Clip UTF-16 offsets in text for range (#20968)

When launching the Pinyin keyboard, macOS will sometimes try to peek one
character back in the string.

This caused a panic if the preceding character was an emoji. The docs
say
"don't assume the range is valid", so now we don't.

Release Notes:

- (macOS) Fixed a panic when using the Pinyin keyboard with emojis

crates/gpui/src/input.rs

@@ -9,8 +9,12 @@ use std::ops::Range;
 /// See [`InputHandler`] for details on how to implement each method.
 pub trait ViewInputHandler: 'static + Sized {
     /// See [`InputHandler::text_for_range`] for details
-    fn text_for_range(&mut self, range: Range<usize>, cx: &mut ViewContext<Self>)
-        -> Option<String>;
+    fn text_for_range(
+        &mut self,
+        range: Range<usize>,
+        adjusted_range: &mut Option<Range<usize>>,
+        cx: &mut ViewContext<Self>,
+    ) -> Option<String>;
 
     /// See [`InputHandler::selected_text_range`] for details
     fn selected_text_range(
@@ -89,10 +93,12 @@ impl<V: ViewInputHandler> InputHandler for ElementInputHandler<V> {
     fn text_for_range(
         &mut self,
         range_utf16: Range<usize>,
+        adjusted_range: &mut Option<Range<usize>>,
         cx: &mut WindowContext,
     ) -> Option<String> {
-        self.view
-            .update(cx, |view, cx| view.text_for_range(range_utf16, cx))
+        self.view.update(cx, |view, cx| {
+            view.text_for_range(range_utf16, adjusted_range, cx)
+        })
     }
 
     fn replace_text_in_range(

crates/gpui/src/platform.rs

@@ -643,9 +643,13 @@ impl PlatformInputHandler {
     }
 
     #[cfg_attr(any(target_os = "linux", target_os = "freebsd"), allow(dead_code))]
-    fn text_for_range(&mut self, range_utf16: Range<usize>) -> Option<String> {
+    fn text_for_range(
+        &mut self,
+        range_utf16: Range<usize>,
+        adjusted: &mut Option<Range<usize>>,
+    ) -> Option<String> {
         self.cx
-            .update(|cx| self.handler.text_for_range(range_utf16, cx))
+            .update(|cx| self.handler.text_for_range(range_utf16, adjusted, cx))
             .ok()
             .flatten()
     }
@@ -712,6 +716,7 @@ impl PlatformInputHandler {
 
 /// A struct representing a selection in a text buffer, in UTF16 characters.
 /// This is different from a range because the head may be before the tail.
+#[derive(Debug)]
 pub struct UTF16Selection {
     /// The range of text in the document this selection corresponds to
     /// in UTF16 characters.
@@ -749,6 +754,7 @@ pub trait InputHandler: 'static {
     fn text_for_range(
         &mut self,
         range_utf16: Range<usize>,
+        adjusted_range: &mut Option<Range<usize>>,
         cx: &mut WindowContext,
     ) -> Option<String>;
 

crates/gpui/src/platform/mac/window.rs

@@ -38,6 +38,7 @@ use std::{
     cell::Cell,
     ffi::{c_void, CStr},
     mem,
+    ops::Range,
     path::PathBuf,
     ptr::{self, NonNull},
     rc::Rc,
@@ -1754,15 +1755,21 @@ extern "C" fn attributed_substring_for_proposed_range(
     this: &Object,
     _: Sel,
     range: NSRange,
-    _actual_range: *mut c_void,
+    actual_range: *mut c_void,
 ) -> id {
     with_input_handler(this, |input_handler| {
         let range = range.to_range()?;
         if range.is_empty() {
             return None;
         }
+        let mut adjusted: Option<Range<usize>> = None;
 
-        let selected_text = input_handler.text_for_range(range.clone())?;
+        let selected_text = input_handler.text_for_range(range.clone(), &mut adjusted)?;
+        if let Some(adjusted) = adjusted {
+            if adjusted != range {
+                unsafe { (actual_range as *mut NSRange).write(NSRange::from(adjusted)) };
+            }
+        }
         unsafe {
             let string: id = msg_send![class!(NSAttributedString), alloc];
             let string: id = msg_send![string, initWithString: ns_string(&selected_text)];