def loadTerms():
global LOADEDTERMS
if not LOADEDTERMS:
LOADEDTERMS = True
print("Loading triples files")
SdoTermSource.loadSourceGraph("default")
print ("loaded %s triples - %s terms" % (len(SdoTermSource.sourceGraph()),len(SdoTermSource.getAllTerms())) )
def exportrdf(exportType):
global allGraph, currentGraph
if not allGraph:
allGraph = rdflib.Graph()
allGraph.bind("schema",VOCABURI)
currentGraph = rdflib.Graph()
currentGraph.bind("schema",VOCABURI)
allGraph += SdoTermSource.sourceGraph()
protocol, altprotocol = protocols()
deloddtriples = """DELETE {?s ?p ?o}
WHERE {
?s ?p ?o.
FILTER (! strstarts(str(?s), "%s://schema.org") ).
}""" % (protocol)
allGraph.update(deloddtriples)
currentGraph += allGraph
desuperseded="""PREFIX schema: <%s://schema.org/>
DELETE {?s ?p ?o}
WHERE{
?s ?p ?o;
schema:supersededBy ?sup.
}""" % (protocol)
#Currenty superseded terms are not suppressed from 'current' file dumps
#Whereas they are suppressed from the UI
#currentGraph.update(desuperseded)
delattic="""PREFIX schema: <%s://schema.org/>
DELETE {?s ?p ?o}
WHERE{
?s ?p ?o;
schema:isPartOf <%s://attic.schema.org>.
}""" % (protocol,protocol)
currentGraph.update(delattic)
formats = ["json-ld", "turtle", "nt", "nquads", "rdf"]
extype = exportType[len("RDFExport."):]
if exportType == "RDFExports":
for format in sorted(formats):
_exportrdf(format,allGraph,currentGraph)
elif extype in formats:
_exportrdf(extype,allGraph,currentGraph)
else:
raise Exception("Unknown export format: %s" % exportType)
TRIPLESFILESGLOB = ["data/*.ttl", "data/ext/*/*.ttl"]
EXAMPLESFILESGLOB = ["data/*examples.txt", "data/ext/*/*examples.txt"]
schema_path = './data/schema.ttl'
examples_path = './data/examples.txt'
andstr = "\n AND\n "
TYPECOUNT_UPPERBOUND = 1500
TYPECOUNT_LOWERBOUND = 500
logging.basicConfig(level=logging.INFO)
log = logging.getLogger(__name__)
SdoTermSource.loadSourceGraph("default")
print("loaded %s triples - %s terms" %
(len(SdoTermSource.sourceGraph()), len(SdoTermSource.getAllTerms())))
print("Loading examples files")
SchemaExamples.loadExamplesFiles("default")
print("Loaded %d examples" % SchemaExamples.count())
# Tests to probe the health of both schemas and code.
# Note that known failings can be annotated with @unittest.expectedFailure or @skip("reason...")
class BallparkCountTests(unittest.TestCase):
def test_alltypes(self):
# ballpark estimates.
self.assertTrue(
len(SdoTermSource.getAllTypes()) > TYPECOUNT_LOWERBOUND,
def loadGraph(self):
self.list(SdoTermSource.sourceGraph())
def loadGraphs(self):
SdoTermSource.loadSourceGraph("default")
self.rdflib_data = SdoTermSource.sourceGraph()
def createcontext():
"""Generates a basic JSON-LD context file for schema.org."""
SCHEMAURI = "http://schema.org/"
jsonldcontext = []
jsonldcontext.append("{\n \"@context\": {\n")
jsonldcontext.append(" \"type\": \"@type\",\n")
jsonldcontext.append(" \"id\": \"@id\",\n")
jsonldcontext.append(" \"HTML\": { \"@id\": \"rdf:HTML\" },\n")
#jsonldcontext.append(" \"@vocab\": \"%s\",\n" % SdoTermSource.vocabUri())
jsonldcontext.append(" \"@vocab\": \"%s\",\n" % SCHEMAURI)
ns = SdoTermSource.sourceGraph().namespaces()
done = []
for n in ns:
for n in ns:
pref, pth = n
pref = str(pref)
if not pref in done:
done.append(pref)
if pref == "schema":
pth = SCHEMAURI #Overide vocab setting to maintain http compatability
jsonldcontext.append(" \"%s\": \"%s\",\n" % (pref, pth))
datatypepre = "schema:"
vocablines = ""
externalines = ""
typins = ""
for t in SdoTermSource.getAllTerms(expanded=True, supressSourceLinks=True):
if t.termType == SdoTerm.PROPERTY:
range = t.rangeIncludes
types = []
#If Text in range don't output a @type value
if not "Text" in range:
if "URL" in range:
types.append("@id")
if "Date" in range:
types.append("Date")
if "Datetime" in range:
types.append("DateTime")
typins = ""
for typ in types:
typins += ", \"@type\": \"" + typ + "\""
line = " \"" + t.id + "\": { \"@id\": \"" + prefixedIdFromUri(
t.uri) + "\"" + typins + "},"
elif t.termType == SdoTerm.REFERENCE:
continue
else:
line = " \"" + t.id + "\": {\"@id\": \"" + prefixedIdFromUri(
t.uri) + "\"},"
if t.id.startswith("http:") or t.id.startswith("https:"):
externalines += line
else:
vocablines += line
jsonldcontext.append(vocablines)
#jsonldcontext.append(externalines)
jsonldcontext.append("}}\n")
ret = "".join(jsonldcontext)
ret = ret.replace("},}}", "}\n }\n}")
ret = ret.replace("},", "},\n")
return ret