class Checker(object):
def __init__(self, input_dir, output_dir=None, tmp_dir=None):
self.input_dir = input_dir
self.output_dir = output_dir
self.tmp_dir = tmp_dir
self.storer = Storer()
self.name = self.__class__.__name__
self.repok = Reporter(prefix="[%s - INFO] " % self.name)
self.repok.new_article()
self.reper = Reporter(prefix="[%s - ERROR] " % self.name)
self.reper.new_article()
def process(self):
for cur_dir, cur_subdir, cur_files in os.walk(self.input_dir):
for cur_file in cur_files:
self.repok.new_article()
self.reper.new_article()
cur_rdf_path = cur_dir + os.sep + cur_file
try:
self.repok.add_sentence("Processing '%s'" % cur_rdf_path)
g = self.storer.load(cur_rdf_path, tmp_dir=self.tmp_dir)
if self.output_dir is None:
self.repok.add_sentence("The RDF graph has been converted in TRIG as follows:\n%s"
% g.serialize(format="trig"))
else:
if not os.path.exists(self.output_dir):
os.makedirs(self.output_dir)
output_file = self.output_dir + os.sep + "converted_" + cur_file + ".ttl"
self.repok.add_sentence("The RDF graph has been stored in %s"
% (output_file, g.serialize(output_file, format="trig")))
except Exception:
self.reper.add_sentence("The file '%s' doesn't contain RDF statements", False)
class Checker(object):
def __init__(self, input_dir, output_dir=None, tmp_dir=None):
self.input_dir = input_dir
self.output_dir = output_dir
self.tmp_dir = tmp_dir
self.storer = Storer()
self.name = self.__class__.__name__
self.repok = Reporter(prefix="[%s - INFO] " % self.name)
self.repok.new_article()
self.reper = Reporter(prefix="[%s - ERROR] " % self.name)
self.reper.new_article()
def process(self):
for cur_dir, cur_subdir, cur_files in os.walk(self.input_dir):
for cur_file in cur_files:
self.repok.new_article()
self.reper.new_article()
cur_rdf_path = cur_dir + os.sep + cur_file
try:
self.repok.add_sentence("Processing '%s'" % cur_rdf_path)
g = self.storer.load(cur_rdf_path, tmp_dir=self.tmp_dir)
if self.output_dir is None:
self.repok.add_sentence(
"The RDF graph has been converted in TRIG as follows:\n%s"
% g.serialize(format="trig"))
else:
if not os.path.exists(self.output_dir):
os.makedirs(self.output_dir)
output_file = self.output_dir + os.sep + "converted_" + cur_file + ".ttl"
self.repok.add_sentence(
"The RDF graph has been stored in %s" %
(output_file,
g.serialize(output_file, format="trig")))
except Exception:
self.reper.add_sentence(
"The file '%s' doesn't contain RDF statements", False)
s = Storer(context_map={context_path: context_file_path})
all_files = []
if args.rec:
for cur_dir, cur_subdir, cur_files in os.walk(args.i_dir):
for cur_file in cur_files:
if cur_file.endswith(".json"):
all_files += [cur_dir + os.sep + cur_file]
else:
for cur_file in os.listdir(args.i_dir):
if cur_file.endswith(".json"):
all_files += [args.i_dir + os.sep + cur_file]
for rdf_path in all_files:
cur_g = s.load(rdf_path, tmp_dir=temp_dir_for_rdf_loading)
try:
for o in cur_g.objects(None, GraphEntity.has_identifier):
o_local_path = str(o).replace(base_iri, base_dir) + ".json"
id_g = s.load(o_local_path, tmp_dir=temp_dir_for_rdf_loading)
for v in id_g.objects(None, GraphEntity.has_literal_value):
if v not in id_doc:
id_doc[v] = []
id_doc[v] += [rdf_path]
except:
pass # Do nothing
result = []
for id_lit in id_doc:
cur_list = id_doc[id_lit]
if len(cur_list) > 1:
class DatasetHandler(object):
DCTERMS = Namespace("http://purl.org/dc/terms/")
DCAT = Namespace("http://www.w3.org/ns/dcat#")
VOID = Namespace("http://rdfs.org/ns/void#")
MTT = Namespace("https://w3id.org/spar/mediatype/text/")
DBR = Namespace("http://dbpedia.org/resource/")
dataset = DCAT.Dataset
datafile = DCAT.Distribution
title = DCTERMS.title
description = DCTERMS.description
issued = DCTERMS.issued
modified = DCTERMS.modified
keyword = DCAT.keyword
subject = DCAT.theme
landing_page = DCAT.landingPage
subset = VOID.subset
sparql_endpoint = VOID.sparqlEndpoint
distribution = DCAT.distribution
license = DCTERMS.license
download_url = DCAT.downloadURL
media_type = DCAT.mediaType
byte_size = DCAT.byte_size
label = RDFS.label
a = RDF.type
turtle = MTT.turtle
bibliographic_database = DBR.Bibliographic_database
open_access = DBR.Open_access
scholary_communication = DBR.Scholarly_communication
citations = DBR.Citation
def __init__(self, tp_url_real, context_path, context_file_path,
base_iri, base_dir, info_dir, dataset_home, tmp_dir, triplestore_url=None):
self.tp_url = triplestore_url
self.base_iri = base_iri
self.base_dir = base_dir
self.info_dir = info_dir
self.context_path = context_path
self.dataset_home = URIRef(dataset_home)
self.tmp_dir = tmp_dir
self.tp_res = URIRef(tp_url_real)
self.repok = Reporter(prefix="[DatasetHandler: INFO] ")
self.reperr = Reporter(prefix="[DatasetHandler: ERROR] ")
self.st = Storer(context_map={context_path: context_file_path},
repok=self.repok, reperr=self.reperr)
self.st.set_preface_query(
u"DELETE { ?res <%s> ?date } WHERE { ?res a <%s> ; <%s> ?date }" %
(str(DatasetHandler.modified), str(DatasetHandler.dataset), str(DatasetHandler.modified)))
# /START Create Literal
def create_label(self, g, res, string):
return create_literal(g, res, RDFS.label, string)
def create_publication_date(self, g, res, string):
return create_literal(g, res, self.issued, string, XSD.dateTime)
def update_modification_date(self, g, res, string):
g.remove((res, self.modified, None))
return create_literal(g, res, self.modified, string, XSD.dateTime)
def create_title(self, g, res, string):
return create_literal(g, res, self.title, string)
def create_description(self, g, res, string):
return create_literal(g, res, self.description, string)
def create_keyword(self, g, res, string):
return create_literal(g, res, self.keyword, string)
def create_byte_size(self, g, res, string):
return create_literal(g, res, self.byte_size, string, XSD.decimal)
# /END Create Literal
# /START Create Complex Attributes
def has_subject(self, g, res, obj):
g.add((res, self.subject, obj))
def has_landing_page(self, g, res, obj):
g.add((res, self.landing_page, obj))
def has_subset(self, g, res, obj):
g.add((res, self.subset, obj))
def has_sparql_endpoint(self, g, res, obj):
g.add((res, self.sparql_endpoint, obj))
def has_distribution(self, g, res, obj):
g.add((res, self.distribution, obj))
def has_license(self, g, res, obj):
g.add((res, self.license, obj))
def has_download_url(self, g, res, obj):
g.add((res, self.download_url, obj))
def has_media_type(self, g, res, obj):
g.add((res, self.media_type, obj))
# /END Create Complex Attributes
# /START Types
def dataset_type(self, g, res):
create_type(g, res, self.dataset)
def distribution_type(self, g, res):
create_type(g, res, self.datafile)
# /END Types
def update_dataset_info(self, graph_set):
cur_time = datetime.now().strftime('%Y-%m-%dT%H:%M:%S')
subgraphs_to_update = set()
all_graphs = []
for g in graph_set.graphs():
cur_id = g.identifier
if cur_id not in subgraphs_to_update:
subgraphs_to_update.add(cur_id)
cur_dataset_res = URIRef(cur_id)
cur_dataset = self.get_dataset_graph(cur_dataset_res, cur_time)
self.update_modification_date(cur_dataset, cur_dataset_res, cur_time)
all_graphs += [cur_dataset]
if subgraphs_to_update:
cur_occ_res = URIRef(self.base_iri)
cur_occ = self.get_dataset_graph(cur_occ_res, cur_time)
self.update_modification_date(cur_occ, cur_occ_res, cur_time)
for subgraph_id in subgraphs_to_update:
self.has_subset(cur_occ, cur_occ_res, URIRef(subgraph_id))
all_graphs += [cur_occ]
if all_graphs: # Store everything and upload to triplestore
if self.tp_url is None:
self.st.store_all(
self.base_dir, self.base_iri, self.context_path,
self.tmp_dir, all_graphs, True)
else:
self.st.upload_and_store(
self.base_dir, self.tp_url, self.base_iri, self.context_path,
self.tmp_dir, all_graphs, True)
def get_dataset_graph(self, res, cur_time):
dataset_path = self.get_metadata_path_from_resource(res)
if os.path.exists(dataset_path):
return list(self.st.load(dataset_path, tmp_dir=self.tmp_dir).contexts())[0]
else:
dataset_label = "OCC"
dataset_title = "The OpenCitations Corpus"
dataset_description = "The OpenCitations Corpus is an open repository of scholarly " \
"citation data made available under a Creative Commons public " \
"domain dedication, which provides in RDF accurate citation " \
"information (bibliographic references) harvested from the " \
"scholarly literature (described using the SPAR Ontologies) " \
"that others may freely build upon, enhance and reuse for any " \
"purpose, without restriction under copyright or database law."
if re.search("/../$", str(res)) is not None:
g = Graph(identifier=str(res))
dataset_short_name = str(res)[-3:-1]
dataset_name = GraphSet.labels[dataset_short_name]
dataset_title += ": %s dataset" % dataset_name.title()
dataset_description += " This sub-dataset contains all the '%s' resources." % \
dataset_name
dataset_label += " / %s" % dataset_short_name
self.create_keyword(g, res, dataset_name)
else:
g = Graph()
self.has_landing_page(g, res, self.dataset_home)
self.has_sparql_endpoint(g, res, self.tp_res)
self.dataset_type(g, res)
self.create_label(g, res, dataset_label)
self.create_title(g, res, dataset_title)
self.create_description(g, res, dataset_description)
self.create_publication_date(g, res, cur_time)
self.create_keyword(g, res, "OCC")
self.create_keyword(g, res, "OpenCitations")
self.create_keyword(g, res, "OpenCitations Corpus")
self.create_keyword(g, res, "SPAR Ontologies")
self.create_keyword(g, res, "bibliographic references")
self.create_keyword(g, res, "citations")
self.has_subject(g, res, self.bibliographic_database)
self.has_subject(g, res, self.scholary_communication)
self.has_subject(g, res, self.open_access)
self.has_subject(g, res, self.citations)
return g
def get_metadata_path_from_resource(self, dataset_res):
return self.get_metadata_path_from_iri(str(dataset_res))
def get_metadata_path_from_iri(self, dataset_iri):
return re.sub("^%s" % self.base_iri, self.base_dir, dataset_iri) + "index.json"
class DatasetHandler(object):
DCTERMS = Namespace("http://purl.org/dc/terms/")
DCAT = Namespace("http://www.w3.org/ns/dcat#")
VOID = Namespace("http://rdfs.org/ns/void#")
MTT = Namespace("https://w3id.org/spar/mediatype/text/")
DBR = Namespace("http://dbpedia.org/resource/")
dataset = DCAT.Dataset
datafile = DCAT.Distribution
title = DCTERMS.title
description = DCTERMS.description
issued = DCTERMS.issued
modified = DCTERMS.modified
keyword = DCAT.keyword
subject = DCAT.theme
landing_page = DCAT.landingPage
subset = VOID.subset
sparql_endpoint = VOID.sparqlEndpoint
distribution = DCAT.distribution
license = DCTERMS.license
download_url = DCAT.downloadURL
media_type = DCAT.mediaType
byte_size = DCAT.byte_size
label = RDFS.label
a = RDF.type
turtle = MTT.turtle
bibliographic_database = DBR.Bibliographic_database
open_access = DBR.Open_access
scholary_communication = DBR.Scholarly_communication
citations = DBR.Citation
def __init__(self,
tp_url_real,
context_path,
context_file_path,
base_iri,
base_dir,
info_dir,
dataset_home,
tmp_dir,
triplestore_url=None):
self.tp_url = triplestore_url
self.base_iri = base_iri
self.base_dir = base_dir
self.info_dir = info_dir
self.context_path = context_path
self.dataset_home = URIRef(dataset_home)
self.tmp_dir = tmp_dir
self.tp_res = URIRef(tp_url_real)
self.repok = Reporter(prefix="[DatasetHandler: INFO] ")
self.reperr = Reporter(prefix="[DatasetHandler: ERROR] ")
self.st = Storer(context_map={context_path: context_file_path},
repok=self.repok,
reperr=self.reperr)
self.st.set_preface_query(
u"DELETE { ?res <%s> ?date } WHERE { ?res a <%s> ; <%s> ?date }" %
(str(DatasetHandler.modified), str(
DatasetHandler.dataset), str(DatasetHandler.modified)))
# /START Create Literal
def create_label(self, g, res, string):
return create_literal(g, res, RDFS.label, string)
def create_publication_date(self, g, res, string):
return create_literal(g, res, self.issued, string, XSD.dateTime)
def update_modification_date(self, g, res, string):
g.remove((res, self.modified, None))
return create_literal(g, res, self.modified, string, XSD.dateTime)
def create_title(self, g, res, string):
return create_literal(g, res, self.title, string)
def create_description(self, g, res, string):
return create_literal(g, res, self.description, string)
def create_keyword(self, g, res, string):
return create_literal(g, res, self.keyword, string)
def create_byte_size(self, g, res, string):
return create_literal(g, res, self.byte_size, string, XSD.decimal)
# /END Create Literal
# /START Create Complex Attributes
def has_subject(self, g, res, obj):
g.add((res, self.subject, obj))
def has_landing_page(self, g, res, obj):
g.add((res, self.landing_page, obj))
def has_subset(self, g, res, obj):
g.add((res, self.subset, obj))
def has_sparql_endpoint(self, g, res, obj):
g.add((res, self.sparql_endpoint, obj))
def has_distribution(self, g, res, obj):
g.add((res, self.distribution, obj))
def has_license(self, g, res, obj):
g.add((res, self.license, obj))
def has_download_url(self, g, res, obj):
g.add((res, self.download_url, obj))
def has_media_type(self, g, res, obj):
g.add((res, self.media_type, obj))
# /END Create Complex Attributes
# /START Types
def dataset_type(self, g, res):
create_type(g, res, self.dataset)
def distribution_type(self, g, res):
create_type(g, res, self.datafile)
# /END Types
def update_dataset_info(self, graph_set):
cur_time = datetime.now().strftime('%Y-%m-%dT%H:%M:%S')
subgraphs_to_update = set()
all_graphs = []
for g in graph_set.graphs():
cur_id = g.identifier
if cur_id not in subgraphs_to_update:
subgraphs_to_update.add(cur_id)
cur_dataset_res = URIRef(cur_id)
cur_dataset = self.get_dataset_graph(cur_dataset_res, cur_time)
self.update_modification_date(cur_dataset, cur_dataset_res,
cur_time)
all_graphs += [cur_dataset]
if subgraphs_to_update:
cur_occ_res = URIRef(self.base_iri)
cur_occ = self.get_dataset_graph(cur_occ_res, cur_time)
self.update_modification_date(cur_occ, cur_occ_res, cur_time)
for subgraph_id in subgraphs_to_update:
self.has_subset(cur_occ, cur_occ_res, URIRef(subgraph_id))
all_graphs += [cur_occ]
if all_graphs: # Store everything and upload to triplestore
if self.tp_url is None:
self.st.store_all(self.base_dir, self.base_iri,
self.context_path, self.tmp_dir, all_graphs,
True)
else:
self.st.upload_and_store(self.base_dir, self.tp_url,
self.base_iri, self.context_path,
self.tmp_dir, all_graphs, True)
def get_dataset_graph(self, res, cur_time):
dataset_path = self.get_metadata_path_from_resource(res)
if os.path.exists(dataset_path):
return list(
self.st.load(dataset_path, tmp_dir=self.tmp_dir).contexts())[0]
else:
dataset_label = "OCC"
dataset_title = "The OpenCitations Corpus"
dataset_description = "The OpenCitations Corpus is an open repository of scholarly " \
"citation data made available under a Creative Commons public " \
"domain dedication, which provides in RDF accurate citation " \
"information (bibliographic references) harvested from the " \
"scholarly literature (described using the SPAR Ontologies) " \
"that others may freely build upon, enhance and reuse for any " \
"purpose, without restriction under copyright or database law."
if re.search("/../$", str(res)) is not None:
g = Graph(identifier=str(res))
dataset_short_name = str(res)[-3:-1]
dataset_name = GraphSet.labels[dataset_short_name]
dataset_title += ": %s dataset" % dataset_name.title()
dataset_description += " This sub-dataset contains all the '%s' resources." % \
dataset_name
dataset_label += " / %s" % dataset_short_name
self.create_keyword(g, res, dataset_name)
else:
g = Graph()
self.has_landing_page(g, res, self.dataset_home)
self.has_sparql_endpoint(g, res, self.tp_res)
self.dataset_type(g, res)
self.create_label(g, res, dataset_label)
self.create_title(g, res, dataset_title)
self.create_description(g, res, dataset_description)
self.create_publication_date(g, res, cur_time)
self.create_keyword(g, res, "OCC")
self.create_keyword(g, res, "OpenCitations")
self.create_keyword(g, res, "OpenCitations Corpus")
self.create_keyword(g, res, "SPAR Ontologies")
self.create_keyword(g, res, "bibliographic references")
self.create_keyword(g, res, "citations")
self.has_subject(g, res, self.bibliographic_database)
self.has_subject(g, res, self.scholary_communication)
self.has_subject(g, res, self.open_access)
self.has_subject(g, res, self.citations)
return g
def get_metadata_path_from_resource(self, dataset_res):
return self.get_metadata_path_from_iri(str(dataset_res))
def get_metadata_path_from_iri(self, dataset_iri):
return re.sub("^%s" % self.base_iri, self.base_dir,
dataset_iri) + "index.json"
class ResourceFinder(object):
def __init__(self, g_set=None, ts_url=None, base_dir=None, base_iri=None,
tmp_dir=None, context_map={}):
self.g = Graph()
self.base_dir = base_dir
self.base_iri = base_iri
self.storer = Storer(context_map=context_map)
self.tmp_dir = tmp_dir
self.name = "SPACIN " + self.__class__.__name__
self.loaded = set()
if g_set is not None:
self.update_graph_set(g_set)
if ts_url is None:
self.ts = None
else:
self.ts = ConjunctiveGraph('SPARQLUpdateStore')
self.ts.open((ts_url, ts_url))
def add_prov_triples_in_filesystem(self, res_iri, prov_entity_type=None):
if self.base_dir is not None and self.base_iri is not None:
cur_file_path = find_paths(res_iri, self.base_dir, self.base_iri,
dir_split_number, items_per_file)[1]
if cur_file_path.endswith("index.json"):
cur_path = cur_file_path.replace("index.json", "") + "prov"
else:
cur_path = cur_file_path[:-5] + os.sep + "prov"
file_list = []
if os.path.isdir(cur_path):
for cur_dir, cur_subdir, cur_files in os.walk(cur_path):
for cur_file in cur_files:
if cur_file.endswith(".json") and \
(prov_entity_type is None or cur_file.startswith(prov_entity_type)):
file_list += [cur_dir + os.sep + cur_file]
for file_path in file_list:
if file_path not in self.loaded:
self.loaded.add(file_path)
cur_g = self.storer.load(file_path, tmp_dir=self.tmp_dir)
self.add_triples_in_graph(cur_g)
def add_triples_in_graph(self, g):
if g is not None:
for s, p, o in g.triples((None, None, None)):
self.g.add((s, p, o))
def update_graph_set(self, g_set):
for g in g_set.graphs():
self.add_triples_in_graph(g)
def retrieve(self, id_dict):
for id_type in id_dict:
for id_string in id_dict[id_type]:
res = self.__id_with_type(id_string, id_type)
if res is not None:
return res
def retrieve_provenance_agent_from_name(self, string):
query = """
SELECT DISTINCT ?pa WHERE {
?pa a <%s> ;
<%s> "%s"
} LIMIT 1
""" % (ProvEntity.prov_agent,
GraphEntity.name, string)
return self.__query(query)
def retrieve_from_orcid(self, string):
return self.__id_with_type(string, GraphEntity.orcid)
def retrieve_citing_from_doi(self, string):
return self.__id_with_type(
string.lower(), GraphEntity.doi, "?res <%s> ?cited" % GraphEntity.cites)
def retrieve_citing_from_pmid(self, string):
return self.__id_with_type(
string, GraphEntity.pmid, "?res <%s> ?cited" % GraphEntity.cites)
def retrieve_citing_from_pmcid(self, string):
return self.__id_with_type(
string, GraphEntity.pmcid, "?res <%s> ?cited" % GraphEntity.cites)
def retrieve_citing_from_url(self, string):
return self.__id_with_type(
string.lower(), GraphEntity.url, "?res <%s> ?cited" % GraphEntity.cites)
def retrieve_from_doi(self, string):
return self.__id_with_type(string.lower(), GraphEntity.doi)
def retrieve_from_pmid(self, string):
return self.__id_with_type(string, GraphEntity.pmid)
def retrieve_from_pmcid(self, string):
return self.__id_with_type(string, GraphEntity.pmcid)
def retrieve_from_url(self, string):
return self.__id_with_type(string.lower(), GraphEntity.url)
def retrieve_from_issn(self, string):
return self.__id_with_type(string, GraphEntity.issn)
def retrieve_from_isbn(self, string):
return self.__id_with_type(string, GraphEntity.isbn)
def retrieve_issue_from_journal(self, id_dict, issue_id, volume_id):
if volume_id is None:
return self.__retrieve_from_journal(id_dict, GraphEntity.journal_issue, issue_id)
else:
retrieved_volume = self.retrieve_volume_from_journal(id_dict, volume_id)
if retrieved_volume is not None:
query = """
SELECT DISTINCT ?br WHERE {
?br a <%s> ;
<%s> <%s> ;
<%s> "%s"
} LIMIT 1
""" % (GraphEntity.journal_issue,
GraphEntity.part_of, str(retrieved_volume),
GraphEntity.has_sequence_identifier, issue_id)
return self.__query(query)
def retrieve_volume_from_journal(self, id_dict, volume_id):
return self.__retrieve_from_journal(id_dict, GraphEntity.journal_volume, volume_id)
def retrieve_br_url(self, res, string):
return self.__retrieve_res_id_by_type(res, string.lower(), GraphEntity.url)
def retrieve_br_doi(self, res, string):
return self.__retrieve_res_id_by_type(res, string.lower(), GraphEntity.doi)
def retrieve_br_pmid(self, res, string):
return self.__retrieve_res_id_by_type(res, string, GraphEntity.pmid)
def retrieve_br_pmcid(self, res, string):
return self.__retrieve_res_id_by_type(res, string, GraphEntity.pmcid)
def retrieve_last_snapshot(self, prov_subj):
query = """
SELECT DISTINCT ?se WHERE {
?se <%s> <%s> .
FILTER NOT EXISTS {?se <%s> ?ca }
} LIMIT 1
""" % (ProvEntity.specialization_of, str(prov_subj),
ProvEntity.was_invalidated_by)
return self.__query(query)
def __retrieve_res_id_by_type(self, res, id_string, id_type):
if id_string is not None:
query = """
SELECT DISTINCT ?id WHERE {
<%s> <%s> ?id .
?id <%s> <%s> ;
<%s> "%s"
}""" % (
res, GraphEntity.has_identifier,
GraphEntity.uses_identifier_scheme, id_type,
GraphEntity.has_literal_value, id_string)
return self.__query(query)
def __retrieve_from_journal(self, id_dict, part_type, part_seq_id):
for id_type in id_dict:
for id_string in id_dict[id_type]:
query = """
SELECT DISTINCT ?res WHERE {
?j <%s> ?id .
?id
<%s> <%s> ;
<%s> "%s" .
?res a <%s> ;
<%s>+ ?j ;
<%s> "%s"
}""" % (
GraphEntity.has_identifier,
GraphEntity.uses_identifier_scheme, id_type,
GraphEntity.has_literal_value, id_string,
part_type,
GraphEntity.part_of,
GraphEntity.has_sequence_identifier, part_seq_id
)
return self.__query(query)
def __id_with_type(self, id_string, id_type, extras=""):
query = """
SELECT DISTINCT ?res WHERE {
?res <%s> ?id .
?id
<%s> <%s> ;
<%s> "%s" .
%s
}""" % (
GraphEntity.has_identifier,
GraphEntity.uses_identifier_scheme, id_type,
GraphEntity.has_literal_value, id_string, extras)
return self.__query(query)
def __query(self, query):
if self.ts is not None:
result = self.ts.query(query)
for res, in result:
return res
# If nothing has been returned, check if there is something
# in the current graph set
result = self.g.query(query)
for res, in result:
return res
class ResourceFinder(object):
def __init__(self,
g_set=None,
ts_url=None,
base_dir=None,
base_iri=None,
tmp_dir=None,
context_map={}):
self.g = Graph()
self.base_dir = base_dir
self.base_iri = base_iri
self.storer = Storer(context_map=context_map)
self.tmp_dir = tmp_dir
self.name = "SPACIN " + self.__class__.__name__
self.loaded = set()
if g_set is not None:
self.update_graph_set(g_set)
if ts_url is None:
self.ts = None
else:
self.ts = ConjunctiveGraph('SPARQLUpdateStore')
self.ts.open((ts_url, ts_url))
def add_prov_triples_in_filesystem(self, res_iri, prov_entity_type=None):
if self.base_dir is not None and self.base_iri is not None:
cur_file_path = find_paths(res_iri, self.base_dir, self.base_iri,
dir_split_number, items_per_file)[1]
if cur_file_path.endswith("index.json"):
cur_path = cur_file_path.replace("index.json", "") + "prov"
else:
cur_path = cur_file_path[:-5] + os.sep + "prov"
file_list = []
if os.path.isdir(cur_path):
for cur_dir, cur_subdir, cur_files in os.walk(cur_path):
for cur_file in cur_files:
if cur_file.endswith(".json") and \
(prov_entity_type is None or cur_file.startswith(prov_entity_type)):
file_list += [cur_dir + os.sep + cur_file]
for file_path in file_list:
if file_path not in self.loaded:
self.loaded.add(file_path)
cur_g = self.storer.load(file_path, tmp_dir=self.tmp_dir)
self.add_triples_in_graph(cur_g)
def add_triples_in_graph(self, g):
if g is not None:
for s, p, o in g.triples((None, None, None)):
self.g.add((s, p, o))
def update_graph_set(self, g_set):
for g in g_set.graphs():
self.add_triples_in_graph(g)
def retrieve(self, id_dict):
for id_type in id_dict:
for id_string in id_dict[id_type]:
res = self.__id_with_type(id_string, id_type)
if res is not None:
return res
def retrieve_provenance_agent_from_name(self, string):
query = """
SELECT DISTINCT ?pa WHERE {
?pa a <%s> ;
<%s> "%s"
} LIMIT 1
""" % (ProvEntity.prov_agent, GraphEntity.name, string)
return self.__query(query)
def retrieve_from_orcid(self, string):
return self.__id_with_type(string, GraphEntity.orcid)
def retrieve_citing_from_doi(self, string):
return self.__id_with_type(string.lower(), GraphEntity.doi,
"?res <%s> ?cited" % GraphEntity.cites)
def retrieve_citing_from_pmid(self, string):
return self.__id_with_type(string, GraphEntity.pmid,
"?res <%s> ?cited" % GraphEntity.cites)
def retrieve_citing_from_pmcid(self, string):
return self.__id_with_type(string, GraphEntity.pmcid,
"?res <%s> ?cited" % GraphEntity.cites)
def retrieve_citing_from_url(self, string):
return self.__id_with_type(string.lower(), GraphEntity.url,
"?res <%s> ?cited" % GraphEntity.cites)
def retrieve_from_doi(self, string):
return self.__id_with_type(string.lower(), GraphEntity.doi)
def retrieve_from_pmid(self, string):
return self.__id_with_type(string, GraphEntity.pmid)
def retrieve_from_pmcid(self, string):
return self.__id_with_type(string, GraphEntity.pmcid)
def retrieve_from_url(self, string):
return self.__id_with_type(string.lower(), GraphEntity.url)
def retrieve_from_issn(self, string):
return self.__id_with_type(string, GraphEntity.issn)
def retrieve_from_isbn(self, string):
return self.__id_with_type(string, GraphEntity.isbn)
def retrieve_issue_from_journal(self, id_dict, issue_id, volume_id):
if volume_id is None:
return self.__retrieve_from_journal(id_dict,
GraphEntity.journal_issue,
issue_id)
else:
retrieved_volume = self.retrieve_volume_from_journal(
id_dict, volume_id)
if retrieved_volume is not None:
query = """
SELECT DISTINCT ?br WHERE {
?br a <%s> ;
<%s> <%s> ;
<%s> "%s"
} LIMIT 1
""" % (GraphEntity.journal_issue, GraphEntity.part_of,
str(retrieved_volume),
GraphEntity.has_sequence_identifier, issue_id)
return self.__query(query)
def retrieve_volume_from_journal(self, id_dict, volume_id):
return self.__retrieve_from_journal(id_dict,
GraphEntity.journal_volume,
volume_id)
def retrieve_br_url(self, res, string):
return self.__retrieve_res_id_by_type(res, string.lower(),
GraphEntity.url)
def retrieve_br_doi(self, res, string):
return self.__retrieve_res_id_by_type(res, string.lower(),
GraphEntity.doi)
def retrieve_br_pmid(self, res, string):
return self.__retrieve_res_id_by_type(res, string, GraphEntity.pmid)
def retrieve_br_pmcid(self, res, string):
return self.__retrieve_res_id_by_type(res, string, GraphEntity.pmcid)
def retrieve_last_snapshot(self, prov_subj):
query = """
SELECT DISTINCT ?se WHERE {
?se <%s> <%s> .
FILTER NOT EXISTS {?se <%s> ?ca }
} LIMIT 1
""" % (ProvEntity.specialization_of, str(prov_subj),
ProvEntity.was_invalidated_by)
return self.__query(query)
def __retrieve_res_id_by_type(self, res, id_string, id_type):
if id_string is not None:
query = """
SELECT DISTINCT ?id WHERE {
<%s> <%s> ?id .
?id <%s> <%s> ;
<%s> "%s"
}""" % (res, GraphEntity.has_identifier,
GraphEntity.uses_identifier_scheme, id_type,
GraphEntity.has_literal_value, id_string)
return self.__query(query)
def __retrieve_from_journal(self, id_dict, part_type, part_seq_id):
for id_type in id_dict:
for id_string in id_dict[id_type]:
query = """
SELECT DISTINCT ?res WHERE {
?j <%s> ?id .
?id
<%s> <%s> ;
<%s> "%s" .
?res a <%s> ;
<%s>+ ?j ;
<%s> "%s"
}""" % (GraphEntity.has_identifier,
GraphEntity.uses_identifier_scheme, id_type,
GraphEntity.has_literal_value, id_string, part_type,
GraphEntity.part_of,
GraphEntity.has_sequence_identifier, part_seq_id)
return self.__query(query)
def __id_with_type(self, id_string, id_type, extras=""):
query = """
SELECT DISTINCT ?res WHERE {
?res <%s> ?id .
?id
<%s> <%s> ;
<%s> "%s" .
%s
}""" % (GraphEntity.has_identifier, GraphEntity.uses_identifier_scheme,
id_type, GraphEntity.has_literal_value, id_string, extras)
return self.__query(query)
def __query(self, query):
if self.ts is not None:
result = self.ts.query(query)
for res, in result:
return res
# If nothing has been returned, check if there is something
# in the current graph set
result = self.g.query(query)
for res, in result:
return res
s = Storer(context_map={context_path: context_file_path})
all_files = []
if args.rec:
for cur_dir, cur_subdir, cur_files in os.walk(args.i_dir):
for cur_file in cur_files:
if cur_file.endswith(".json"):
all_files += [cur_dir + os.sep + cur_file]
else:
for cur_file in os.listdir(args.i_dir):
if cur_file.endswith(".json"):
all_files += [args.i_dir + os.sep + cur_file]
for rdf_path in all_files:
cur_g = s.load(rdf_path, tmp_dir=temp_dir_for_rdf_loading)
try:
for o in cur_g.objects(None, GraphEntity.has_identifier):
o_local_path = str(o).replace(base_iri, base_dir) + ".json"
id_g = s.load(o_local_path, tmp_dir=temp_dir_for_rdf_loading)
for v in id_g.objects(None, GraphEntity.has_literal_value):
if v not in id_doc:
id_doc[v] = []
id_doc[v] += [rdf_path]
except:
pass # Do nothing
result = []
for id_lit in id_doc:
cur_list = id_doc[id_lit]
if len(cur_list) > 1:
