Struct DetailedWorkspaceStats 

pub struct DetailedWorkspaceStats {
    pub total_repos: usize,
    pub indexed_repos: usize,
    pub unindexed_repos: usize,
    pub total_symbols: u64,
    pub freshness: FreshnessBuckets,
    pub avg_symbols_per_repo: f64,
    pub unknown_symbol_count_repos: usize,
}

Detailed workspace statistics including staleness tracking.

Fields

total_repos: usize

Total number of repositories in the workspace.

indexed_repos: usize

Number of repositories that have been indexed.

unindexed_repos: usize

Number of repositories that have never been indexed.

total_symbols: u64

Total symbol count across all indexed repositories whose manifest.json sidecar could be read. Repositories counted in unknown_symbol_count_repos do not contribute here.

freshness: FreshnessBuckets

Freshness buckets categorizing repositories by age.

avg_symbols_per_repo: f64

Average symbols per repository with a known symbol count (total_symbols divided by the repos that contributed to it, not by indexed_repos, so an unreadable manifest cannot silently drag the average down).

unknown_symbol_count_repos: usize

Indexed repositories (last_indexed_at.is_some()) whose .sqry/graph/manifest.json could not be read or parsed, so their symbol count is unknown rather than zero. Surfaced separately so sqry workspace stats can report them instead of silently folding them into total_symbols as zero contributions.

Implementations

impl DetailedWorkspaceStats

pub fn from_registry(registry: &WorkspaceRegistry) -> Self

Compute detailed statistics from a workspace registry.

Reads every indexed member’s symbol count live from its .sqry/graph/manifest.json sidecar (via WorkspaceRepository::symbol_count_from_manifest), not from the registry’s cached symbol_count_at_registration field. See the module-level docs for why: this is the fix for issue #515’s staleness gap (stats reporting a stale count after a direct sqry index --force reindex of a member).

pub fn from_registry_with_resolver( registry: &WorkspaceRegistry, resolve_symbol_count: impl Fn(&WorkspaceRepository) -> Option<u64>, ) -> Self

Compute detailed statistics from a workspace registry using a caller-supplied symbol-count resolver instead of always reading .sqry/graph/manifest.json from disk.

Self::from_registry is the production entry point (used by sqry workspace stats) and always resolves live from each member’s manifest. This variant exists so the aggregation logic (freshness split, total_symbols sum, unknown_symbol_count_repos bucketing, avg_symbols_per_repo denominator) stays unit-testable without needing real manifest fixtures on disk for every test: callers can inject a resolver that returns canned known/unknown counts per repository.

pub fn stale_repos<'a>( &self, registry: &'a WorkspaceRegistry, threshold: Duration, ) -> Vec<&'a WorkspaceRepository>

Get repositories that need reindexing (older than threshold).

pub fn health_score(&self) -> f64

Calculate a health score (0.0-1.0) based on freshness.

Score factors:

• Fresh repos (< 1 hour): 1.0 weight
• Recent repos (< 1 day): 0.8 weight
• Stale repos (< 1 week): 0.5 weight
• Very stale repos (> 1 week): 0.2 weight
• Never indexed: 0.0 weight

pub fn health_status(&self) -> &'static str

Get a human-readable health status.

Trait Implementations

impl Clone for DetailedWorkspaceStats

fn clone(&self) -> DetailedWorkspaceStats

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for DetailedWorkspaceStats

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

Formats the value using the given formatter.