def run_query_new(query, endpoint):
"""Query a SPARQL endpoint and return reportable results in text format.
Modified for specific queries.
Keyword arguments:
query -- a SPARQL query file on disk
endpoint -- an open SPARQL endpoint
"""
result = ""
data = vivotools.vivo_sparql_query(query, baseURL=endpoint)
#print data
#if (not(any(data["results"]["bindings"])))
if len(data["results"]["bindings"]) == 0:
result = "0 \t\t" + (sparql_variable_rename(data["head"]["vars"][0])) + "\n"
return result
else:
result += str(len(data["results"]["bindings"])/2) +" \
"+ (sparql_variable_rename(data["head"]["vars"][0])) + "\n"
for var in data["head"]["vars"]:
result += ("\t"+sparql_variable_rename(var)) + "\t\t\t"
result += "\n"
for item in data["results"]["bindings"]:
for var in data["head"]["vars"]:
result += "\t" + item[var]["value"] + "\t"
result += "\n"
#print result
return result
def run_query(query, endpoint, query_file):
#print query
# assume initially there are no errors
error = False
logging.info("Sending query " + query_file)
data = vivotools.vivo_sparql_query(query, baseURL = endpoint)
#print data
for var in data["head"]["vars"]:
#print var
if (len(data["results"]["bindings"]) == 0):
num_error_records = 0
report = str(num_error_records) + "\t\t" + sparql_variable_rename(var) + "\n"
else:
report = ""
error = True
num_error_records = len(data["results"]["bindings"])
report = str(num_error_records) + "\t\t" + sparql_variable_rename(var) + "\n"
for error_record in data["results"]["bindings"]:
#print error_record
error_value = error_record[var.encode('ascii', 'ignore')]["value"]
#print error_value
# insert variable name and error value
report += "\t\t\t" + error_value + "\n"
#print report
"""# insert code here to re-run reporting function
if error:
query_error = str(query + "_error")
print run_error_query(query)
"""
return report
def make_phone_dictionary(phone_dictionary={}, debug=False):
"""
Extract all phone numbers in VIVO and organize them into a dictionary
keyed by uri and each with a dictionary of phone and primary phone
"""
query = tempita.Template("""
SELECT ?uri ?phone ?primary
WHERE {
{?uri vivo:phoneNumber ?phone .}
UNION {?uri vivo:primaryPhoneNumber ?primary .}
}""")
query = query.substitute()
result = vt.vivo_sparql_query(query)
try:
count = len(result["results"]["bindings"])
except:
count = 0
if debug:
print query, count,\
result["results"]["bindings"][0], result["results"]["bindings"][1]
#
i = 0
while i < count:
b = result["results"]["bindings"][i]
uri = b['uri']['value']
dict = {}
if 'phone' in b:
dict['phone'] = b['phone']['value']
if 'primary' in b:
dict['primary'] = b['primary']['value']
if dict != {}:
phone_dictionary[uri] = dict
i = i + 1
return phone_dictionary
def run_query(query, endpoint, query_file):
"""Query a SPARQL endpoint and return reportable results in text format.
Keyword arguments:
query -- a SPARQL query file on disk
endpoint -- an open SPARQL endpoint
query_file -- the SPARQL query file name
"""
#print query
logging.info("Sending query " + query_file)
data = vivotools.vivo_sparql_query(query, baseURL=endpoint)
#print data
for var in data["head"]["vars"]:
#print var
if len(data["results"]["bindings"]) == 0:
num_error_records = 0
report = str(num_error_records) + "\t\t" + sparql_variable_rename(var) + "\n"
else:
report = ""
num_error_records = len(data["results"]["bindings"])
report = str(num_error_records) + "\t\t" + sparql_variable_rename(var) + "\n"
for error_record in data["results"]["bindings"]:
#print error_record
error_value = error_record[var.encode('ascii', 'ignore')]["value"]
#print error_value
# insert variable name and error value
report += "\t\t\t" + error_value + "\n"
#print report
return report
def t32_dept_counts(uri):
"""
Given a URI of a unit, count five things the NIH wants to know about
the unit for a T32 application
"""
query = tempita.Template("""
#
# Count the faculty, graduate students and postdocs in a unit
#
SELECT
(COUNT(DISTINCT ?fac) AS ?faculty_count)
(COUNT(DISTINCT ?pre) AS ?predoc_count)
(COUNT(DISTINCT ?pos) AS ?postdoc_count)
(COUNT(DISTINCT ?pre_sup) AS ?predoc_supported)
(COUNT(DISTINCT ?pos_sup) AS ?postdoc_supported)
WHERE {
{
?fac ufVivo:homeDept <{{uri}}> .
?fac a vivo:FacultyMember .
}
UNION {
?pre ufVivo:homeDept <{{uri}}> .
?pre a vivo:GraduateStudent .
}
UNION {
?pos ufVivo:homeDept <{{uri}}> .
?pos a vivo:Postdoc .
}
UNION {
?pre_sup ufVivo:homeDept <{{uri}}> .
?pre_sup a vivo:GraduateStudent .
?pre_sup vivo:hasPrincipalInvestigatorRole ?role .
?role vivo:roleIn ?grant .
?grant a vivo:Grant .
}
UNION {
?pos_sup ufVivo:homeDept <{{uri}}> .
?pos_sup a vivo:Postdoc .
?pos_sup vivo:hasPrincipalInvestigatorRole ?role .
?role vivo:roleIn ?grant .
?grant a vivo:Grant .
}
}
GROUP BY ?uri
""")
query = query.substitute(uri=uri)
result = vt.vivo_sparql_query(query)
t32 = {}
for vname in ['faculty_count', 'predoc_count', 'postdoc_count',
'predoc_supported','postdoc_supported']:
try:
t32[vname] = \
result['results']['bindings'][0][vname]['value']
except:
pass
return t32
def run_error_query(query_error):
# portable for *NIX/Windows
if os.name == "nt":
path = os.getcwd() + "\\error_queries\\"
elif os.name == "posix":
path = os.getcwd() + "/error_queries/"
logging.info("Reading error query from " + path + query_error)
data = vivotools.vivo_sparql_query(query_error, baseURL=endpoint)
print data
for var in data["head"]["vars"]:
report = sparql_variable_rename(var) + ":" +"\t\t\t" + data["results"]["bindings"][0][var.encode('ascii', 'ignore')]["value"] + "\n"
return report
def case0(last,first,middle,debug):
query = tempita.Template("""
#search on last name
SELECT ?x ?lname WHERE
{
?x rdf:type core:FacultyMember .
?x foaf:lastName ?lname .
FILTER (regex(?lname, '''{{last}}''', "i"))
}""")
query = query.substitute(last=last)
result = vivotools.vivo_sparql_query(query)
if debug:
print query,result
return result
def case01(last,first,middle,debug):
query = tempita.Template("""
#search on last name
SELECT ?x ?lname ?ufid WHERE
{
?x rdf:type foaf:Person .
?x foaf:lastName ?lname .
?x ufVivo:ufid ?ufid .
FILTER (regex(?lname, '''{{last}}''', "i"))
}""")
query = query.substitute(last=last)
result = vivotools.vivo_sparql_query(query)
if debug:
print query,result
return result
def find_person(ufid):
"""
Given a UFID, find the URI of the person that has that ufidid
"""
query = tempita.Template("""
SELECT ?uri
WHERE {
?uri ufVivo:ufid "{{ufid}}" .
}
""")
query = query.substitute(ufid=ufid)
result = vt.vivo_sparql_query(query)
try:
uri = result['results']['bindings'][0]['uri']['value']
except:
uri = None
return uri
def find_dept(deptid):
"""
Given a UF deptid, find the URI of the org that has that deptid
"""
query = tempita.Template("""
SELECT ?uri
WHERE {
?uri ufVivo:deptID "{{deptid}}" .
}
""")
query = query.substitute(deptid=deptid)
result = vt.vivo_sparql_query(query)
try:
uri = result['results']['bindings'][0]['uri']['value']
except:
uri = None
return uri
def case3(last,first,middle,debug):
query = tempita.Template("""
#search on last, first initial, middle initial
SELECT ?x ?fname ?lname ?mname WHERE
{
?x rdf:type core:FacultyMember .
?x foaf:firstName ?fname .
?x foaf:lastName ?lname .
?x core:middleName ?mname .
FILTER (regex(?fname, '''^{{first}}''', "i"))
FILTER (regex(?mname, '''^{{middle}}''', "i"))
FILTER (regex(?lname, '''{{last}}''', "i"))
}""")
query = query.substitute(last=last,first=first,middle=middle)
result = vivotools.vivo_sparql_query(query)
if debug:
print query,result
return result
def make_items(debug=False):
"""
Extract all the papers for the feed and organize them as feed items
in a list
"""
query = tempita.Template("""SELECT DISTINCT ?x ?dt ?label WHERE {
?x rdf:type bibo:Document .
?x core:dateTimeValue ?dtv .
?x rdfs:label ?label .
?x ufVivo:dateHarvested ?dh .
?dtv core:dateTime ?dt .
FILTER regex(?dh,"{{expression}}")
}
ORDER BY DESC(?dt)""")
query = query.substitute(expression=make_date_expression())
#print query
result = vivotools.vivo_sparql_query(query)
#print result
try:
count = len(result["results"]["bindings"])
except:
count = 0
if debug:
print query, count, result["results"]["bindings"][0],\
result["results"]["bindings"][1]
i = 0
date_cutoff = TODAY - datetime.timedelta(days=DAYS_SINCE_PUBLICATION)
if debug:
print "Cutoff date for publications is ", date_cutoff
items = []
while i < count:
b = result["results"]["bindings"][i]
title = b['label']['value']
uri = b['x']['value']
dt = b['dt']['value']
date_published = datetime.date(int(dt[0:4]), int(dt[5:7]),
int(dt[8:10]))
if date_published >= date_cutoff:
items.append(PyRSS2Gen.RSSItem(title=title,
link=uri,
pubDate=datetime.datetime.now()))
i = i + 1
return items
def run_query_new(query,endpoint):
error = False
result=""
data = vivotools.vivo_sparql_query(query, baseURL = endpoint)
#print data
if (len(data["results"]["bindings"]) == 0):
result = "0 \t\t" + (sparql_variable_rename(data["head"]["vars"][0])) + "\n"
return result
else:
result += str(len(data["results"]["bindings"])) +" "+ (sparql_variable_rename(data["head"]["vars"][0])) + "\n"
print data["head"]["vars"]
for var in data["head"]["vars"]:
result += ("\t\t"+sparql_variable_rename(var)) + "\t\t\t\t\t"
result += "\n"
for item in data["results"]["bindings"]:
for var in data["head"]["vars"]:
result += item[var]["value"] + "\t\t\t\t\t\t"
result += "\n"
#print result
return result
def make_data_dictionary(pred="vivo:totalAwardAmount", debug=False):
"""
Create a dictionary of the specified data pred. Key is uri.
"""
query = """
SELECT ?uri ?data
WHERE {
?uri {{pred}} ?data .
}
"""
query = query.replace('{{pred}}', pred)
result = vivo_sparql_query(query)
if debug:
print "Query = ", query
print "Result = ", len(result['results']['bindings'])
dictionary = {}
for row in result['results']['bindings']:
uri = row['uri']['value']
data = row['data']['value']
dictionary[uri] = data
return dictionary
def make_data_dictionary(pred="vivo:totalAwardAmount", debug=False):
"""
Create a dictionary of the specified data pred. Key is uri.
"""
query = """
SELECT ?uri ?data
WHERE {
?uri {{pred}} ?data .
}
"""
query = query.replace('{{pred}}', pred)
result = vivo_sparql_query(query)
if debug:
print "Query = ", query
print "Result = ", len(result['results']['bindings'])
dictionary = {}
for row in result['results']['bindings']:
uri = row['uri']['value']
data = row['data']['value']
dictionary[uri] = data
return dictionary
def run_query(query, endpoint, query_file):
# assume initially there are no errors
report = ""
error_report = ""
error = False
data = vivotools.vivo_sparql_query(query, baseURL=endpoint)
for var in data["head"]["vars"]:
error_value = \
data["results"]["bindings"][0][var.encode('ascii', 'ignore')]["value"]
# insert variable name and error value
report = error_value + "\t\t" + sparql_variable_rename(var) + "\n"
if int((data["results"]["bindings"][0][var.encode('ascii', 'ignore')]["value"])) != 0:
error = True
if error:
# if we've encountered a non-zero value, run the corresponding error query
# make sure the file exists
query_error = str(query_file[:-3] + "_error" + ".rq")
try:
error_report = run_error_query(query_error)
except IOError:
logging.critical("Critical error: " + query_error + ".rq does not exist.")
return {"report":report, "error_report":error_report}
# VIVO URI tester
# usage: python uritester.py (integer) > outputfile
# specify the number of URIs desired in (integer), and write to outputfile
import vivotools
print vivotools.get_vivo_uri()
print vivotools.get_vivo_uri()
query="""
#search on last, first initial, middle initial
SELECT ?x ?fname ?lname ?mname WHERE
{
?x rdf:type core:FacultyMember .
?x foaf:firstName ?fname .
?x foaf:lastName ?lname .
FILTER (regex(?fname, "^D", "i"))
FILTER (regex(?lname, "Nelson", "i"))
}"""
result = vivotools.vivo_sparql_query(query)
print result
FILTER (?uritype IN (vivo:College, vivo:University, vivo:Department,
ufVivo:AdministrativeUnit,vivo:Center, vivo:Institute, vivo:School, vivo:ExtensionUnit, vivo:Library, vivo:Program, vivo:Committee,
vivo:Foundation, vivo:Laboratory, vivo:Division, vivo:Company,
vivo:Association, vivo:ClinicalOrganization, vivo:Hospital,
vivo:Publisher, ufVivo:NonGovernmentalOrganization,
vivo:ResearchOrganization,
vivo:StudentOrganization, vivo:Team, vivo:Consortium,
vivo:Museum, vivo:AcademicDepartment)) .
FILTER (!bound(?successor)) .
}
GROUP BY ?uri ?childuri
ORDER BY ?uri ?childuri"""
# Data from VIVO SPARQL query
org_result = vivo_sparql_query(query)["results"]["bindings"]
org_dict = {}
for org in org_result:
uri = org["uri"]["value"]
org_dict[uri] = org_dict.get(uri, 0)+1
print len(org_dict), "orgs"
result = {}
uri = "http://vivo.ufl.edu/individual/n1278130" # UF
#uri = "http://vivo.ufl.edu/individual/n8763427" # CTSI
#uri = "http://vivo.ufl.edu/individual/n142500" # CTSI Biobehavioral core
result = tree_build(uri, org_result)
result["org_count"] = len(org_dict)
orgs_file = open("orgs.json", "w")
print >>orgs_file, json.dumps(result, indent=4)
the result as a JSON object
Version 0.1 MC 2013-12-28
-- Initial version.
"""
__author__ = "Michael Conlon"
__copyright__ = "Copyright 2013, University of Florida"
__license__ = "BSD 3-Clause license"
__version__ = "0.1"
import vivotools as vt
import json
from datetime import datetime
print datetime.now(),"Start"
query = """
SELECT ?p ?o
WHERE {
<http://vivo.ufl.edu/individual/n25562> ?p ?o
}
ORDER BY ?p
"""
data=vt.vivo_sparql_query(query,debug=True) # show the encoded query # issue the query, return the data
print "Retrieved data:\n" + json.dumps(data, sort_keys=True, indent=4)
print "Items found = ",len(data["results"]["bindings"])
print datetime.now(),"Finish"
import json, vivotools
query = """
SELECT ?x ?lname WHERE
{
?x rdf:type foaf:Person .
?x foaf:lastName ?lname .
FILTER (regex(?lname,"Conlon","i"))
}
"""
print vivotools.get_vivo_uri()
print vivotools.get_vivo_uri()
data=vivotools.vivo_sparql_query(query) # issue the query, return the data
print "Retrieved data:\n" + json.dumps(data, sort_keys=True, indent=4) # show the returned json object
print "Items found = ",len(data["results"]["bindings"]) # count the items in the result set
for item in data["results"]["bindings"]: # for each item, show the uri and last name
print item["x"]["value"],item["lname"]["value"]
def update_conc(conc, concept, debug=False):
"""
for a concept, update the entry in the concordance for the concept
or create a new entry if one does not exist.
"""
concept_uri = str(concept['uri']['value'])
concept_name = concept['concept_name']['value']
npubs = concept['npub']['value']
if concept_uri in conc:
entry = conc[concept_uri]
else:
entry = {'name' : concept_name,
'npubs' : npubs,
'concepts' : {},
'authors': {}}
# First we get the concordant concepts
query = """
#
# For a specified concept, find all the concepts that co-occur with the
# specified concept in one or more academic articles. For each
# co-occuring concept, return the name, uri and count of papers in which
# the concept and the specified concept co-occur
#
SELECT ?concept_uri (MIN(DISTINCT ?xconcept_name) AS ?concept_name)
(COUNT(DISTINCT ?pub_uri) AS ?count)
WHERE {
?pub_uri vivo:hasSubjectArea <{uri}> .
?pub_uri a bibo:AcademicArticle .
?pub_uri vivo:hasSubjectArea ?concept_uri .
?concept_uri rdfs:label ?xconcept_name .
FILTER(str(?concept_uri) !=
"{uri}")
}
GROUP BY ?concept_uri
ORDER BY DESC(?count)
"""
query = query.replace("{uri}", concept_uri)
result = vivo_sparql_query(query)
if 'results' in result and 'bindings' in result['results'] and \
'count' in result['results']['bindings'][0] and \
int(result['results']['bindings'][0]['count']['value']) != 0:
rows = result['results']['bindings']
print 'concept',len(rows)
# Replace concept content with current content
concept_dict = {}
for row in rows:
concept_name = row['concept_name']['value']
concept_dict[concept_name] = {'concept_uri':
row['concept_uri']['value'],
'count':
row['count']['value']}
entry['concepts'] = concept_dict
# Second we get the concordant authors
query = """
#
# For a specified concept, find all the current UF authors that co-occur with the
# specified concept in one or more academic articles. For each
# co-occuring author, return the name, uri and count of papers in which
# the author and the specified concept co-occur
#
SELECT ?author_uri (MIN(DISTINCT ?xauthor_name) AS ?author_name)
(COUNT(DISTINCT ?pub_uri) AS ?count)
WHERE {
?pub_uri vivo:hasSubjectArea <{uri}> .
?pub_uri a bibo:AcademicArticle .
?pub_uri vivo:informationResourceInAuthorship ?a .
?a vivo:linkedAuthor ?author_uri .
?author_uri a ufVivo:UFCurrentEntity .
?author_uri rdfs:label ?xauthor_name .
}
GROUP BY ?author_uri
ORDER BY DESC(?count)
"""
query = query.replace("{uri}", concept_uri)
result = vivo_sparql_query(query)
if 'results' in result and 'bindings' in result['results'] and \
'count' in result['results']['bindings'][0] and \
int(result['results']['bindings'][0]['count']['value']) != 0:
rows = result['results']['bindings']
print 'author',len(rows)
# Replace concept content with current content
author_dict = {}
for row in rows:
author_name = row['author_name']['value']
author_dict[author_name] = {'author_uri':
row['author_uri']['value'],
'count':
row['count']['value']}
entry['authors'] = author_dict
conc[concept_uri] = entry
return conc
__author__ = "Michael Conlon"
__copyright__ = "Copyright 2013, University of Florida"
__license__ = "BSD 3-Clause license"
__version__ = "0.1"
import vivotools as vt
import random
from datetime import datetime
query = """
SELECT ?uri WHERE
{
?uri a vivo:FacultyMember .
?uri a ufVivo:UFCurrentEntity .
}
"""
print datetime.now(),"Gathering Current UF Faculty from VIVO"
data = vt.vivo_sparql_query(query)
print datetime.now(),"Current UF Faculty found = ",len(data["results"]["bindings"])
print datetime.now(),"Load data structure with results"
d = []
for item in data["results"]["bindings"]:
d.append(item["uri"]["value"])
print datetime.now(),"Select random sample"
random.shuffle(d)
print datetime.now(),"Show selected faculty by VIVO URI"
for i in range(100):
print d[i]
print datetime.now(),"Finished"
def run_error_query(query_error):
data = vivotools.vivo_sparql_query(query_error)
for var in data["head"]["vars"]:
report = sparql_variable_rename(var) + ":" +"\t\t\t" + data["results"]["bindings"][0][var.encode('ascii', 'ignore')]["value"] + "\n"
return report
