@captiveorigins/onetaxonomy-core

The scientific core of OneTaxonomy: taxon concepts, names, classifications and phylogenetic hypotheses, with validation, bounded traversal, MRCA and tri-state monophyly. No dependencies; no DOM, Node, React, CMS or database. It runs in browsers, Node, workers and desktop webviews.

import { assessTaxon, findMRCA, indexDataset, isMonophyletic, sealSnapshot, validateDataset } from '@captiveorigins/onetaxonomy-core'

const problems = validateDataset(snapshot)          // diagnostics; nothing is repaired silently
const sealed = sealSnapshot(snapshot)               // checksummed and frozen
const data = indexDataset(sealed)

const tree = data.phylogeny('archelosauria')        // a rooted hypothesis
findMRCA(tree, ['t-testudines', 't-aves'])          // { status: 'found', node: 'archelosauria', hypothesis }
isMonophyletic(tree, ['t-crocodylia', 't-aves'])    // { status: 'monophyletic', reason: 'complete-clade', … }

// A classification taxon against a hypothesis, through explicit concept mappings only:
assessTaxon(data.classification('traditional'), 'c:reptilia', tree)
// { status: 'not-monophyletic', missing: ['t-aves'], … }
assessTaxon(data.classification('traditional'), 'c:reptilia', null)
// { status: 'indeterminate', reason: 'classification-only' }

• Results are relative to one hypothesis ({ id, version }) and carry caveats (partial-sampling, unmapped-members, …).
• Unrooted trees give no MRCA or clade claims; forests get no invented common ancestor; polytomies are unresolved (indeterminate) unless you pass { polytomies: 'hard' }.
• Traversal (getChildren, getAncestors, getLineage, getDescendants) is iterative and accepts maxDepth / limit, returning truncated when cut.

Curation, coverage and releases

import { applySelection, diffSnapshots, extractSubset, getCoverageReport, referenceImpact } from '@captiveorigins/onetaxonomy-core'

const result = applySelection(data, selection)            // curated + ancestor paths, omitted branches, diagnostics
const coverage = getCoverageReport(data, {
  classification: selection.classification, included: result, phylogeny: 'my-tree',
  knownToScience: { [turtles]: { count: describedSpecies, rank: 'species', source: 'the checklist you cite' } },
})                                                        // dataset · hypothesis · known to science (only as you state it)
const subset = extractSubset(snapshot, { id, version, createdAt, classification: selection.classification, include: result })
const changes = referenceImpact(diffSnapshots(oldRelease, newRelease), mySavedIds) // renamed, relocated, split/merge candidates…

schema/snapshot.schema.json (exported as @captiveorigins/onetaxonomy-core/schema/snapshot.schema.json) describes the snapshot format for non-TypeScript tools.

See the data schema and invariants.