make diagnostics stick to word boundaries

Diagnostics are currently extended if text is inserted at their end. This is
desirable when inserting text after an identifier. For example consider:

let foo = 2;
    --- unused variable

Renaming the identifier should extend the diagnostic:

let foobar = 2;
    ------ unused variable

This is currently implemented in helix but as a consequence adding whitespaces
or a type hint also extends the diagnostic:

let foo      = 2;
    -------- unused variable
let foo: Bar = 2;
    -------- unused variable

In these cases the diagnostic should remain unchanged:

let foo      = 2;
    --- unused variable
let foo: Bar = 2;
    --- unused variable

As a heuristic helix will now only extend diagnostics that end on a word char
if new chars are appended to the word (so not for punctuation/ whitespace).
The idea for this mapping was inspired for the word level tracking vscode uses
for many positions. While VSCode doesn't currently update diagnostics after
receiving publishDiagnostic it does use this system for inlay hints for example.
Similarly, the new association mechanism implemented here can be used for word
level tracking of inlay hints.

A similar mapping function is implemented for word starts. Together
these can be used to make a diagnostic stick to a word. If that word
is removed that diagnostic is automatically removed too. This is the exact
same behavior VSCode inlay hints eixibit.

helix-core/src/diagnostic.rs

@@ -39,6 +39,10 @@ pub enum DiagnosticTag {
 #[derive(Debug, Clone)]
 pub struct Diagnostic {
     pub range: Range,
+    // whether this diagnostic ends at the end of(or inside) a word
+    pub ends_at_word: bool,
+    pub starts_at_word: bool,
+    pub zero_width: bool,
     pub line: usize,
     pub message: String,
     pub severity: Option<Severity>,

helix-core/src/transaction.rs

@@ -1,7 +1,7 @@
 use ropey::RopeSlice;
 use smallvec::SmallVec;
 
-use crate::{Range, Rope, Selection, Tendril};
+use crate::{chars::char_is_word, Range, Rope, Selection, Tendril};
 use std::{borrow::Cow, iter::once};
 
 /// (from, to, replacement)
@@ -23,6 +23,30 @@ pub enum Operation {
 pub enum Assoc {
     Before,
     After,
+    /// Acts like `After` if a word character is inserted
+    /// after the position, otherwise acts like `Before`
+    AfterWord,
+    /// Acts like `Before` if a word character is inserted
+    /// before the position, otherwise acts like `After`
+    BeforeWord,
+}
+
+impl Assoc {
+    /// Whether to stick to gaps
+    fn stay_at_gaps(self) -> bool {
+        !matches!(self, Self::BeforeWord | Self::AfterWord)
+    }
+
+    fn insert_offset(self, s: &str) -> usize {
+        let chars = s.chars().count();
+        match self {
+            Assoc::After => chars,
+            Assoc::AfterWord => s.chars().take_while(|&c| char_is_word(c)).count(),
+            // return position before inserted text
+            Assoc::Before => 0,
+            Assoc::BeforeWord => chars - s.chars().rev().take_while(|&c| char_is_word(c)).count(),
+        }
+    }
 }
 
 #[derive(Debug, Default, Clone, PartialEq, Eq)]
@@ -415,8 +439,6 @@ impl ChangeSet {
                     map!(|pos, _| (old_end > pos).then_some(new_pos), i);
                 }
                 Insert(s) => {
-                    let ins = s.chars().count();
-
                     // a subsequent delete means a replace, consume it
                     if let Some((_, Delete(len))) = iter.peek() {
                         iter.next();
@@ -424,13 +446,13 @@ impl ChangeSet {
                         old_end = old_pos + len;
                         // in range of replaced text
                         map!(
-                            |pos, assoc| (old_end > pos).then(|| {
+                            |pos, assoc: Assoc| (old_end > pos).then(|| {
                                 // at point or tracking before
-                                if pos == old_pos || assoc == Assoc::Before {
+                                if pos == old_pos && assoc.stay_at_gaps() {
                                     new_pos
                                 } else {
                                     // place to end of insert
-                                    new_pos + ins
+                                    new_pos + assoc.insert_offset(s)
                                 }
                             }),
                             i
@@ -438,20 +460,15 @@ impl ChangeSet {
                     } else {
                         // at insert point
                         map!(
-                            |pos, assoc| (old_pos == pos).then(|| {
+                            |pos, assoc: Assoc| (old_pos == pos).then(|| {
                                 // return position before inserted text
-                                if assoc == Assoc::Before {
+                                new_pos + assoc.insert_offset(s)
-                                    new_pos
-                                } else {
-                                    // after text
-                                    new_pos + ins
-                                }
                             }),
                             i
                         );
                     }
 
-                    new_pos += ins;
+                    new_pos += s.chars().count();
                 }
             }
             old_pos = old_end;
@@ -884,6 +901,48 @@ mod test {
         let mut positions = [4, 2];
         cs.update_positions(positions.iter_mut().map(|pos| (pos, Assoc::After)));
         assert_eq!(positions, [4, 2]);
+        // stays at word boundary
+        let cs = ChangeSet {
+            changes: vec![
+                Retain(2), // <space><space>
+                Insert(" ab".into()),
+                Retain(2), // cd
+                Insert("de ".into()),
+            ],
+            len: 4,
+            len_after: 10,
+        };
+        assert_eq!(cs.map_pos(2, Assoc::BeforeWord), 3);
+        assert_eq!(cs.map_pos(4, Assoc::AfterWord), 9);
+        let cs = ChangeSet {
+            changes: vec![
+                Retain(1), // <space>
+                Insert(" b".into()),
+                Delete(1), // c
+                Retain(1), // d
+                Insert("e ".into()),
+                Delete(1), // <space>
+            ],
+            len: 5,
+            len_after: 7,
+        };
+        assert_eq!(cs.map_pos(1, Assoc::BeforeWord), 2);
+        assert_eq!(cs.map_pos(3, Assoc::AfterWord), 5);
+        let cs = ChangeSet {
+            changes: vec![
+                Retain(1), // <space>
+                Insert("a".into()),
+                Delete(2), // <space>b
+                Retain(1), // d
+                Insert("e".into()),
+                Delete(1), // f
+                Retain(1), // <space>
+            ],
+            len: 5,
+            len_after: 7,
+        };
+        assert_eq!(cs.map_pos(2, Assoc::BeforeWord), 1);
+        assert_eq!(cs.map_pos(4, Assoc::AfterWord), 4);
     }
 
     #[test]

helix-term/src/application.rs

@@ -1,6 +1,7 @@
 use arc_swap::{access::Map, ArcSwap};
 use futures_util::Stream;
 use helix_core::{
+    chars::char_is_word,
     diagnostic::{DiagnosticTag, NumberOrString},
     path::get_relative_path,
     pos_at_coords, syntax, Selection,
@@ -811,7 +812,6 @@ impl Application {
                                         log::warn!("lsp position out of bounds - {:?}", diagnostic);
                                         return None;
                                     };
-
                                     let severity =
                                         diagnostic.severity.map(|severity| match severity {
                                             DiagnosticSeverity::ERROR => Error,
@@ -863,8 +863,17 @@ impl Application {
                                         Vec::new()
                                     };
 
+                                    let ends_at_word = start != end
+                                        && end != 0
+                                        && text.get_char(end - 1).map_or(false, char_is_word);
+                                    let starts_at_word = start != end
+                                        && text.get_char(start).map_or(false, char_is_word);
+
                                     Some(Diagnostic {
                                         range: Range { start, end },
+                                        ends_at_word,
+                                        starts_at_word,
+                                        zero_width: start == end,
                                         line: diagnostic.range.start.line as usize,
                                         message: diagnostic.message.clone(),
                                         severity,

helix-view/src/document.rs

@@ -1213,20 +1213,32 @@ impl Document {
 
             let changes = transaction.changes();
 
-            changes.update_positions(
+            // map diagnostics over changes too
-                self.diagnostics
+            changes.update_positions(self.diagnostics.iter_mut().map(|diagnostic| {
-                    .iter_mut()
+                let assoc = if diagnostic.starts_at_word {
-                    .map(|diagnostic| (&mut diagnostic.range.start, Assoc::After)),
+                    Assoc::BeforeWord
-            );
+                } else {
-            changes.update_positions(
+                    Assoc::After
-                self.diagnostics
+                };
-                    .iter_mut()
+                (&mut diagnostic.range.start, assoc)
-                    .map(|diagnostic| (&mut diagnostic.range.end, Assoc::After)),
+            }));
-            );
+            changes.update_positions(self.diagnostics.iter_mut().map(|diagnostic| {
-            // map state.diagnostics over changes::map_pos too
+                let assoc = if diagnostic.ends_at_word {
-            for diagnostic in &mut self.diagnostics {
+                    Assoc::AfterWord
-                diagnostic.line = self.text.char_to_line(diagnostic.range.start);
+                } else {
+                    Assoc::Before
+                };
+                (&mut diagnostic.range.end, assoc)
+            }));
+            self.diagnostics.retain_mut(|diagnostic| {
+                if diagnostic.range.start > diagnostic.range.end
+                    || (!diagnostic.zero_width && diagnostic.range.start == diagnostic.range.end)
+                {
+                    return false;
                 }
+                diagnostic.line = self.text.char_to_line(diagnostic.range.start);
+                true
+            });
 
             self.diagnostics
                 .sort_unstable_by_key(|diagnostic| diagnostic.range);