Struct QualitySample 

pub struct QualitySample { /* private fields */ }

The rows a sampled run read, kept beside its results: an acquisition.

What decides these rows is the acquisition’s identity — the dataset, the view, the scope, the method, the size and the seed — which the caller keys it by. Everything else a plan says (grain, comparison, time roles, text read as time, the latency threshold) is the report’s, and a run that changes only those cuts these rows again rather than reading the source. Every column of the scope is kept, and where each row sat, so any role, format or row-chunk grain finds what it needs here.

Implementations

impl QualitySample

pub fn needs_segment_count(&self, plan: &DataQualityPlan) -> bool

Whether plan’s segments need a count this sample does not hold: a partition or time-window grain on a sample, counted neither while sampling nor by an earlier run, nor summed from a finer count.

pub fn segment_count(&self, plan: &DataQualityPlan) -> SegmentCount

Where a run of plan over these rows gets its segment totals.

pub fn df(&self) -> &DataFrame

The rows themselves, as the sample every tool reads.

pub fn estimated_bytes(&self) -> usize

Memory the rows, their positions and their counts hold, near enough to budget by.

pub fn analysis_rows(&self, df: DataFrame) -> AnalysisRows

df, cut from these rows, described as the sampler described them.

Trait Implementations

impl Clone for QualitySample

fn clone(&self) -> Self

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for QualitySample

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

Formats the value using the given formatter.