Extract Data from BNode

Question

I'm using a SPARQL to extract a node from an RDF file, the node in the rdf file is as follows:

 <dc:description>Birds are a class of vertebrates. They are bipedal, warm-blooded, have a covering of feathers, and their front limbs are modified into wings. Some birds, such as penguins and ostriches, have lost the power of flight. All birds lay eggs. Because birds are warm-blooded, their eggs have to be incubated to keep the embryos inside warm, or they will perish.^M
    <br />
    <br />
    <a href="/nature/19700707">All you need to know about British birds.</a>
</dc:description>

I'm using python RDFLib to get this node. It is returning as

rdflib.term.BNode('Nfc3f01b2567a4b3ea23dbd01394929df')

How is it possible to extract the text from dc:description rdflib.term.BNode('Nfc3f01b2567a4b3ea23dbd01394929df')

Something I tried based on the answers:

from rdflib import *
import rdfextras
import json

#load the ontology
rdfextras.registerplugins()
g=Graph()

g.parse("http://www.bbc.co.uk/nature/life/Bird.rdf")


#define the predixes
PREFIX = ''' PREFIX dc:<http://purl.org/dc/terms/>
             .......
             PREFIX po:<http://purl.org/ontology/po/>
             PREFIX owl:<http://www.w3.org/2002/07/owl#>
         '''

def exe(query):
        query = PREFIX + query
        return g.query(query)

def getEntity(entity_type,entity):
        #getting the description
        entity_url = "<http://www.bbc.co.uk/nature/life/" + entity.capitalize() + "#" + entity_type.lower() +">"
    query = ''' SELECT ?description
                    WHERE { ''' + entity_url + ''' dc:description ?description . }'''
    result_set = exe(query)
    dc = Namespace("http://purl.org/dc/terms/")
        for row in result_set:
                description = row[0]
            print description.value(dc.description)

getEntity("class","bird")

I'm getting the following error:

Traceback (most recent call last):
  File "test_bird1.py", line 40, in <module>
    getEntity("class","bird")
  File "test_bird1.py", line 38, in getEntity
    print description.value(dc.description)
AttributeError: 'BNode' object has no attribute 'value'

score 2 · Accepted Answer · edited May 23 '17 at 12:17

2

BNodes (and URIrefs, too) are resources, so the resource module documentation is probably the most useful documentation for you here. Based on that documentation, it looks like something like this should take care of things for you. Where x is the blank node, and g is the graph, it would look like this:

>>> from rdflib import *
>>> DC = Namespace("http://purl.org/dc/terms/")
>>> r = Resource( g, x )
>>> r.value(DC.description)

As pointed out in this answer to your other question, SPARQL not returning correct result, it's not actually legal to have those <br /> where they're appearing (perhaps you'll need to work with the another serialization, e.g., NTriples, N3, Turtle), so it's hard to predict what different libraries will do with the malformed input. You might let the content producer know that they're publishing ill-formed data.

edited May 23 '17 at 12:17

Community

1
1

answered Jan 03 '14 at 17:50

Joshua Taylor

84,998
9
154
353

Hi, thanks for you answer, I've tried it but getting an error, I'm updated my question with the codes – Noor Jan 03 '14 at 18:10
@Noor Oops, sorry about that; I missed a bit in the documentation. You have to create a resource from the graph and the bnode (or uriref), and the use _its_ value attribute. I've updated my code. – Joshua Taylor Jan 03 '14 at 18:19
Error Trace:Traceback (most recent call last): File "test_bird1.py", line 41, in getEntity("class","bird") File "test_bird1.py", line 38, in getEntity r = Resource(g,description) NameError: global name 'Resource' is not defined noor:test_bird noor$ NameError: global name 'Resource' is not defined – Noor Jan 03 '14 at 18:27
@Noor Did you import the [resource module](https://rdflib.readthedocs.org/en/latest/apidocs/rdflib.html#module-rdflib.resource)? – Joshua Taylor Jan 03 '14 at 18:29
@Noor (I'm not much of a Python user myself, I'm just cribbing from the documentation, and the example in my answer looks pretty close to what's in the docs. Are you running an old version of rdflib?) – Joshua Taylor Jan 03 '14 at 18:31
Ok, no prob man, thanks for the help, u pointed to me to the right doc, thanks for this ;) – Noor Jan 03 '14 at 18:33
Yeah, if you get this working, be sure to post back and let us know what worked – Joshua Taylor Jan 03 '14 at 18:45

score 0 · Answer 2 · edited Mar 05 '17 at 13:19

0

from rdflib import Graph, BNode
g = Graph()
g.parse("http://www.bbc.co.uk/nature/life/Bird.rdf")

for objects in g.objects(subject=BNode(add the BNode code here)):
   print (objects)

edited Mar 05 '17 at 13:19

Tunaki

132,869
46
340
423

answered Mar 05 '17 at 13:18

Albert Sanz

1

Please explain how your solution solves the problem so that future readers would benefit. – Illegal Argument Mar 05 '17 at 13:38

Extract Data from BNode

2 Answers2