feat(lsp): Implement textDocument/foldingRange #9900
Conversation
cli/lsp/tsc.rs
Outdated
| fn adjust_folding_end_line( | ||
| &self, | ||
| range: &lsp::Range, | ||
| line_index: &LineIndex, | ||
| text_content: &str, | ||
| line_folding_only: bool, | ||
| ) -> u32 { | ||
| if line_folding_only && range.end.character > 0 { | ||
| let offset_start: usize = line_index | ||
| .offset(lsp::Position { | ||
| character: range.end.character - 1, | ||
| ..range.end | ||
| }) | ||
| .unwrap() | ||
| .into(); | ||
| let offset_end: usize = line_index.offset(range.end).unwrap().into(); | ||
| let fold_end_char: &str = | ||
| unsafe { text_content.get_unchecked(offset_start..offset_end) }; | ||
| if FOLD_END_PAIR_CHARACTERS.contains(&fold_end_char) { | ||
| return cmp::max(range.end.line - 1, range.start.line); | ||
| } | ||
| } | ||
|
|
||
| range.end.line | ||
| } |
There was a problem hiding this comment.
This was a really ugly hack in TypeScript to work around an issue, and it is even uglier in Rust. The original issue this was added for: microsoft/vscode#47240 appears to be a hack to workaround some folding issues with tsc. Personally I would like to try to remove this and see if the example mentioned in the original issue still persists in recent versions of tsc.
If it is still a problem, I would really like to find a safe way to basically get a single character at a position within a string and then compare that to a vector of static chars, instead of this. All we want is the end offset - 1 and the end offset.
There was a problem hiding this comment.
This is still needed as vscode only supports line folding (lineFoldingOnly is set to true) and typescript returns line and column
/**
* If set, the client signals that it only supports folding complete lines.
* If set, client will ignore specified `startCharacter` and `endCharacter`
* properties in a FoldingRange.
*/
lineFoldingOnly?: boolean;
I would really like to find a safe way to basically get a single character at a position within a string and then compare that to a vector of static chars, instead of this. All we want is the end offset - 1 and the end offset.
This is what I'm trying to do here extracting a valid utf8 code point but I'll try using as_bytes as we are just looking for ASCII characters. I'm just learning rust so if there is a better way let me know.
There was a problem hiding this comment.
If you don't mind, I will take an attempt at it. I'm still learning Rust too, but might have been learning it a bit longer. 😁
There was a problem hiding this comment.
Sure go ahead 👍
In my last commit I managed to reduce it to
let offset_end: usize = line_index.offset(range.end).unwrap().into();
let fold_end_char = text_content.as_bytes()[offset_end - 1];
Related #8643