def test_construct_query(self):
q = """
CONSTRUCT { v:tmp rdfs:label ?label }
WHERE {
<http://dbpedia.org/resource/Baseball> rdfs:label ?label ;
<http://dbpedia.org/ontology/wikiPageID> ?wikiPageID
}
"""
with self.assertRaises(sparql_dataframe.QueryException):
sparql_dataframe.get(self.endpoint, q)
def test_get_integer(self):
q = """
SELECT DISTINCT ?wikiPageID
WHERE {
<http://dbpedia.org/resource/Baseball> rdfs:label ?label ;
<http://dbpedia.org/ontology/wikiPageID> ?wikiPageID
}
"""
df = sparql_dataframe.get(endpoint, q)
self.assertEqual(df['wikiPageID'].iloc[0], 3850)
def test_get_integer(self):
q = self.prefixes + """
SELECT ?astronaut ?astronautLabel ?birthdate (YEAR(?birthdate) as ?year) ?birthplace WHERE {
?astronaut ?x1 wd:Q11631;
wdt:P569 ?birthdate;
wdt:P19 ?birthplace.
SERVICE wikibase:label { bd:serviceParam wikibase:language "en". }
}
ORDER BY DESC (?birthdate)
LIMIT 10
"""
df = sparql_dataframe.get(self.endpoint, q, post=True)
self.assertEqual(df['year'].iloc[0], 1988)
def test_get_multiple(self):
q = """
SELECT ?label ?wikiPageID
WHERE {
<http://dbpedia.org/resource/Baseball> rdfs:label ?label ;
<http://dbpedia.org/ontology/wikiPageID> ?wikiPageID
}
"""
df = sparql_dataframe.get(endpoint, q)
# Make sure we get 5 or more labels.
self.assertTrue(len(df) > 5)
# Both variables should be of equal length.
self.assertEqual(len(df['label']), len(df['wikiPageID']))
def test_get_multiple(self):
q = self.prefixes + """
SELECT ?human ?humanLabel
WHERE
{
?human wdt:P31 wd:Q5 . #find humans
?human rdf:type wdno:P40 . #with at least one P40 (child) statement defined to be "no value"
SERVICE wikibase:label { bd:serviceParam wikibase:language "[AUTO_LANGUAGE],en" }
}
LIMIT 10
"""
df = sparql_dataframe.get(self.endpoint, q, post=True)
# Make sure we get 5 or more labels.
self.assertTrue(len(df) == 10)
# Both variables should be of equal length.
self.assertEqual(len(df['human']), len(df['humanLabel']))
def aqcheckindoor():
endpoint = ("http://<ip address>/repositories/UNIAQ")
if request.method == "POST":
post_code = request.form["postcodeID"]
pollutant = request.form["pollutantID"]
fromdate = request.form["fromdateID"]
todate = request.form["todateID"]
reading_limit = request.form["limitID"]
q = ("""
PREFIX rdf: <http://www.w3.org/1999/02/22-rdf-syntax-ns#>
PREFIX UNIAQ: <http://score-bradford.com/ontologies/UNIAQ.owl#>
PREFIX : <http://score-bradford.com/ontologies/UNIAQ.owl#>
select ?obs ?val ?uni ?time where {
?s UNIAQ:postcode \"""" + post_code + """\".
?s UNIAQ:observes ?obs.
?obs rdf:type UNIAQ:Observation.
?obs UNIAQ:pollutant UNIAQ:""" + pollutant + """.
?obs UNIAQ:value ?val.
?obs UNIAQ:unit ?uni.
?obs UNIAQ:timestamp ?time
FILTER ((?time> \"""" + fromdate +
""":00\"^^xsd:dateTime)&&(?time< \"""" + todate +
""":00\"^^xsd:dateTime))
} limit """ + reading_limit + """
""")
# Converting Results into Dataframes
result = sparql_dataframe.get(endpoint, q, post=True)
# Standard Deviation
sdv = str(result.loc[:, "val"].std())
# Average
avg = str(result.loc[:, "val"].mean())
# Count
cnt = str(result.loc[:, "val"].count())
# Passing Value to HTML page for Display
return render_template('result.html',
value1=sdv,
value2=avg,
value3=cnt,
value4=pollutant)
# Front End form to get the variable entries from user
return render_template('post_code.html')
def plot():
endpoint = ("http://<ip address>/repositories/INDOOR_TEST")
if request.method == "POST":
post_code = request.form["postcodeID"]
pollutant = request.form["pollutantID"]
fromdate = request.form["fromdateID"]
todate = request.form["todateID"]
reading_limit = request.form["limitID"]
q = ("""
PREFIX rdf: <http://www.w3.org/1999/02/22-rdf-syntax-ns#>
PREFIX UNIAQ: <http://score-bradford.com/ontologies/UNIAQ.owl#>
PREFIX : <http://score-bradford.com/ontologies/UNIAQ.owl#>
select ?val ?time where {
?s UNIAQ:postcode \"""" + post_code + """\".
?s UNIAQ:observes ?obs.
?obs rdf:type UNIAQ:Observation.
?obs UNIAQ:pollutant UNIAQ:""" + pollutant + """.
?obs UNIAQ:value ?val.
?obs UNIAQ:unit ?uni.
?obs UNIAQ:timestamp ?time
FILTER ((?time> \"""" + fromdate +
""":00\"^^xsd:dateTime)&&(?time< \"""" + todate +
""":00\"^^xsd:dateTime))
} limit """ + reading_limit + """
""")
result = sparql_dataframe.get(endpoint, q, post=True)
img = BytesIO()
y = result['val']
x = result['time']
plt.plot(y)
plt.savefig(img, format='png')
plt.close()
img.seek(0)
plot_url = base64.b64encode(img.getvalue()).decode('utf8')
print(y)
return render_template('aqplot.html', plot_url=plot_url)
# Front End form to get the variable entries from user
return render_template('post_code.html')
def test_bad_query(self):
with self.assertRaises(SPARQLWrapper.SPARQLExceptions.QueryBadFormed):
sparql_dataframe.get(self.endpoint, "SELECT ?label FROM {}")
}}
""".format(language, abbre))
sparql.setReturnFormat(JSON)
results = sparql.query().convert()
results_df = pd.io.json.json_normalize(results['results']['bindings'])
results_df[['doid.value', 'item.value', 'label.value', 'wpLang.value']].head()
import sparql_dataframe
endpoint = "https://query.wikidata.org/sparql"
q = """
SELECT DISTINCT ?doid ?item ?label ?wpLang
WHERE {
?item wdt:P699 ?doid ;
rdfs:label ?label ;
rdfs:label ?english .
?article schema:about ?item ;
schema:inLanguage ?wpLang .
FILTER (lang(?label) = ?wpLang)
FILTER (lang(?english) = "en")
}
"""
df = sparql_dataframe.get(endpoint, q)
df
def retrieveData(self, cube, dims, meas, hierdims=[]):
__doc__ = """
retrieveData - SPARQL query builder to retrieve data from SPARQL endpoint.
...
Attributes
--------------
cube: str
Cube to retrieve data
dims: list[str]
list of Strings (dimension names) to retrieve
meas: list[str]
list of measures to retrieve
hierdims: dict{hierdim:{"selected_level":[value]}}
hierarchical dimension (if provided) to retrieve data from specific
hierarchical level
...
Output
------------
Pandas Dataframe
"""
skosInSchemeString = "<http://www.w3.org/2004/02/skos/core#inScheme>"
queryString = ""
queryTemplate = """%s"""
selectString = "SELECT "
groupByString = "GROUP BY "
whereString = "WHERE \n {\n ?s ?p ?o. \n"
hierarchyLevelString = ""
i = 1
for dimension in dims:
selectString += ("(str(?dimLabel" + str(i) + ") as ?" +
self.metaDataDict[cube]["dimensions"][dimension]
["dimension_title"] + ") ")
groupByString += ("?dimLabel" + str(i) + " ")
whereString += "?s <" + self.metaDataDict[cube]["dimensions"][
dimension]["dimension_url"] + "> ?dim" + str(
i) + " .\n ?dim" + str(i) + " rdfs:label ?dimLabel" + str(
i) + ". \n"
i = i + 1
for hierdimension in hierdims:
selectString += ("(str(?dimLabel" + str(i) + ") as ?" +
self.metaDataDict[cube]["hierarchical_dimensions"]
[hierdimension]["dimension_title"] + ") ")
groupByString += ("?dimLabel" + str(i) + " ")
if (hierdims[hierdimension]["selected_level"]):
hierarchyLevelString += "?dim" + str(
i) + " " + skosInSchemeString + " <" + self.metaDataDict[
cube]["hierarchical_dimensions"][hierdimension][
"dimension_prefix"] + hierdims[hierdimension][
"selected_level"] + "> .\n"
whereString += "?s <" + self.metaDataDict[cube][
"hierarchical_dimensions"][hierdimension][
"dimension_url"] + "> ?dim" + str(i) + " .\n ?dim" + str(
i) + " rdfs:label ?dimLabel" + str(i) + ". \n"
i = i + 1
i = 1
for measure in meas:
selectString += (
" (SUM(?measure" + str(i) + ") as ?" +
self.metaDataDict[cube]["measures"][measure]["measure_title"] +
") ")
whereString += (
"?s <" +
self.metaDataDict[cube]["measures"][measure]["measure_url"] +
"> ?measure" + str(i) + " . \n")
whereString += hierarchyLevelString + "} \n"
queryString = selectString + whereString + groupByString
queryTemplate = '''%s '''
sparqldata = sparql_dataframe.get(self.sparqlEndpoint,
queryTemplate % (queryString))
return sparqldata
def queryMe(query):
# print(query)
endpoint = "http://<ip address>/repositories/INDOOR_TEST"
df = sparql_dataframe.get(endpoint, query, post=True)
meanvalue = mean(df)
return meanvalue
def sq2df(qry_str):
"sparql to df"
import sparql_dataframe
endpoint = "https://graph.geodex.org/blazegraph/namespace/cdf/sparql"
q=q1+qs+q2
df = sparql_dataframe.get(endpoint, q)
