@larsgw/formica 0.7.2 → 0.8.0

package/src/bin/generate-linked-data.ts

@@ -0,0 +1,709 @@
+#!/usr/bin/env node
+
+import { existsSync as doesFileExist, promises as fs } from 'fs'
+import * as path from 'path'
+import * as util from 'util'
+
+import type { JsonLdDocument, NodeObject } from 'jsonld'
+import * as jsonld from 'jsonld'
+import * as N3 from 'n3'
+
+import { catalog } from '../index'
+import { WorkResource } from '../resources/resource'
+import { parseCsv } from '../csv'
+
+const SHEETS = ['catalog', 'authors', 'places', 'publishers', 'taxa'] as const
+
+interface Resource {
+    metadata: ResourceMetadata,
+    taxa: AmendedTaxon[],
+}
+
+interface Catalog {
+    catalog: catalog.Entities,
+    authors: catalog.Entities,
+    places: catalog.Entities,
+    publishers: catalog.Entities,
+    taxa: catalog.Entities,
+
+    resources: Record<ResourceId, Resource>,
+}
+
+const PREFIX = 'https://purl.org/identification-resources/'
+const HANDLE_PREFIX = 'https://hdl.handle.net/'
+const SCOPES: Record<string, [string, string]> = {
+    // animal life stage
+    'adults': ['dwciri:lifeStage', 'http://rs.gbif.org/vocabulary/gbif/life_stage/adult'],
+    'pupae': ['dwciri:lifeStage', 'http://rs.gbif.org/vocabulary/gbif/life_stage/pupa'],
+    'juveniles': ['dwciri:lifeStage', 'http://rs.gbif.org/vocabulary/gbif/life_stage/juvenile'],
+    'subimagos': ['dwciri:lifeStage', 'http://rs.gbif.org/vocabulary/gbif/life_stage/juvenile'],
+    'larvae': ['dwciri:lifeStage', 'http://rs.gbif.org/vocabulary/gbif/life_stage/larva'],
+    'larvae (instar V)': ['dwciri:lifeStage', 'http://rs.gbif.org/vocabulary/gbif/life_stage/larva'],
+    'larvae (instar IV)': ['dwciri:lifeStage', 'http://rs.gbif.org/vocabulary/gbif/life_stage/larva'],
+    'larvae (instar III)': ['dwciri:lifeStage', 'http://rs.gbif.org/vocabulary/gbif/life_stage/larva'],
+    'larvae (instar I)': ['dwciri:lifeStage', 'http://rs.gbif.org/vocabulary/gbif/life_stage/larva'],
+    'nymphs': ['dwciri:lifeStage', 'http://rs.gbif.org/vocabulary/gbif/life_stage/larva'],
+    'nypmhs': ['dwciri:lifeStage', 'http://rs.gbif.org/vocabulary/gbif/life_stage/larva'],
+    'nymphs (instar V)': ['dwciri:lifeStage', 'http://rs.gbif.org/vocabulary/gbif/life_stage/larva'],
+    'eggs': ['dwciri:lifeStage', 'http://rs.gbif.org/vocabulary/gbif/life_stage/embryo'],
+
+    // plant life stage
+    'flowering plants': ['dwciri:lifeStage', 'http://rs.gbif.org/vocabulary/gbif/life_stage/adult'],
+    'fruiting plants': ['dwciri:lifeStage', 'http://rs.gbif.org/vocabulary/gbif/life_stage/adult'],
+    'without sporangia': ['dwciri:lifeStage', 'http://rs.gbif.org/vocabulary/gbif/life_stage/juvenile'],
+    'with sporangia': ['dwciri:lifeStage', 'http://rs.gbif.org/vocabulary/gbif/life_stage/adult'],
+    'teleomorphs': ['dwciri:lifeStage', 'http://rs.gbif.org/vocabulary/gbif/life_stage/adult'],
+
+    // sex
+    'females': ['dwciri:sex', 'http://rs.gbif.org/vocabulary/gbif/sex/female'],
+    'males': ['dwciri:sex', 'http://rs.gbif.org/vocabulary/gbif/sex/male'],
+
+    // caste
+    'queens': ['dwc:caste', 'queen'],
+    'workers': ['dwc:caste', 'worker'],
+    'soldiers': ['dwc:caste', 'soldier'],
+    'alates': ['dwc:caste', 'alate'],
+    'alatae': ['dwc:caste', 'alate'],
+    'apterae': ['dwc:caste', 'aptera'],
+    'viviparae': ['dwc:caste', 'vivipara'],
+
+    // evidence
+    'nests': ['ac:subjectPartLiteral', 'nest'],
+    'galls': ['ac:subjectPartLiteral', 'gall'],
+    'puparia': ['ac:subjectPartLiteral', 'puparium'],
+    'eggcases': ['ac:subjectPart', 'http://rs.tdwg.org/acpart/values/p0031'],
+    'bones': ['ac:subjectPartLiteral', 'skeleton'],
+    'bones (skulls)': ['ac:subjectPart', 'http://rs.tdwg.org/acpart/values/p0027'],
+    'bones (upper jaws)': ['ac:subjectPart', 'http://rs.tdwg.org/acpart/values/p0028'],
+    'bones (lower jaws)': ['ac:subjectPart', 'http://rs.tdwg.org/acpart/values/p0029'],
+}
+const PREFIXES = {
+    'ac': 'http://rs.tdwg.org/ac/terms/',
+    'bibo': 'http://purl.org/ontology/bibo/',
+    'dcterms': 'http://purl.org/dc/terms/',
+    'dwc': 'http://rs.tdwg.org/dwc/terms/',
+    'dwciri': 'http://rs.tdwg.org/dwc/iri/',
+    'foaf': 'http://xmlns.com/foaf/0.1/',
+    'owl': 'http://www.w3.org/2002/07/owl#',
+    'schema': 'https://schema.org/',
+}
+const DWC_FIELDS: Record<string, string> = {
+    scientificName: 'dwc:scientificName',
+    scientificNameAuthorship: 'dwc:scientificNameAuthorship',
+    genericName: 'dwc:genericName',
+    infragenericEpithet: 'dwc:infragenericEpithet',
+    specificEpithet: 'dwc:specificEpithet',
+    infraspecificEpithet: 'dwc:infraspecificEpithet',
+    taxonRank: 'dwc:taxonRank',
+    taxonRemarks: 'dwc:taxonRemarks',
+    taxonomicStatus: 'dwc:taxonomicStatus',
+    verbatimIdentification: 'dwc:verbatimIdentification',
+}
+const GBIF_RANKS = ['kingdom', 'phylum', 'class', 'order', 'family', 'genus', 'species', 'subspecies']
+
+function getCoveringTaxon (taxa: catalog.Entity[]): string|null {
+    if (taxa.length === 1 && !taxa[0].has('gbif')) {
+        return `${PREFIX}taxon/${taxa[0].get('id')}`
+    }
+
+    function getAncestors (taxon: catalog.Entity): string[] {
+        const [...parents] = (taxon.get('parent_taxa') ?? []) as string[]
+        if (taxon.has('gbif')) {
+            parents.push(taxon.get('gbif') as string)
+        }
+        return parents
+    }
+
+    const ancestors = getAncestors(taxa[0])
+    for (const taxon of taxa.slice(1)) {
+        const otherAncestors = getAncestors(taxon)
+
+        for (let i = 0; i < ancestors.length; i++) {
+            if (ancestors[i] !== otherAncestors[i]) {
+                ancestors.splice(i)
+            }
+        }
+    }
+
+    return ancestors.length ? makeGbifUri(ancestors[ancestors.length - 1])['@id'] as string : null
+}
+
+function mapEntities (names: string[], entities: catalog.Entity[]): catalog.Entity[] {
+    const result = []
+
+    for (const name of names) {
+        const entity = entities.find(entity => {
+            const values = entity.get('name')
+            return Array.isArray(values) ? values.includes(name) : values === name
+        })
+
+        if (!entity) {
+            console.error('Unmapped entity:', name)
+            continue
+        }
+
+        result.push(entity as catalog.Entity)
+    }
+
+    return result
+}
+
+function makeWikidataUri (qid: string): NodeObject {
+    return { '@id': `http://www.wikidata.org/entity/${qid}` }
+}
+
+function makeGbifUri (id: string): NodeObject {
+    return { '@id': `https://gbif.org/species/${id}` }
+}
+
+function makeWorkUri (id: string): NodeObject {
+    return { '@id': `${PREFIX}catalog/${id}` }
+}
+
+function makeScientificNameUri (id: string): NodeObject {
+    const resource = id.split(':').slice(0, -1).join(':')
+    return { '@id': `${PREFIX}resource/${resource}#${id}` }
+}
+
+function makeLinkedDataForAuthor (author: catalog.Entity): NodeObject {
+    const node: NodeObject = {
+        '@id': `${PREFIX}author/${author.get('id')}`,
+        '@type': 'foaf:Person',
+        'foaf:name': author.get('display_name')
+    }
+
+    if (author.has('qid')) {
+        node['owl:sameAs'] = makeWikidataUri(author.get('qid') as string)
+    }
+
+    return node
+}
+
+function makeLinkedDataForPlace (place: catalog.Entity): NodeObject {
+    const node: NodeObject = {
+        '@id': `${PREFIX}place/${place.get('id')}`,
+        '@type': 'dcterms:Location',
+        'dcterms:title': place.get('display_name'),
+    }
+
+    if (place.has('qid')) {
+        node['owl:sameAs'] = makeWikidataUri(place.get('qid') as string)
+    }
+
+    return node
+}
+
+function makeLinkedDataForPublisher (publisher: catalog.Entity): NodeObject {
+    const node: NodeObject = {
+        '@id': `${PREFIX}author/${publisher.get('id')}`,
+        '@type': 'foaf:Organization',
+        'foaf:name': publisher.get('display_name')
+    }
+
+    if (publisher.has('qid')) {
+        node['owl:sameAs'] = makeWikidataUri(publisher.get('qid') as string)
+    }
+
+    return node
+}
+
+function makeLinkedDataForTaxon (taxon: catalog.Entity): NodeObject {
+    const node: NodeObject = {
+        '@id': `${PREFIX}taxon/${taxon.get('id')}`,
+        '@type': 'dwc:Taxon',
+        'dwc:scientificName': taxon.get('display_name'),
+    }
+
+    if (taxon.has('rank')) {
+        node['dwc:taxonRank'] = taxon.get('rank')
+    }
+
+    const ids = []
+    if (taxon.has('qid')) {
+        ids.push(makeWikidataUri(taxon.get('qid') as string))
+    }
+    if (taxon.has('gbif')) {
+        ids.push(makeGbifUri(taxon.get('gbif') as string))
+    }
+    if (ids.length) {
+        node['owl:sameAs'] = ids
+    }
+
+    return node
+}
+
+function makeLinkedDataForTaxa (files: Catalog): NodeObject[] {
+    const nodes = []
+    const gbifTaxa: Record<string, NodeObject> = {}
+
+    for (const taxon of files.taxa.entities) {
+        const node = makeLinkedDataForTaxon(taxon)
+
+        const ancestors = taxon.get('ancestors_gbif') ?? []
+        if (ancestors.length) {
+            node['dwc:parentNameUsageID'] = makeGbifUri(ancestors[ancestors.length - 1])
+        } else if (taxon.get('id') !== 'T141') {
+            node['dwc:parentNameUsageID'] = { '@id': `${PREFIX}taxon/T141` }
+        }
+
+        for (let i = 0; i < ancestors.length; i++) {
+            gbifTaxa[ancestors[i]] = {
+                ...makeGbifUri(ancestors[i]),
+                'dwc:taxonRank': GBIF_RANKS[i],
+                'dwc:parentNameUsageID': i ? makeGbifUri(ancestors[i - 1]) : { '@id': `${PREFIX}taxon/T141` }
+            }
+        }
+
+        if (taxon.has('children_gbif')) {
+            const children = taxon.get('children_gbif') as string[]
+            node['@reverse'] = { 'dwc:parentNameUsageID': children.map(makeGbifUri) as unknown as string }
+
+            const childRank = GBIF_RANKS[ancestors.length]
+            for (const child of children) {
+                gbifTaxa[child] = {
+                    ...makeGbifUri(child),
+                    'dwc:taxonRank': childRank,
+                }
+            }
+        }
+
+        nodes.push(node)
+    }
+
+    nodes.push(...Object.values(gbifTaxa))
+
+    return nodes
+}
+
+function makeLinkedDataForScientificName (name: AmendedTaxon): NodeObject {
+    const node: NodeObject = {
+        ...makeScientificNameUri(name.scientificNameID),
+        '@type': 'dwc:Taxon',
+    }
+
+    for (const field in DWC_FIELDS) {
+        const value = name[field as keyof AmendedTaxon]
+        if (value) {
+            node[DWC_FIELDS[field]] = value as string
+        }
+    }
+
+    if (name.acceptedNameUsageID) {
+        node['dwc:acceptedNameUsageID'] = makeScientificNameUri(name.acceptedNameUsageID)
+    }
+
+    if (name.parentNameUsageID) {
+        node['dwc:parentNameUsageID'] = makeScientificNameUri(name.parentNameUsageID)
+    }
+
+    const identifiers = []
+    if (name.gbifTaxonID) {
+        identifiers.push(makeGbifUri(name.gbifTaxonID))
+    }
+    if (name.colTaxonID) {
+        identifiers.push({ '@id': `https://www.checklistbank.org/dataset/309120/taxon/${name.colTaxonID}` })
+    }
+
+    if (identifiers.length) {
+        node['dwc:taxonID'] = identifiers
+    }
+
+    return node
+}
+
+function makeLinkedDataForResource (work: catalog.Entity, files: Catalog, resourceId?: ResourceId): NodeObject {
+    const resource = new WorkResource({})
+
+    if (!resourceId) {
+        resourceId = `${work.get('id')}:0`
+    }
+
+    if (files.resources[resourceId] && files.resources[resourceId].metadata.catalog) {
+        resource.fields = files.resources[resourceId].metadata.catalog as Record<string, Value>
+    }
+
+    if (!resource.has('language')) {
+        resource.fields.language = work.get('language') as Value
+    }
+
+    const node: NodeObject = {
+        ...makeLinkedDataForWork(resource, files),
+        '@id': `${PREFIX}resource/${resourceId}`,
+        '@type': 'bibo:DocumentPart',
+    }
+
+    const types = resource.get('key_type') ?? work.get('key_type') ?? []
+
+    if (types.includes('matrix')) {
+        node['dcterms:type'] = { '@id': 'http://purl.org/dc/dcmitype/Software' }
+    } else if (types.includes('key') || types.includes('reference') || types.includes('supplement')) {
+        node['dcterms:type'] = { '@id': 'http://purl.org/dc/dcmitype/Text' }
+    } else if (types.includes('gallery') || types.includes('collection')) {
+        node['dcterms:type'] = { '@id': 'http://purl.org/dc/dcmitype/Collection' }
+    }
+
+    if (types.includes('key') || types.includes('matrix')) {
+        node['ac:subtype'] = { '@id': 'http://rs.tdwg.org/acsubtype/values/IdentificationKey' }
+    }
+
+    const taxonNames = resource.get('taxon') ?? work.get('taxon')
+    if (taxonNames) {
+        const taxa = mapEntities(taxonNames as string[], files.taxa.entities)
+        const coveringTaxon = taxa.length ? getCoveringTaxon(taxa) : null
+        if (coveringTaxon !== null) {
+            node['ac:taxonCoverage'] = { '@id': coveringTaxon }
+        }
+
+        node['dwc:taxonID'] = taxa.map(taxon => ({ '@id': `${PREFIX}taxon/${taxon.get('id')}` }))
+    }
+
+    const scopes = resource.get('scope') ?? work.get('scope')
+    if (scopes) {
+        for (const scope of scopes as string[]) {
+            if (!SCOPES[scope]) {
+                console.error('Unmapped scope:', scope)
+                continue
+            }
+
+            const [property, ...values] = SCOPES[scope]
+
+            if (!Array.isArray(node[property])) {
+                node[property] = []
+            }
+
+            for (const value of values) {
+                if (value.startsWith('http')) {
+                    (node[property] as unknown[]).push({ '@id': value })
+                } else {
+                    (node[property] as unknown[]).push(value)
+                }
+            }
+        }
+    }
+
+    const region = resource.get('region') ?? work.get('region')
+    if (region) {
+        const places = mapEntities(region as string[], files.places.entities).map(place => ({ '@id': `${PREFIX}place/${place.get('id')}` }))
+        node['dcterms:spatial'] = places
+    }
+
+    const tags: string[] = []
+
+    const taxonScopes = resource.get('taxon_scope') ?? work.get('taxon_scope')
+    if (taxonScopes) {
+        tags.push(...taxonScopes as string[])
+    } else if ((resource.get('complete') ?? work.get('complete')) === 'FALSE') {
+        tags.push('not intended to be complete')
+    }
+
+    const targetTaxa = resource.get('target_taxa') ?? work.get('target_taxa')
+    if (targetTaxa) {
+        const first = (targetTaxa as string[]).slice(0, -1).map(rank => rank + ',')
+        const last = targetTaxa[targetTaxa.length - 1]
+        const list = first.length ? first.join('') + ' or ' + last : last
+        tags.push(`for identification to ${list}`)
+    }
+
+    if (tags.length) {
+        node['ac:tag'] = tags
+    }
+
+    if (files.resources[resourceId].taxa) {
+        const leafs = new Set(files.resources[resourceId].taxa.map(taxon => taxon.scientificNameID))
+        const taxa = []
+        for (const taxon of files.resources[resourceId].taxa) {
+            taxa.push(makeLinkedDataForScientificName(taxon))
+
+            if (taxon.parentNameUsageID) {
+                leafs.delete(taxon.parentNameUsageID)
+            }
+        }
+
+        node['dcterms:subject'] = taxa
+        node['ac:taxonCount'] = leafs.size
+    }
+
+    return node
+}
+
+function makeLinkedDataForWork (work: catalog.Entity, files: Catalog): NodeObject {
+    const id = work.get('id') as string
+    const node: NodeObject = makeWorkUri(id)
+
+    const languages = work.get('language') as string[]
+    node['dcterms:language'] = languages.map(language => ({ '@id': `http://id.loc.gov/vocabulary/iso639-1/${language}` }))
+
+    if (work.has('title')) {
+        const title = work.get('title') as string[]
+        node['dcterms:title'] = title.map((value, i) => ({ '@value': value, '@lang': languages[i] }))
+    }
+
+    if (work.has('pages') && (work.get('pages') as string).includes('-')) {
+        const containers = work.has('part_of') ? (work.get('part_of') as string[]).map(id => files.catalog.get(id)) : []
+        if (containers.find(container => container.get('entry_type') === 'online')) {
+            node['@type'] = 'bibo:BookSection'
+        } else {
+            node['@type'] = 'bibo:AcademicArticle'
+        }
+    } else if (work.get('entry_type') === 'online') {
+        node['@type'] = 'bibo:Website'
+    } else {
+        node['@type'] = 'bibo:Book'
+    }
+
+    if (work.has('author')) {
+        const authors = mapEntities(work.get('author') as string[], files.authors.entities).map(author => ({ '@id': `${PREFIX}author/${author.get('id')}` }))
+        node['bibo:authorList'] = { '@list': authors }
+        node['dcterms:creator'] = authors
+    }
+
+    if (work.has('url')) {
+        const urls = work.get('url') as string[]
+
+        const handle = urls.find(url => url.startsWith(HANDLE_PREFIX))
+        if (handle) {
+            node['bibo:handle'] = handle.slice(HANDLE_PREFIX.length)
+        }
+
+        node['schema:url'] = urls.map(url => ({ '@id': url }))
+    }
+
+    if (work.has('fulltext_url')) {
+        const urls = work.get('fulltext_url') as string[]
+
+        node['schema:encoding'] = urls.map(url => ({ '@type': 'schema:MediaObject', 'schema:contentUrl': { '@id': url } }))
+    }
+
+    if (work.has('archive_url')) {
+        node['schema:archivedAt'] = (work.get('archive_url') as string[]).map(url => ({ '@id': url }))
+    }
+
+    if (work.has('date')) {
+        node['dcterms:issued'] = work.get('date')
+    }
+
+    if (work.has('publisher')) {
+        const publishers = mapEntities(work.get('publisher') as string[], files.publishers.entities).map(publisher => ({ '@id': `${PREFIX}publisher/${publisher.get('id')}` }))
+        node['dcterms:publisher'] = publishers
+    }
+
+    if (work.has('ISSN')) {
+        node['bibo:issn'] = work.get('ISSN')
+    }
+
+    if (work.has('ISBN')) {
+        const isbns = work.get('ISBN') as string[]
+
+        for (const isbn of isbns) {
+            if (isbn.length === 13) {
+                node['bibo:isbn13'] = isbn
+            } else if (isbn.length === 10) {
+                node['bibo:isbn10'] = isbn
+            } else {
+                node['bibo:isbn'] = isbn
+            }
+        }
+    }
+
+    if (work.has('QID')) {
+        node['bibo:uri'] = makeWikidataUri(work.get('QID') as string)
+    }
+
+    if (work.has('DOI')) {
+        node['bibo:doi'] = work.get('DOI')
+    }
+
+    if (work.has('volume')) {
+        node['bibo:volume'] = work.get('volume')
+    }
+
+    if (work.has('issue')) {
+        node['bibo:issue'] = work.get('issue')
+    }
+
+    if (work.has('pages')) {
+        const pages = work.get('pages') as string
+        const range = pages.split('-')
+
+        if (pages.match(/^\d+$/)) {
+            node['bibo:numPages'] = parseInt(pages)
+        } else if (range.length === 2) {
+            node['bibo:pages'] = pages
+
+            const [start, end] = range.map(part => parseInt(part))
+            node['bibo:pageStart'] = start
+            node['bibo:pageEnd'] = end
+            node['bibo:numPages'] = end - start + 1
+        } else {
+            node['bibo:pages'] = pages
+        }
+    }
+
+    if (work.has('edition')) {
+        node['bibo:edition'] = work.get('edition')
+    }
+
+    if (work.has('license')) {
+        const licenses = work.get('license') as string[]
+        node['dcterms:rights'] = licenses.map(license => {
+            if (license === '<public domain>' || license.match(/^<.+\?>$/)) {
+                return license.slice(1, -1)
+            } else {
+                return { '@id': `https://spdx.org/licenses/${license}.html` }
+            }
+        })
+    }
+
+    return node
+}
+
+function makeLinkedDataForWorks (files: Catalog): NodeObject[] {
+    const nodes: Record<string, NodeObject> = {}
+    const works = files.catalog
+
+    for (const work of works.entities) {
+        nodes[work.get('id') as string] = makeLinkedDataForWork(work, files)
+    }
+
+    for (const work of works.entities) {
+        const id = work.get('id') as string
+        const node = nodes[id]
+        const language = work.get('language') as string[]
+
+        if (work.has('part_of')) {
+            const containers = work.get('part_of') as string[]
+
+            if ((work.get('key_type') as string[]).includes('supplement')) {
+                node['bibo:annotates'] = containers.map(makeWorkUri)
+            } else {
+                node['dcterms:isPartOf'] = containers.map(makeWorkUri)
+
+                for (const container of containers) {
+                    if (!Array.isArray(nodes[container]['dcterms:hasPart'])) {
+                        nodes[container]['dcterms:hasPart'] = []
+                    }
+                    (nodes[container]['dcterms:hasPart'] as NodeObject[]).push(makeWorkUri(id))
+                }
+            }
+        }
+
+        if (work.has('listed_in')) {
+            const referers = work.get('listed_in') as string[]
+            node['bibo:citedBy'] = node['dcterms:isReferencedBy'] = referers.map(makeWorkUri)
+
+            for (const referer of referers) {
+                nodes[referer]['bibo:cites'] = nodes[referer]['dcterms:references'] = makeWorkUri(id)
+            }
+        }
+
+        if (work.has('version_of')) {
+            const originals = work.get('version_of') as string[]
+            const originalLanguages = originals.map(id => ((works.get(id) as catalog.Entity).get('language') as string[]).join())
+            node['bibo:translationOf'] = originals.filter((_, i) => originalLanguages[i] !== language.join()).map(makeWorkUri)
+            node['dcterms:isVersionOf'] = originals.map(makeWorkUri)
+        }
+
+        let resourceIndex = 1
+        let resourceId
+        while ((resourceId = `${id}:${resourceIndex++}`) in files.resources) {
+            const resource = makeLinkedDataForResource(work, files, resourceId)
+
+            if (!Array.isArray(node['dcterms:hasPart'])) {
+                node['dcterms:hasPart'] = []
+            }
+            (node['dcterms:hasPart'] as NodeObject[]).push(resource)
+
+            if (!Array.isArray(resource['dcterms:isPartof'])) {
+                resource['dcterms:isPartof'] = []
+            }
+            (resource['dcterms:isPartof'] as NodeObject[]).push({ '@id': node['@id'] })
+        }
+    }
+
+    return Object.values(nodes)
+}
+
+async function writeOutput (document: JsonLdDocument, format = 'jsonld'): Promise<void> {
+    if (format === 'jsonld') {
+        process.stdout.write(JSON.stringify(document, null, 2))
+        return
+    }
+
+    const nquads = await jsonld.toRDF(document, { format: 'application/n-quads' }) as string
+    if (format === 'nquads') {
+        process.stdout.write(nquads)
+        return
+    }
+
+    const parser = new N3.StreamParser()
+    const writer = new N3.StreamWriter({ prefixes: PREFIXES })
+    parser.write(nquads)
+    parser.pipe(writer)
+    writer.pipe(process.stdout)
+}
+
+async function main (): Promise<void> {
+    const args = util.parseArgs({
+        allowPositionals: true,
+        options: {
+            format: {
+                type: 'string',
+                short: 'f',
+                default: 'jsonld',
+            }
+        },
+    })
+    const directory = path.resolve(args.positionals[0])
+
+    const files = {} as Catalog
+    for (const sheet of SHEETS) {
+        const filePath = path.join(directory, `${sheet}.csv`)
+        if (!doesFileExist(filePath)) {
+            throw new Error(`File "${sheet}.csv" must be provided`)
+        }
+
+        const file = await fs.readFile(filePath, 'utf8')
+        files[sheet] = catalog.loadData(file, sheet)
+    }
+
+    files.resources = {}
+    const resources = JSON.parse(await fs.readFile(path.join(directory, 'resources', 'index.json'), 'utf8'))
+    for (const id in resources) {
+        const filePath = path.join(directory, 'resources', 'dwc', id.split(':').join('-') + '.csv')
+        const file = await fs.readFile(filePath, 'utf8')
+        const [header, ...rows] = parseCsv(file)
+        const taxa = rows.map(row => row.reduce((object, value, index) => {
+            object[header[index]] = value
+            return object
+        }, {} as Record<string, string>) as unknown as AmendedTaxon)
+
+        files.resources[id] = { metadata: resources[id], taxa }
+    }
+
+    const graph: NodeObject[] = [
+        ...makeLinkedDataForWorks(files),
+        ...makeLinkedDataForTaxa(files),
+    ]
+
+    for (const entity of files.authors.entities) {
+        graph.push(makeLinkedDataForAuthor(entity))
+    }
+    for (const entity of files.places.entities) {
+        graph.push(makeLinkedDataForPlace(entity))
+    }
+    for (const entity of files.publishers.entities) {
+        graph.push(makeLinkedDataForPublisher(entity))
+    }
+
+    const document: JsonLdDocument = {
+        '@context': PREFIXES,
+        '@graph': graph
+    }
+
+    writeOutput(document, args.values.format)
+}
+
+main().catch((error: Error) => {
+    console.error(error)
+    process.exit(1)
+})