def import_gadm():
total = 0
limit = 2000
f_path = "C:\Users\Al\PycharmProjects\AlignmentUI\src\UploadedFiles\gadm.ttl"
b_path = "C:\Users\Al\PycharmProjects\AlignmentUI\src\UploadedFiles\gadm{}".format(
Ut.batch_extension())
# CREATE THE WRITERS
writer = codecs.open(f_path, "wb", "utf-8")
batch_writer = codecs.open(b_path, "wb", "utf-8")
# GENERATING THE BATCH FILE TEXT
graph = "{}gadm".format(Ns.dataset)
stardog_path = '' if Ut.OPE_SYS == "windows" else Svr.settings[
St.stardog_path]
load_text = """echo "Loading data"
{}stardog data add {} -g {} "{}"
""".format(stardog_path, Svr.settings[St.database], graph,
f_path)
batch_writer.write(to_unicode(load_text))
batch_writer.close()
print "1. GET THE TOTAL NUMBER OF TRIPLES TO LOAD"
count_query = import_gadm_query(is_count=True)
# print count_query
count_res = Qry.virtuoso_request(count_query)
result = count_res['result']
if result is None:
print "NO RESULT FOR THIS ENRICHMENT."
return count_res
print "2. PROCESSING THE COUNT RESULT"
g = rdflib.Graph()
g.parse(data=result, format="turtle")
attribute = rdflib.URIRef("http://www.w3.org/2005/sparql-results#value")
for subject, predicate, obj in g.triples((None, attribute, None)):
total = int(obj)
iterations = total / limit if total % limit == 0 else total / limit + 1
print "\tTOTAL TRIPLES TO RETREIVE : {} \n\tTOTAL NUMBER OF ITERATIONS : {}\n".format(
total, iterations)
# RUN THE ITERATIONS
try:
for i in range(0, iterations):
offset = i * limit + 1
print "ROUND: {} OFFSET: {}".format(i, offset)
print "\tRUNNING THE QUERY"
import_query = import_gadm_query(limit=limit,
offset=offset,
is_count=False)
response = Qry.virtuoso_request(import_query)
print "RESPONSE SIZE: ".format(response["result"])
print "\tWRITING THE RESULT TO FILE"
writer.write(response["result"])
break
except Exception as err:
print str(err.message)
# CLOSE THE IMPORT WRITER
writer.close()
print "4. RUNNING THE BATCH FILE"
print "THE DATA IS BEING LOADED OVER HTTP POST." if Svr.settings[St.split_sys] is True \
else "THE DATA IS BEING LOADED AT THE STARDOG LOCAL HOST FROM BATCH."
print "PATH:", b_path
os.system(b_path)
print "JOB DONE!!!"
def enrich(specs, directory, endpoint):
# TODO RUN IT IF THERE IS NOT GRAPH ENRICHED WITH THE SAME NAME
# specs[St.graph] = "http://grid.ac/20170712"
print "ENRICHING DATA/GRAPH FROM EXPORT-ALIGNMENT"
print "GRAPH:", specs[St.graph]
print "ENTITY TYPE:", specs[St.entity_datatype]
print "LAT PREDICATE:", specs[St.long_predicate]
print "LONG PREDICATE:", specs[St.lat_predicate]
print "FILE DIRECTORY:", directory
name = Ut.get_uri_local_name(specs[St.graph])
print endpoint
data_1 = Qry.virtuoso_request(
"ask {{ GRAPH <{}> {{ ?x ?y ?z . }} }}".format(specs[St.graph]),
endpoint)
data_1 = regex.findall("rs:boolean[ ]*(.*)[ ]*\.", data_1["result"])
if len(data_1) > 0:
data_1 = data_1[0].strip() == "true"
if data_1 is False:
print "GRAPH: {} {}".format(
specs[St.graph], "DOES NOT EXIST AT THE REMOTE VIRTUOSO SITE.")
# CHECKING WHETHER BOTH DATASETS ARE AT THE VIRTUOSO TRIPLE STORE
data_2 = Qry.virtuoso_request(
"ask {GRAPH <http://geo.risis.eu/gadm>{ ?x ?y ?z . }}", endpoint)
data_2 = regex.findall("rs:boolean[ ]*(.*)[ ]*\.", data_2["result"])
if len(data_2) > 0:
data_2 = data_2[0].strip() == "true"
if data_2 is False:
print "GRAPH: {} {}".format(
specs[St.graph], "DOES NOT EXIST AT THE REMOTE VIRTUOSO SITE.")
if data_1 is False or data_2 is False:
message = "BECAUSE BOTH DATASETS NEED TO BE PRESENT AT OUR TRIPLES STORE, WE ARE UNABLE TO EXECUTE THE REQUEST."
return {
St.message:
message,
St.result:
'The dataset {} '
'cannot be enriched with GADM boundary at the moment.'.format(
specs[St.graph])
}
total = 0
limit = 20000
date = datetime.date.isoformat(datetime.date.today()).replace('-', '')
f_path = "{0}{1}{1}{2}_enriched_{3}.ttl".format(directory, os.path.sep,
name, date)
b_path = "{0}{1}{1}{2}_enriched_{3}{4}".format(directory, os.path.sep,
name, date,
Ut.batch_extension())
# MAKE SURE THE FOLDER EXISTS
try:
if not os.path.exists(directory):
os.makedirs(directory)
except OSError as err:
print "\n\t[utility_LOAD_TRIPLE_STORE:]", err
return
print "\n1. GETTING THE TOTAL NUMBER OF TRIPLES."
count_query = enrich_query(specs, limit=0, offset=0, is_count=True)
print count_query
count_res = Qry.virtuoso_request(count_query, endpoint)
result = count_res['result']
# GET THE TOTAL NUMBER OF TRIPLES
if result is None:
print "NO RESULT FOR THIS ENRICHMENT."
return count_res
g = rdflib.Graph()
g.parse(data=result, format="turtle")
attribute = rdflib.URIRef("http://www.w3.org/2005/sparql-results#value")
for subject, predicate, obj in g.triples((None, attribute, None)):
total = int(obj)
# NUMBER OF REQUEST NEEDED
iterations = total / limit if total % limit == 0 else total / limit + 1
print "\n2. TOTAL TRIPLES TO RETREIVE : {} \n\tTOTAL NUMBER OF ITERATIONS : {}\n".format(
total, iterations)
writer = codecs.open(f_path, "wb", "utf-8")
batch_writer = codecs.open(b_path, "wb", "utf-8")
print "3. GENERATING THE BATCH FILE TEXT"
enriched_graph = "{}_enriched".format(specs[St.graph])
stardog_path = '' if Ut.OPE_SYS == "windows" else Svr.settings[
St.stardog_path]
load_text = """echo "Loading data"
{}stardog data add {} -g {} "{}"
""".format(stardog_path, Svr.settings[St.database], enriched_graph,
f_path)
batch_writer.write(to_unicode(load_text))
batch_writer.close()
# RUN THE ITERATIONS
for i in range(0, iterations):
offset = i * 20000 + 1
print "\tROUND: {} OFFSET: {}".format(i + 1, offset)
# print "\t\t1. GENERATING THE ENRICHMENT QUERY"
virtuoso = enrich_query(specs,
limit=limit,
offset=offset,
is_count=False)
# print virtuoso
# exit(0)
# print Qry.virtuoso(virtuoso)["result"]
# print "\t\t2. RUNNING THE QUERY + WRITE THE RESULT TO FILE"
writer.write(Qry.virtuoso_request(virtuoso, endpoint)["result"])
writer.close()
print "\n4. RUNNING THE BATCH FILE"
print "\tTHE DATA IS BEING LOADED OVER HTTP POST." if Svr.settings[St.split_sys] is True \
else "\tTHE DATA IS BEING LOADED AT THE STARDOG LOCAL HOST FROM BATCH."
# os.system(b_path)
# RUN THE BATCH FILE
print "\tFILE: {}".format(f_path)
print "\tBATCH: {}\n".format(b_path)
os.chmod(b_path, 0o777)
Ut.batch_load(b_path)
if os.path.exists(b_path) is True:
os.remove(b_path)
# TODO 1. REGISTER THE DATASET TO BE ENRICHED IF NOT YET REGISTER
# TODO 2. ADD THE ENRICHED DATASET TO THE RESEARCH QUESTION (REGISTER).
# TODO 3. MAYBE, CREATE THE LINKSET BETWEEN THE SOURCE AND THE RESULTING
size = Qry.get_namedgraph_size(enriched_graph)
print "JOB DONE...!!!!!!"
return {
St.message:
"The select dataset was enriched with the GADM boundary as {}. "
"{} triples were created.".format(enriched_graph, size),
St.result:
enriched_graph
}
