Exemplo n.º 1
0
    def get(self, request, *args, **kwargs):
        target_uri = os.path.splitext(request.build_absolute_uri())[0].replace('/data/', '/resource/')
        if not self.request.path.startswith("/data"):
            target_uri = re.sub('/[a-z]{2}/resource/', '/resource/', target_uri, count=1)
        if target_uri.endswith('graph'):
            target_uri = re.sub("/graph", "", target_uri)
        extension_ = self.kwargs.get('extension')

        rdf_format = mime_to_extension(get_lod_mime_type(extension_, self.request))
        if rdf_format == "rdf":
            rdf_format = "xml"

        resolved_uri = RDFRecord.get_internal_rdf_base_uri(target_uri)
        if "/resource/cache/" in target_uri:
            # old lookup rdfstore.get_rdfstore().get_cached_source_uri(target_uri)
            target_uri = target_uri.split('/resource/cache/')[-1]
            if 'geonames.org' in target_uri:
                target_uri = '{}/'.format(target_uri)
            if CacheResource.objects.filter(document_uri=target_uri).exists():
                cache_object = CacheResource.objects.filter(document_uri=target_uri).first()
                content = cache_object.get_graph().serialize(format=rdf_format)
            else:
                raise UnknownGraph("URI {} is not known in our graph store".format(target_uri))
        elif settings.RDF_USE_LOCAL_GRAPH:
            mode = self.request.REQUEST.get('mode', 'default')
            acceptance = True if mode == 'acceptance' else False
            local_object = ElasticSearchRDFRecord(source_uri=resolved_uri)
            local_object.get_graph_by_source_uri(uri=resolved_uri)
            if not local_object.exists():
                # todo: temporary work around for EDMRecords not saved with subjects
                logger.warn("Unable to find graph for: {}".format(resolved_uri))
                raise UnknownGraph("URI {} is not known in our graph store".format(resolved_uri))
            mode = self.get_mode(request)
            if mode in ['context', 'api', 'api-flat']:
                # get_graph(with_mappings=True, include_mapping_target=True, acceptance=acceptance)
                content = local_object.get_context_graph(with_mappings=True, include_mapping_target=True)
                if mode in ['api', 'api-flat']:
                    bindings = GraphBindings(about_uri=resolved_uri, graph=content)
                    index_doc = bindings.to_index_doc() if mode == 'api' else bindings.to_flat_index_doc()
                    content = json.dumps(index_doc)
                    rdf_format = 'json-ld'
                else:
                    content = content.serialize(format=rdf_format)
            else:
                content = local_object.get_graph()
                content = content.serialize(format=rdf_format)
        elif self.store.ask(uri=resolved_uri):
            target_uri = resolved_uri
            content = self.get_content(target_uri, rdf_format, request)
        return HttpResponse(
            content,
            content_type='{}; charset=utf8'.format(result_extension_to_mime(rdf_format))
        )
Exemplo n.º 2
0
 def get_content(self, target_uri, rdf_format, request):
     if '/resource/aggregation' in target_uri:
         target_named_graph = "{}/graph".format(target_uri.rstrip('/'))
         mode = self.get_mode(request, 'graph')
         describe = self.get_graph(mode=mode, named_graph=target_named_graph)
     else:
         mode = self.get_mode(request, "describe")
         describe = self.get_graph(mode=mode, uri=target_uri)
     if not describe:
         logger.warn("Unable to find graph for: {}".format(target_uri))
         raise Http404()
     if mode in ['api', 'api-flat']:
         bindings = GraphBindings(about_uri=target_uri, graph=describe)
         index_doc = bindings.to_index_doc() if mode == 'api' else bindings.to_flat_index_doc()
         content = json.dumps(index_doc)
         rdf_format = 'json'
     else:
         content = describe.serialize(format=rdf_format)
     return content, rdf_format
Exemplo n.º 3
0
    def create_es_action(self, action="index", record_type=None, index=settings.SITE_NAME, store=None, doc_type=None,
                         context=True, flat=True, exclude_fields=None, acceptance=False):
        if doc_type is None:
            doc_type = self._generate_doc_type()
        if record_type is None:
            record_type = self.get_rdf_type()
        if not store:
            store = rdfstore.get_rdfstore()

        if acceptance:
            index = "{}_acceptance".format(index)

        if record_type == "http://www.openarchives.org/ore/terms/Aggregation":
            record_type = "mdr"

        if action == "delete":
            return {
                '_op_type': action,
                '_index': index,
                '_type': doc_type,
                '_id': self.hub_id
            }

        graph = None

        if not context:
            graph = self.get_graph()
        else:
            graph, nr_levels = self.get_context_graph(store=store, named_graph=self.named_graph)
            graph.namespace_manager = namespace_manager

        bindings = GraphBindings(
            about_uri=self.source_uri,
            graph=graph
        )
        index_doc = bindings.to_flat_index_doc() if flat else bindings.to_index_doc()
        if exclude_fields:
            index_doc = {k: v for k, v in index_doc.items() if k not in exclude_fields}
        # add delving spec for default searchability
        index_doc["delving_spec"] = [
            {'@type': "Literal",
             'value': self.get_spec_name(),
             'raw': self.get_spec_name(),
             'lang': None}
        ]
        logger.debug(index_doc)
        mapping = {
            '_op_type': action,
            '_index': index,
            '_type': doc_type,
            '_id': self.hub_id,
            '_source': index_doc
        }
        thumbnail = bindings.get_about_thumbnail
        mapping['_source']['system'] = {
            'slug': self.hub_id,
            'spec': self.get_spec_name(),
            'thumbnail': thumbnail if thumbnail else "",
            'preview': "detail/foldout/{}/{}".format(doc_type, self.hub_id),
            'caption': bindings.get_about_caption if bindings.get_about_caption else "",
            'about_uri': self.document_uri,
            'source_uri': self.source_uri,
            'graph_name': self.named_graph,
            'created_at': datetime.datetime.now().isoformat(),
            'modified_at': datetime.datetime.now().isoformat(),
            'source_graph': graph.serialize(format='nt', encoding="utf-8").decode(encoding="utf-8"),
            'proxy_resource_graph': None,
            'web_resource_graph': None,
            # 'about_type': [rdf_type.qname for rdf_type in bindings.get_about_resource().get_types()]
            # 'collections': None, todo find a way to add collections via link
        }
        data_owner = self.dataset.data_owner if hasattr(self, 'dataset') else None
        dataset_name = self.dataset.name if hasattr(self, 'dataset') else None
        mapping['_source']['legacy'] = {
            'delving_hubId': self.hub_id,
            'delving_recordType': record_type,
            'delving_spec': self.get_spec_name(),
            'delving_owner': data_owner,
            'delving_orgId': settings.ORG_ID,
            'delving_collection': dataset_name,
            'delving_title': self.get_first_literal(DC.title, graph),
            'delving_creator': self.get_first_literal(DC.creator, graph),
            'delving_description': self.get_first_literal(DC.description, graph),
            'delving_provider': index_doc.get('edm_provider')[0].get('value') if 'edm_provider' in index_doc else None,
            'delving_hasGeoHash': "true" if bindings.has_geo() else "false",
            'delving_hasDigitalObject': "true" if thumbnail else "false",
            'delving_hasLandingePage': "true" if 'edm_isShownAt' in index_doc else "false",
            'delving_hasDeepZoom': "true" if 'nave_deepZoom' in index_doc else "false",
        }
        return mapping
Exemplo n.º 4
0
    def retrieve(self, request, pk=None, format=None, *args, **kwargs):
        def get_mode(default=None):
            params = request.GET
            return params.get('schema', default)

        self._clean_callback(request)

        query = NaveESQuery(
            index_name=self.get_index_name,
            doc_types=self.doc_types,
            default_facets=self.facets,
            cluster_geo=False,
            size=1,
            converter=self.get_converter()
        )
        try:
            query = query.build_item_query(query, request.query_params, pk)
        except ValueError as ve:
            logger.error("Unable to build request because: {}".format(ve))
            # todo display error message when bad/unknown hubId is given
            return HttpResponseBadRequest()
        mlt = True if request.query_params.get('mlt', 'false') == "true" else False
        mlt_count = int(request.query_params.get('mlt.count', 5))
        mlt_filter_queries = request.query_params.getlist('mlt.qf', [])
        mlt_fq_dict = {}
        for fq in mlt_filter_queries:
            if ":" in fq:
                k, v = fq.split(":", maxsplit=1)
                mlt_fq_dict[k] = v
        record = ElasticSearchRDFRecord(hub_id=pk)
        record.get_graph_by_id(hub_id=pk)
        response = NaveItemResponse(
            query,
            self,
            index=self.get_index_name,
            mlt=mlt,
            mlt_count=mlt_count,
            mlt_filter_query=mlt_fq_dict,
            rdf_record=record
        )
        renderer_format = request.accepted_renderer.format
        if renderer_format in list(EXTENSION_TO_MIME_TYPE.keys()) and renderer_format not in ['xml', 'json']:
            graph = record.get_graph()
            graph_string = graph.serialize(format=renderer_format).decode('utf-8')
            mime_type = EXTENSION_TO_MIME_TYPE.get(renderer_format)
            return Response(data=graph_string, content_type=mime_type)
        target_uri = record.document_uri
        if settings.RDF_USE_LOCAL_GRAPH:
            graph = record.get_graph()
        else:
            store = rdfstore.get_rdfstore()
            graph, _ = RDFModel.get_context_graph(store, named_graph=record.named_graph)
        if not graph:
            from django.http import HttpResponseNotFound
            return HttpResponseNotFound()
        mode = get_mode(self.default_converter)
        bindings = GraphBindings(about_uri=target_uri, graph=graph)
        delving_fields = False if request.GET.get("delving_fields") == 'false' else True
        converter = None
        if mode in ['api', 'api-flat']:
            index_doc = bindings.to_index_doc() if mode == 'api' else bindings.to_flat_index_doc()
        elif mode in REGISTERED_CONVERTERS.keys():
            converter = REGISTERED_CONVERTERS.get(mode)
            index_doc = converter(
                bindings=bindings,
                graph=graph,
                about_uri=bindings.about_uri()
            ).convert(add_delving_fields=delving_fields)
        elif self.default_converter in REGISTERED_CONVERTERS.keys():
            converter = REGISTERED_CONVERTERS.get(self.default_converter)
            index_doc = converter(
                bindings=bindings,
                graph=graph,
                about_uri=bindings.about_uri()
            ).convert(add_delving_fields=delving_fields)
        else:
            logger.warn("unable to convert results to schema {}".format(mode))
            index_doc = bindings.to_index_doc()
        layout_fields = OrderedDict()
        layout_fields['layout'] = converter().get_layout_fields() if converter else []
        if response.get_mlt():
            mlt = {"item": [NaveESItemSerializer(item).data for item in response.get_mlt()]}
        else:
            mlt = ""
        result = {'result': {
            'layout': layout_fields,
            'item': {'fields': index_doc},
            "relatedItems": mlt}}
        return Response(result)