Struct WatcherStats 

#[non_exhaustive]
pub struct WatcherStats {
    pub requeries: u64,
    pub changes: u64,
    pub skipped: u64,
    pub retries: u64,
    pub recoveries: u64,
    pub terminal_failures: u64,
    pub last_error: Option<WatcherErrorKind>,
    pub watch_errors: u64,
}

A cheap point-in-time copy of the watcher’s health counters — see RepoWatcher::stats. Lets a long-running consumer notice a watcher that is silently skipping re-queries (e.g. a permanently wedged repository) instead of inferring health from event silence.

Fields (Non-exhaustive)

Non-exhaustive structs could have additional fields added in future. Therefore, non-exhaustive structs cannot be constructed in external crates using the traditional Struct { .. } syntax; cannot be matched against without a wildcard ..; and struct update syntax will not work.

requeries: u64

Re-query attempts started, including automatic retry attempts.

changes: u64

Re-queries that emitted a RepoChange (the rest found no difference).

skipped: u64

Re-queries skipped — transient query failures plus deadline overruns.

retries: u64

Automatic re-query retries scheduled after transient failures/timeouts.

recoveries: u64

Failed re-query sequences that later succeeded during automatic retry.

terminal_failures: u64

Terminal filesystem-watch backend failures. When this increments, the output channel is closed and RepoWatcher::recv returns None.

last_error: Option<WatcherErrorKind>

What the most recent skip failed on; None when nothing was ever skipped.

watch_errors: u64

Filesystem-watch errors reported by the OS backend (via notify). A non-zero — especially climbing — count means the underlying watch is failing: most often the watched .git/.jj directory was removed and re-created (a re-clone / jj git init), which invalidates the OS watch on the old directory. Such a reported error terminates the watch: recv returns None (and the stream ends), so the consumer can rebuild it.

Best-effort, platform-dependent. It is reliable on Windows, where removing the watched directory fails ReadDirectoryChangesW and notify reports an error. On Linux (inotify) a removed/re-created directory may surface as an ordinary event or a silent watch teardown rather than an error, so watch_errors can stay 0 even as the watcher goes deaf — don’t rely on it as the sole liveness signal there.

Trait Implementations

impl Clone for WatcherStats

fn clone(&self) -> WatcherStats

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Copy for WatcherStats

impl Debug for WatcherStats

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

Formats the value using the given formatter.