# Extract few fields from a WoS file.
# Usage:
# npx ezs extract-fields.ini < corpus_WoS_vol1.json | jq

[use]
plugin = basics

[JSONParse]

[assign]
path = tmp.reprint_addresses
value = get('static_data.fullrecord_metadata.reprint_addresses.address_name', []).castArray().map(name => name.address_spec.full_address)

path = tmp.addresses
value = get('static_data.fullrecord_metadata.addresses.address_name', []).castArray().map(name => name.address_spec.full_address)

[replace]
path = uri
value = get("UID")

path = doi
value = get("dynamic_data.cluster_related.identifiers.identifier").filter(i => i.type === "doi").map("value").uniq()

path = authors
value = get("static_data.summary.names.name").filter(person => person.role === "author").map(author => author.first_name ? author.first_name + ", " + author.last_name : null).compact()

path = title
value = get('static_data.summary.titles.title').filter(title => title.type === "item").map(title => title.content).join('|')

path = abstract
value = get('static_data.fullrecord_metadata.abstracts.abstract.abstract_text.p', []).castArray().map(s => String(s).replace(/<[^>]*>/g, "").replace("Key Points", "")).join(" ; ")

path = publication_year
value = get('static_data.summary.pub_info.pubyear')

path = source
value = get('static_data.summary.titles.title').filter(title => title.type === "source").map(title => title.content).join('|')

path = affiliations
value = get('tmp.addresses', []).concat(_.get(self, 'tmp.reprint_addresses')).remove(null).uniq()

path = countries
value = get('static_data.fullrecord_metadata.addresses.address_name', []).castArray().map("address_spec.country", []).uniq()

path = document_type
value = get('static_data.summary.doctypes.doctype')

path = language
value = get('static_data.fullrecord_metadata.languages.language.content')

path = keywords
value = get('static_data.item.keywords_plus.keyword')

path = subjects
value = get('static_data.fullrecord_metadata.category_info.subjects.subject').map('content').uniq()

path = subheadings
value = get('static_data.fullrecord_metadata.category_info.subheadings.subheading').castArray().uniq()

path = headings
value = get('static_data.fullrecord_metadata.category_info.headings.heading').castArray().uniq()

path = fund_text
value = get('static_data.fullrecord_metadata.fund_ack.fund_text.p')

path = grants
value = get('static_data.fullrecord_metadata.fund_ack.grants.grant').castArray().map(grant => ({ \
    ids:_.chain(grant).get('grant_ids').castArray().map('grant_id').compact().flatten(), \
    agency: _.chain(grant).get('grant_agency_names').castArray().map('content').compact() \
}))

path = publisher
value = get('static_data.summary.publishers.publisher.names.name.unified_name')

[dump]
indent = true