 6, 2023
Merged

Fix offset conversion for end of file #100

Changes from all commits
Commits
Show all changes
1 commit
Select commit
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
35 changes: 20 additions & 15 deletions crates/nil/src/vfs.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -153,11 +153,11 @@ impl Vfs {
#[derive(Debug, PartialEq, Eq)]
pub struct LineMap {
/// Invariant:
/// - Have at least two elements.
/// - Have at least one element.
/// - The first must be 0.
/// - The last must be the length of original text.
line_starts: Vec<u32>,
char_diffs: HashMap<u32, Vec<(u32, CodeUnitsDiff)>>,
len: u32,
}

#[derive(Debug, Clone, Copy, PartialEq, Eq)]
Expand All @@ -169,12 +169,12 @@ enum CodeUnitsDiff {
impl LineMap {
fn normalize(mut text: String) -> (String, Self) {
// Must be valid for `TextSize`.
u32::try_from(text.len()).expect("Text too long");
let text_len = u32::try_from(text.len()).expect("Text too long");

text.retain(|c| c != '\r');
let bytes = text.as_bytes();

let mut line_starts = Some(0)
let line_starts = Some(0)
.into_iter()
.chain(
bytes
Expand All @@ -184,10 +184,12 @@ impl LineMap {
.map(|(_, i)| i + 1),
)
.collect::<Vec<_>>();
line_starts.push(text.len() as u32);

let mut char_diffs = HashMap::new();
for ((&start, &end), i) in line_starts.iter().zip(&line_starts[1..]).zip(0u32..) {

let start_pos_iter = line_starts.iter().copied();
let end_pos_iter = line_starts[1..].iter().copied().chain(Some(text_len));
for ((start, end), i) in start_pos_iter.zip(end_pos_iter).zip(0u32..) {
let mut diffs = Vec::new();
for (&b, pos) in bytes[start as usize..end as usize].iter().zip(0u32..) {
let diff = match b {
Expand All @@ -207,12 +209,13 @@ impl LineMap {
let this = Self {
line_starts,
char_diffs,
len: text_len,
};
(text, this)
}

pub fn last_line(&self) -> u32 {
self.line_starts.len() as u32 - 2
self.line_starts.len() as u32 - 1
}

pub fn pos_for_line_col(&self, line: u32, mut col: u32) -> TextSize {
Expand Down Expand Up @@ -245,15 +248,16 @@ impl LineMap {
}

pub fn end_col_for_line(&self, line: u32) -> u32 {
let mut len = self.line_starts[line as usize + 1] - self.line_starts[line as usize];
let mut len = if line + 1 >= self.line_starts.len() as u32 {
self.len - self.line_starts[line as usize]
} else {
// Minus the trailing `\n` for non-last-lines.
self.line_starts[line as usize + 1] - self.line_starts[line as usize] - 1
};

if let Some(diffs) = self.char_diffs.get(&line) {
len -= diffs.iter().map(|&(_, diff)| diff as u32).sum::<u32>();
}
// Lines except the last one has a trailing `\n`.
// Note that `line_starts` has one element more than actual total lines.
if line + 1 + 1 != self.line_starts.len() as u32 {
len -= 1;
}
len
}
}
Expand All @@ -268,7 +272,7 @@ mod tests {
let s = "hello\nworld\nend";
let (norm, map) = LineMap::normalize(s.into());
assert_eq!(norm, s);
assert_eq!(&map.line_starts, &[0, 6, 12, 15]);
assert_eq!(&map.line_starts, &[0, 6, 12]);

let mapping = [
(0, 0, 0),
Expand All @@ -277,6 +281,7 @@ mod tests {
(6, 1, 0),
(11, 1, 5),
(12, 2, 0),
(15, 2, 3),
];
for (pos, line, col) in mapping {
assert_eq!(map.line_col_for_pos(pos.into()), (line, col));
Expand All @@ -294,7 +299,7 @@ mod tests {
let s = "_A_ß_ℝ_💣_";
let (norm, map) = LineMap::normalize(s.into());
assert_eq!(norm, s);
assert_eq!(&map.line_starts, &[0, 15]);
assert_eq!(&map.line_starts, &[0]);
assert_eq!(
&map.char_diffs,
&HashMap::from([(
Expand Down