def load_statements():
a = datetime.datetime.now()
logger.info(f"start loading ds at: {a}")
ds = Dataset()
ds.parse(STATEMENTS, format=TYPE)
b = datetime.datetime.now()
logger.info(f"finished loading ds at: {b}")
logger.info(f"ds loaded: {ds}")
logger.info(f"ds loaded in {b - a}")
return ds
def test_load_from_file(self):
ds = Dataset()
ds.parse("geoStatements.trig", format="trig")
async def f():
await self.aiotest.addN(
(i for i in ds.quads((None, None, None, None))))
print("ds loaded")
self.loop.run_until_complete(asyncio.gather(f()))
def _get_single_graph_from_trig(trig_file: Optional[str] = None,
data: Optional[str] = None) -> rdflib.Graph:
if trig_file is None and data is None:
raise RuntimeError("One of trig_file OR data *must* be specified.")
dataset = Dataset()
dataset.parse(format="trig", source=trig_file, data=data)
graphs_with_triples = [g for g in dataset.graphs() if len(g) > 0]
assert (
len(graphs_with_triples) == 1
), f"Found {len(graphs_with_triples)} non-trivial graphs in {trig_file}. Expected one."
return graphs_with_triples[0]
def test_roundtrip():
d = Dataset()
d.parse(Path(__file__).parent / "test_parser_hext_multigraph.ndjson",
format="hext",
publicID=d.default_context.identifier)
d.default_union = True
with open(str(
Path(__file__).parent /
"test_parser_hext_multigraph.ndjson")) as i:
ordered_input = "".join(sorted(i.readlines())).strip()
ordered_output = "\n".join(sorted(
d.serialize(format="hext").split("\n"))).strip()
assert ordered_output == ordered_input
def test_hext_dataset_linecount():
d = Dataset()
assert len(d) == 0
d.parse(Path(__file__).parent / "test_parser_hext_multigraph.ndjson",
format="hext",
publicID=d.default_context.identifier)
total_triples = 0
# count all the triples in the Dataset
for context in d.contexts():
for triple in context.triples((None, None, None)):
total_triples += 1
assert total_triples == 18
# count the number of serialized Hextuples, should be 22, as per the original file
lc = len(d.serialize(format="hext").splitlines())
assert lc == 22
def test_hext_json_representation():
"""Tests to see if every link in the ND-JSON Hextuple result is, in fact, JSON"""
d = Dataset()
trig_data = """
PREFIX ex: <http://example.com/>
PREFIX owl: <http://www.w3.org/2002/07/owl#>
PREFIX rdf: <http://www.w3.org/1999/02/22-rdf-syntax-ns#>
PREFIX xsd: <http://www.w3.org/2001/XMLSchema#>
ex:g1 {
ex:s1
ex:p1 ex:o1 , ex:o2 ;
ex:p2 [
a owl:Thing ;
rdf:value "thingy" ;
] ;
ex:p3 "Object 3" , "Object 4 - English"@en ;
ex:p4 "2021-12-03"^^xsd:date ;
ex:p5 42 ;
ex:p6 "42" ;
.
}
ex:g2 {
ex:s1
ex:p1 ex:o1 , ex:o2 ;
.
ex:s11 ex:p11 ex:o11 , ex:o12 .
}
# default graph triples
ex:s1 ex:p1 ex:o1 , ex:o2 .
ex:s21 ex:p21 ex:o21 , ex:o22 .
"""
d.parse(data=trig_data, format="trig")
out = d.serialize(format="hext")
for line in out.splitlines():
j = json.loads(line)
assert isinstance(j, list)
def performUpdate(options):
endpoint = options['endpoint']
inputFile = options['inputfile']
updateCondition = options['updatecondition']
preprocessupdate = options['preprocessupdate']
limit = int(options['limit'])
offset = int(options['offset'])
inputData = Dataset()
print("Parsing input data...")
devnull = inputData.parse(inputFile, format='trig')
print("Found %d named graphs" % len([
d
for d in list(inputData.contexts()) if d.identifier.startswith("http")
]))
headers = {'Accept': 'text/turtle'}
# Query the endpoint and determine which graphs are new, changed, or unchanged
graphs = {'new': [], 'changed': [], 'unchanged': []}
queryTemplate = """
CONSTRUCT { ?s ?p ?o } WHERE { GRAPH <%s> { ?s ?p ?o }}
"""
print("Comparing with named graphs at endpoint %s" % endpoint)
for context in tqdm([
d for d in list(inputData.contexts())
if d.identifier.startswith("http")
][offset:offset + limit]):
r = requests.get(endpoint,
headers=headers,
params={"query": queryTemplate % context.identifier})
if r.ok:
remoteGraph = Graph()
remoteGraph.parse(data=r.text, format='turtle')
if not len(remoteGraph):
graphs['new'].append((context, False))
elif graphsAreTheSame(context, remoteGraph, preprocessupdate):
graphs['unchanged'].append((context, remoteGraph))
else:
graphs['changed'].append((context, remoteGraph))
else:
print(r.text)
# Output statistics:
print("\nComparison Result:")
print("%d graph%s %s not exist at the endpoint and will be added" %
(len(graphs['new']), "" if len(graphs['new']) == 1 else "s",
"does" if len(graphs['new']) == 1 else "do"))
print("%d graph%s already exist%s but %s different in the input file" %
(len(graphs['changed']), "" if len(graphs['changed']) == 1 else "s",
"s" if len(graphs['changed']) == 1 else "",
"is" if len(graphs['changed']) == 1 else "are"))
print("%d graph%s %s identical in both the input file and endpoint" %
(len(graphs['unchanged']), "" if len(graphs['unchanged']) == 1 else
"s", "is" if len(graphs['unchanged']) == 1 else "are"))
# All new graphs should be included in the update
graphsToUpdate = [d[0] for d in graphs['new']]
# Only graphs where the new graph matches the update condition should be updated
# If no update condition is set, all changed should be updated
if updateCondition:
count = 0
for graphPair in graphs['changed']:
for result in graphPair[1].query(updateCondition):
if result:
graphsToUpdate.append(graphPair[0])
count += 1
print(
"\n%d out of %d graph%s will be overwritten based on the update condition"
% (count, len(graphs['changed']),
"" if len(graphs['changed']) == 1 else "s"))
else:
graphsToUpdate += [d[0] for d in graphs['changed']]
# Perform update
for g in tqdm(graphsToUpdate):
putGraph(g, endpoint)
app = Flask(__name__)
api = Api(app)
# set up a custom formatter to return turtle in text/plain to browsers
custom_formatter = flask_rdf.FormatSelector()
custom_formatter.wildcard_mimetype = 'application/ld+json'
custom_formatter.add_format('application/ld+json', 'json-ld')
custom_decorator = flask_rdf.flask.Decorator(custom_formatter)
ds = Dataset(default_union=True)
with open('./dectectorfinalstate.owl', "r") as f:
result = ds.parse(f, format="application/rdf+xml")
class HelloWorld(Resource):
@custom_decorator
def get(self):
return ds
api.add_resource(HelloWorld, '/detectorfinalstate')
@app.route("/")
def main():
# This is cached, so for development it is better
# to use make_response
# return send_file('templates/index.html')'
return make_response(open('templates/index.html').read())
def test_hext_dataset():
"""Tests context-aware (multigraph) data"""
d = Dataset()
trig_data = """
PREFIX ex: <http://example.com/>
PREFIX owl: <http://www.w3.org/2002/07/owl#>
PREFIX rdf: <http://www.w3.org/1999/02/22-rdf-syntax-ns#>
PREFIX xsd: <http://www.w3.org/2001/XMLSchema#>
ex:g1 {
ex:s1
ex:p1 ex:o1 , ex:o2 ;
ex:p2 [
a owl:Thing ;
rdf:value "thingy" ;
] ;
ex:p3 "Object 3" , "Object 4 - English"@en ;
ex:p4 "2021-12-03"^^xsd:date ;
ex:p5 42 ;
ex:p6 "42" ;
.
}
ex:g2 {
ex:s1
ex:p1 ex:o1 , ex:o2 ;
.
ex:s11 ex:p11 ex:o11 , ex:o12 .
}
# default graph triples
ex:s1 ex:p1 ex:o1 , ex:o2 .
ex:s21 ex:p21 ex:o21 , ex:o22 .
"""
d.parse(data=trig_data,
format="trig",
publicID=d.default_context.identifier)
out = d.serialize(format="hext")
# note: cant' test for BNs in result as they will be different ever time
testing_lines = [
[
False,
'["http://example.com/s21", "http://example.com/p21", "http://example.com/o21", "globalId", "", ""]'
],
[
False,
'["http://example.com/s21", "http://example.com/p21", "http://example.com/o22", "globalId", "", ""]'
],
[
False,
'["http://example.com/s1", "http://example.com/p1", "http://example.com/o2", "globalId", "", ""]'
],
[
False,
'["http://example.com/s1", "http://example.com/p1", "http://example.com/o1", "globalId", "", ""]'
],
[
False,
'["http://example.com/s11", "http://example.com/p11", "http://example.com/o12", "globalId", "", "http://example.com/g2"]'
],
[
False,
'["http://example.com/s1", "http://example.com/p1", "http://example.com/o2", "globalId", "", "http://example.com/g2"]'
],
[
False,
'["http://example.com/s11", "http://example.com/p11", "http://example.com/o11", "globalId", "", "http://example.com/g2"]'
],
[
False,
'["http://example.com/s1", "http://example.com/p1", "http://example.com/o1", "globalId", "", "http://example.com/g2"]'
],
[
False,
'["http://example.com/s1", "http://example.com/p1", "http://example.com/o2", "globalId", "", "http://example.com/g1"]'
],
[False, '["http://example.com/s1", "http://example.com/p2"'],
[
False,
'"http://www.w3.org/1999/02/22-rdf-syntax-ns#value", "thingy", "http://www.w3.org/2001/XMLSchema#string", "", "http://example.com/g1"]'
],
[
False,
'"http://www.w3.org/1999/02/22-rdf-syntax-ns#type", "http://www.w3.org/2002/07/owl#Thing", "globalId", "", "http://example.com/g1"]'
],
[
False,
'["http://example.com/s1", "http://example.com/p3", "Object 4 - English", "http://www.w3.org/1999/02/22-rdf-syntax-ns#langString", "en", "http://example.com/g1"]'
],
[
False,
'["http://example.com/s1", "http://example.com/p6", "42", "http://www.w3.org/2001/XMLSchema#string", "", "http://example.com/g1"]'
],
[
False,
'["http://example.com/s1", "http://example.com/p4", "2021-12-03", "http://www.w3.org/2001/XMLSchema#date", "", "http://example.com/g1"]'
],
[
False,
'["http://example.com/s1", "http://example.com/p1", "http://example.com/o1", "globalId", "", "http://example.com/g1"]'
],
[
False,
'["http://example.com/s1", "http://example.com/p5", "42", "http://www.w3.org/2001/XMLSchema#integer", "", "http://example.com/g1"]'
],
[
False,
'["http://example.com/s1", "http://example.com/p3", "Object 3", "http://www.w3.org/2001/XMLSchema#string", "", "http://example.com/g1"]'
],
]
for line in out.splitlines():
for test in testing_lines:
if test[1] in line:
test[0] = True
assert all([x[0] for x in testing_lines])
