Expand description
Orchestration of an analysis: input → probe → normalisation → counting → statistics → summaries and provenance. Each step lives in its own module; this one only wires them.
Reading runs on its own thread, one batch ahead of normalisation and counting, which run in
parallel on a local rayon pool sized by RunControl::thread_count, or smaller when the
system refuses some of its threads (D110). At most two batches are in memory at a time.
Structs§
- Analysis
- The outcome of
analyze. - Execution
- Execution facts of a run (not part of the deterministic results).
Functions§
- analyze
- Analyses
sourcewithparams.