Native Comparator Scope

DotMatch currently has one native alignment-library comparator: Edlib exhaustive global edit-distance assignment. The generated native report records Edlib through EDLIB_MODE_NW, EDLIB_TASK_DISTANCE, fixed threshold k, and zero assignment mismatches before speedups are reported.

SeqAn and Parasail are not part of the checked README, website, or release-note performance comparison set yet. Before either name belongs in those comparisons, the repository needs all of the following:

  • equivalent global edit-distance or documented semi-global scoring semantics for the exact workload being claimed;

  • fixed threshold k and identical assignment policy for unique, ambiguous, no-match, and invalid reads;

  • native dependency name, version, build flags, and platform in the raw artifact or generated report;

  • raw CSV rows under benchmarks/raw/ plus a generated report under docs/benchmarks/;

  • zero assignment mismatches against DotMatch and the selected comparator;

  • a gate script that fails when only scaffold, smoke, or unmatched-scoring rows are present.

Until that evidence exists, native comparator evidence is limited to Edlib exhaustive global edit-distance assignment scans plus the exact-hash and BK-tree baselines recorded in docs/benchmarks/native/README.md.