Struct CompletionContext 

pub struct CompletionContext {
    pub prefix: String,
    pub cursor_byte: usize,
    pub word_start_byte: usize,
    pub buffer_len: usize,
    pub is_large_file: bool,
    pub scan_range: Range<usize>,
    pub viewport_top_byte: usize,
    pub viewport_bottom_byte: usize,
    pub language_id: Option<String>,
    pub word_chars_extra: String,
    pub prefix_has_uppercase: bool,
    pub other_buffers: Vec<OtherBufferSlice>,
}

Context passed to every provider when completion is requested.

All byte ranges are clamped to valid buffer positions by the service before being handed to providers.

Fields

prefix: String

The prefix the user has already typed (from word start to cursor).

cursor_byte: usize

Byte offset of the cursor in the buffer.

word_start_byte: usize

Byte offset where the current word starts (for replacement range).

buffer_len: usize

Total buffer size in bytes.

is_large_file: bool

Whether this buffer is lazily loaded (multi-gigabyte).

scan_range: Range<usize>

The safe scan window: providers MUST NOT read outside this range. For normal files this covers a generous region around the cursor. For huge files this is clamped to a small neighbourhood.

viewport_top_byte: usize

Byte position of the first visible line in the viewport. Useful for proximity scoring—candidates near the viewport rank higher.

viewport_bottom_byte: usize

Approximate byte position of the last visible line.

language_id: Option<String>

The file extension or language id, if known.

word_chars_extra: String

Extra characters (beyond alphanumeric and _) that are considered part of an identifier in the current language.

Examples:

• Lisp/Clojure/CSS: "-" (kebab-case)
• PHP/Bash: "$" (sigils)
• Ruby: "?!"
• Rust (default): "" (only [A-Za-z0-9_])

Populated from LanguageConfig::word_characters if set, otherwise empty (standard alphanumeric + underscore).

prefix_has_uppercase: bool

Whether the prefix contains at least one uppercase character. When true, providers should use smart-case matching: prefer case-sensitive matches, and penalise case mismatches in scoring rather than filtering them out entirely.

other_buffers: Vec<OtherBufferSlice>

Pre-sliced byte windows from other open buffers, ordered by MRU (most recently used first). Enables multi-buffer dabbrev scanning.

Implementations

impl CompletionContext

pub fn compute_scan_range( cursor_byte: usize, buffer_len: usize, is_large_file: bool, ) -> Range<usize>

Compute the scan range for a given cursor position and buffer size.

Trait Implementations

impl Clone for CompletionContext

fn clone(&self) -> CompletionContext

Returns a duplicate of the value.

fn clone_from(&mut self, source: &Self)

Performs copy-assignment from source.

impl Debug for CompletionContext

fn fmt(&self, f: &mut Formatter<'_>) -> Result

Formats the value using the given formatter.