Python Graph.preferredLabel方法代码示例

# 需要导入模块: from rdflib import Graph [as 别名]
# 或者: from rdflib.Graph import preferredLabel [as 别名]
    def test_language(self):
        rec = '''
          <marc:record xmlns:marc="http://www.loc.gov/MARC21/slim">
            <marc:leader>00000nw  a2200000n  4500</marc:leader>
            <marc:datafield tag="040" ind2=" " ind1=" ">
              <marc:subfield code="a">OCLCD</marc:subfield>
              <marc:subfield code="b">nob</marc:subfield>
              <marc:subfield code="c">OCLCD</marc:subfield>
            </marc:datafield>
            <marc:datafield tag="084" ind2=" " ind1="0">
              <marc:subfield code="a">ddc</marc:subfield>
              <marc:subfield code="c">23no</marc:subfield>
              <marc:subfield code="e">nob</marc:subfield>
            </marc:datafield>
            <marc:datafield tag="153" ind2=" " ind1=" ">
              <marc:subfield code="a">564.58</marc:subfield>
              <marc:subfield code="e">564.5</marc:subfield>
              <marc:subfield code="j">Decapoda (tiarmede blekkspruter)</marc:subfield>
            </marc:datafield>
          </marc:record>
        '''
        graph = Graph()
        process_record(graph, rec, base_uri='http://test/{object}')
        uri = URIRef(u'http://test/564.58')

        assert graph.preferredLabel(uri)[0][0] == SKOS.prefLabel
        assert graph.preferredLabel(uri)[0][1].language == 'nb'

# 需要导入模块: from rdflib import Graph [as 别名]
# 或者: from rdflib.Graph import preferredLabel [as 别名]
#!/usr/bin/env python
# -*- coding: utf-8 -*-
import sys, rdflib
from rdflib import Graph, Namespace, URIRef, Literal, RDF, RDFS
from unidecode import unidecode

SKOS=Namespace("http://www.w3.org/2004/02/skos/core#")

graph = Graph()
graph.parse(sys.argv[1], format='turtle')

for s,p,o in graph.triples( (None, rdflib.RDF.type, SKOS.Concept) ):
    result = graph.preferredLabel(s, lang='fi')
    if result:
        label = result[-1][-1].value
        stripped = unidecode(label)
        if label != stripped and 'ä'.decode('utf-8') not in label and 'ö'.decode('utf-8') not in label:
            graph.add((s, SKOS.altLabel, Literal(stripped, lang='fi')))
    for alt in graph.objects(s, SKOS.altLabel):
        stripped = unidecode(alt.value)
        if stripped != alt.value and 'ä'.decode('utf-8') not in label and 'ö'.decode('utf-8') not in label:
            graph.add((s, SKOS.hiddenLabel, Literal(stripped, lang='fi')))

graph.serialize(format='turtle', destination=sys.stdout)

# 需要导入模块: from rdflib import Graph [as 别名]
# 或者: from rdflib.Graph import preferredLabel [as 别名]
class SemWP(Graph):
    
    def __init__(self, fname='lrmirdfs.ttl',
                       fmat='turtle'):
        if fname != '':
            self.g = Graph().parse(fname, format=fmat)
            schema = Namespace(u'http://schema.org/')
            self.thing = schema.Thing
            semwp_ns = Namespace(u'http://ns.pjjk.net/semwp')
        else:
            self.g = Graph()
            schema = Namespace(u'http://schema.org/')
            self.thing = schema.Thing
            semwp_ns = Namespace(u'http://ns.pjjk.net/semwp')
        self.name = None

    def resource_label(self, r: URIRef, lang='en'):
    # Given SemWP object amd a resource uri and a language code string
    # returns the most appropriate label for the resource as a string.
    # Uses g.preferredLabel with preferred language but defaults first
    # alternative option if possible preferred lang not available
        if self.g.preferredLabel(r, lang):
            resourcelabel = self.g.preferredLabel(r, lang)[0][1].toPython()
        elif self.g.preferredLabel(r):
            resourcelabel = self.g.preferredLabel(r)[0][1].toPython()
        else:
            raise Exception('resource %r has no label' % r)
            return
        return resourcelabel

    def resource_id(self, r: URIRef, lang='en'):
    # Given SemWP object amd a resource uri and a language code string
    # returns a string that can be used as an id for custom post types
    # for that resource.
    # Uses g.preferredLabel with preferred language but defaults first
    # alternative option if possible preferred lang not available
    # bumps to lower case and truncates to 20 chars
        resourceid = self.resource_label(r, lang)[:20].lower()
        return resourceid

    def resource_comment(self, r: URIRef, lang='en'):
    # Given SemWP object and a resource uri returns the rdfs comment as a string
    # stripped of its HTML cruft, new lines and quotes
    # Currently lang  is ignored (all in schema comments are in english)
        s = MLStripper()
        html = self.g.value(r, RDFS.comment, None).toPython()
        html += '<br>'      # bizarrely, will return null if no html tags at all
        s.feed(html)
        comment = s.get_data()
        comment = comment.replace('\n',' ')
        comment = comment.replace("'","")
        comment = comment.replace('"','')
        comment = comment.replace(')','')
        comment = comment.replace('(','')
        return comment    

    def peek(self, iterable):
        try:
            first = next(iterable)
        except StopIteration:
            return None
        return first, chain([first], iterable)

    def top_classes(self):
    # give a SemWP object returns a list of rdfs classes that are
    # not RDFS.subClass of anything
        tclist = []
        for c in self.g.subjects(RDF.type, RDFS.Class):
            parent = self.peek(self.g.objects(c, RDFS.subClassOf))
            if  parent is None:
                tclist.append(c)
        return tclist

    def sub_classes(self, c: URIRef, recurse=True):
    # Given SemWP object and a class returns a list of rdfs
    # subclasses of the class from the current instance of SemWP graph.
    # Will descend down through subclasses of subclasses if recurse is
    # set to default val of True
        sclist = []
        if self.g.subjects(RDFS.subClassOf, c):
            for sc in self.g.subjects(RDFS.subClassOf, c):
                sclist.append(sc)
                if recurse:
                    sclist.extend(self.sub_classes(sc))
            return sclist
        else:
            return sclist
        
    def super_classes(self, c: URIRef, recurse=True):
    # Given SemWP object and a class returns a list of classes of which
    # the class is a sublass from the current instance of SemWP graph.
    # Will ascend up through sperclasses of superclasses if recurse is
    # set to default val of True
        sclist = []
        if self.g.objects(c, RDFS.subClassOf):
            for sc in self.g.objects(c, RDFS.subClassOf):
                sclist.append(sc)
                if recurse:
                    sclist.extend(self.super_classes(sc))
            return sclist
#.........这里部分代码省略.........

# 需要导入模块: from rdflib import Graph [as 别名]
# 或者: from rdflib.Graph import preferredLabel [as 别名]
        return '' # no difference
    else:
        return result

g = Graph()
g.parse(sys.argv[1], format='turtle')

concepts = list(g.subjects(SKOS.inScheme, YSO['']))
random.shuffle(concepts)

writer = csv.writer(sys.stdout)
i = 0

for conc in concepts:
    vals = [conc]
    is_plural = False
    for lang,pl_suffix in PLURAL_SUFFIXES.items():
        try:
            label = g.preferredLabel(conc, lang=lang)[0][1]
            if label.endswith(pl_suffix):
                is_plural = True
            labelSingular = singular(label, lang)
        except:
            label = labelSingular = ''
        vals += [label.encode('UTF-8'), labelSingular.encode('UTF-8')]
    if not is_plural:
        continue
    writer.writerow(vals)
    i += 1
    if i >= 1000: break

# 需要导入模块: from rdflib import Graph [as 别名]
# 或者: from rdflib.Graph import preferredLabel [as 别名]
        return URIRef(pnr_uri.replace(MMLPNR, LDFPNR['P_']))
    elif SUBREG['c'] in pnr_uri:
        return URIRef(pnr_uri.replace(SUBREG['c'], LDFPNR['subregion_']))
    else:
        logging.critical("Cannot convert PNR URI <%s> to LDF equivalent", pnr_uri)
        return None
        
def scaleuri_to_value(scaleuri):
    return int(scaleuri.replace(SCALE['c'], ''))

logging.info("Adding PNR mappings and hierarchy...")
count = 0
for ysauri, target in mappings.subject_objects(SKOS.closeMatch):
    logging.info("---")
    try:
        ysalabel = ysa.preferredLabel(ysauri)[0][1]
    except IndexError:
        logging.warning("YSA URI <%s> has disappeared, skipping", ysauri)
        continue
    pnruri = ldf_to_pnr_uri(target)
    logging.info("YSA: <%s> '%s'", ysauri, ysalabel)
    out.add((ysauri, SKOS.prefLabel, Literal(ysalabel, 'fi')))
    pnrdata = Graph()
    starttime = time.time()
    pnrdata.parse(pnruri, format='xml')
    endtime = time.time()
    logging.debug('Loaded %d triples from %s in %.2f seconds', len(pnrdata), pnruri, endtime-starttime)
    try: # prefer Finnish
        pnrlabel = pnrdata.preferredLabel(pnruri,lang='fi')[0][1]
    except IndexError: # but any language will do
        try:

# 需要导入模块: from rdflib import Graph [as 别名]
# 或者: from rdflib.Graph import preferredLabel [as 别名]
    return uris

with open(sys.argv[1], 'rb') as fh:
    reader = MARCReader(fh)
    for rec in reader:
        changed_fields = set()
        recid = rec['001'].value()
        uri = ysa_uri(recid) # FIXME check 024
        # Virojoki (joki), Itämeri, Joensuu, Joroinen -- Kolma
        if uri not in (YSA['Y107962'], YSA['Y105038'], YSA['Y94166'], YSA['Y112746']):
            continue
        logging.info("YSA URI: <%s>", uri)
        
        # check for BT relationships and add
        for broader in enriched.objects(uri, SKOS.broader):
            btlabel = enriched.preferredLabel(broader, lang='fi')[0][1]
            remove_existing_551(rec, btlabel)
            logging.info("BT: <%s> '%s'", broader, btlabel)
            rec.add_ordered_field(
                Field(
                    tag='551',
                    indicators = [' ', ' '],
                    subfields = ['w', 'g'] + label_to_subfields(btlabel)
                    ))
            changed_fields.add('551')
        
        # check for NT relationships and add
        for narrower in enriched.objects(uri, SKOS.narrower):
            ntlabel = enriched.preferredLabel(narrower, lang='fi')[0][1]
            remove_existing_551(rec, ntlabel)
            logging.info("NT: <%s> '%s'", narrower, ntlabel)

# 需要导入模块: from rdflib import Graph [as 别名]
# 或者: from rdflib.Graph import preferredLabel [as 别名]
label = repo.get_label('uusi')
need_inspection = repo.get_issues(state='closed', labels=[label])
ysa_skos = Graph().parse(ysa_file, format='turtle')
yse_skos = Graph().parse(yse_file, format='turtle')

def delete_triples(uri):
    yse_skos.remove((URIRef(uri), None, None))
    yse_skos.remove((None, None, URIRef(uri)))

def replaced_by(uri, title):
    new_uri = lookup(title)
    if(new_uri != ''):
        yse_skos.remove((None, None, URIRef(uri)))
        yse_skos.add((URIRef(uri), dct.isReplacedBy, URIRef(new_uri)))
        yse_skos.add((URIRef(uri), OWL.deprecated, Literal('true',datatype=XSD.boolean)))

for issue in need_inspection:
    if ('**Termiehdotus Fintossa:**' in issue.body):
        md_href = issue.body.split("**Termiehdotus Fintossa:**")[1]
        if md_href:
            suggestion_uri = ysa + md_href.split('page/')[1].replace(')', '')
            yselab = yse_skos.preferredLabel(suggestion_uri, lang='fi')
            ysalab = yse_skos.preferredLabel(suggestion_uri, lang='fi')
            if yselab != '' and yselab == ysalab: # the suggestion has been taken into YSA with the same prefLabel
                print "deleting: " + issue.title
                delete_triples(suggestion_uri)
            else:
                print('add replacedBy to:' + suggestion_uri)

yse_skos.serialize(destination=yse_file, format='turtle')