def make_documents_broad_match(document_ids):
"""If a work has more than one score, the documents are broadMatches of each other. We should
create relations between *all permutations* of these documents."""
for from_id, to_id in itertools.permutations(document_ids, 2):
query = cequery.document.get_mutation_merge_document_broad_match(
from_id, to_id)
connection.submit_query(query)
def import_or_update_work(work, artist_data):
source = work["Source"]
title = work["Title"]
logger.info("Adding work: %s", title)
existing_work = connection.get_music_composition_by_source(source)
if len(existing_work) == 0:
query = cequery.work.transform_data_create_composition(work)
resp = connection.submit_query(query)
music_composition = resp["data"]["CreateMusicComposition"]
logger.info(" done")
elif len(existing_work) == 1:
logger.info(" already exists, updating")
existing_work = existing_work[0]
identifier = existing_work["identifier"]
query = cequery.work.transform_data_update_composition(
identifier, work)
resp = connection.submit_query(query)
music_composition = resp["data"]["UpdateMusicComposition"]
else:
raise Exception(
"Unexpectedly got more than 1 entry for a given source")
work_identifier = music_composition["identifier"]
logger.info(" work id %s", work_identifier)
composer = _get_composer_data(work, artist_data)
composer_identifer = import_or_update_composer(composer)
join_work_composer(work_identifier, composer_identifer)
add_documents(work_identifier, work)
def join_work_and_documents(work_identifier, document_identifiers):
"""Mark that a score is the subjectOf a work"""
for d_id in document_identifiers:
query = cequery.document.get_query_remove_document_composition(
work_identifier, d_id)
connection.submit_query(query)
query = cequery.document.get_query_add_document_composition(
work_identifier, d_id)
connection.submit_query(query)
def join_work_composer(work_identifier, composer_identifier):
"""Say that a composer wrote a score"""
logger.info("joining work %s and composer %s", work_identifier,
composer_identifier)
# Remove any link that is already there (We'd have to do a query anyway to check if it exists,
# so just do it unconditionally)
# TODO: Could select this query when getting the work originally to check if we update it
query = cequery.work.get_mutation_merge_composition_author(
work_identifier, composer_identifier)
connection.submit_query(query)
logger.info("done")
def add_or_get_digital_document(document):
logger.info("Adding document")
query_resp = connection.get_digital_document_by_source(document["source"])
if len(query_resp) == 0:
logger.info(" doesn't exist, adding")
query = cequery.document.transform_data_create_document(document)
resp = connection.submit_query(query)
resp_document = resp["data"]["CreateDigitalDocument"]
elif len(query_resp) == 1:
logger.info(" exists")
resp_document = query_resp[0]
identifier = resp_document["identifier"]
logger.info(" document id %s", identifier)
return identifier
def import_or_update_composer(composer):
"""Import metdata for a composer from a datafile"""
logger.info("Adding composer: %s", composer["name"])
existing_composer = connection.get_person_by_source(composer["source"])
if len(existing_composer) == 0:
logger.info(" doesn't exist, adding")
query = cequery.person.transform_data_artist(composer)
resp = connection.submit_query(query)
person = resp["data"]["CreatePerson"]
elif len(existing_composer) == 1:
logger.info(" exists, getting")
person = existing_composer[0]
else:
logger.error(
"Unexpectedly got more than 1 composer for a given source")
identifier = person["identifier"]
logger.info(" composer id: %s", identifier)
return identifier
def join_work_and_documents(work_identifier, document_identifiers):
"""Mark that a score is a workExample a work"""
for d_id in document_identifiers:
query = cequery.document.get_mutation_merge_document_composition(
work_identifier, d_id)
connection.submit_query(query)
def import_work(work_mbid):
work = mb.get_artist_by_id(work_mbid, includes=["url-rels"])["work"]
query = person.transform_work(work)
print(work)
submit_query(query)
def import_artist(artist_mbid):
artist = mb.get_artist_by_id(artist_mbid, includes=["url-rels"])["artist"]
query = artist.transform_musicbrainz_artist(artist)
submit_query(query)
