Exemplo n.º 1
0
class ElasticsearchConnection:
    """
    Wrapper class to handle the connection with Elasticsearch.
    Uses the `CEDAElasticsearchClient <https://github.com/cedadev/ceda-elasticsearch-tools>`_
    """

    def __init__(self, **kwargs):
        self.es = CEDAElasticsearchClient(**kwargs)

    def get_hits(self, index, query=None):
        return scan(self.es, query=query, index=index)

    def get_query(self, extensions, path, excludes=[]):
        query_base = {
            "_source": {
                "exclude": ["info.phenomena"]
            },
            "query": {
                "bool": {
                    "must": [
                        {
                            "match_phrase_prefix": {
                                "info.directory.analyzed": path
                            }
                        }
                    ],
                    "must_not": [],
                    "filter": []
                }
            }
        }

        for ext in extensions:
            filter = {
                "term": {
                    "info.type.keyword": ext
                }
            }

            query_base["query"]["bool"]["filter"].append(filter)

        for exclusion in excludes:
            query_base["query"]["bool"]["must_not"].append(exclusion)

        return query_base

    def bulk(self, iterator, *args, generator=False):
        if generator:
                bulk(self.es, iterator(*args), refresh=True)
        else:
            bulk(self.es, iterator, refresh=True)

    def count(self, *args, **kwargs):
        return self.es.count(*args, **kwargs).get('count')

    def mget(self, *args, **kwargs):
        return self.es.mget(*args, **kwargs)

    def search(self, *args, **kwargs):
        return self.es.search(*args, **kwargs)

    def create_collections_index(self, index):
        
        return self.es.indices.create(index=index, ignore=400, body={
            "mappings": {
                "properties": {
                    "time_frame": {
                        "type": "date_range"
                    },
                    "bbox": {
                        "properties": {
                            "coordinates": {
                                "type": "geo_point"
                            }
                        }
                    }
                }
            }
        })
class ElasticsearchConnection:
    """
    Elasticsearch Connection class. Uses `CEDAElasticsearchClient <https://github.com/cedadev/ceda-elasticsearch-tools>`_

    :param index: files index (default: settings.ELASTICSEARCH_INDEX)
    :type index: str

    """
    def __init__(self, index=settings.ELASTICSEARCH_INDEX):
        self.index = index
        self.collection_index = settings.ELASTICSEARCH_COLLECTION_INDEX
        self.es = CEDAElasticsearchClient(
            **settings.ELASTICSEARCH_CONNECTION_PARAMS)

    def search(self, query):
        """
        Search the files index

        :param query: Elasticsearch file query
        :type query: dict

        :return: Elasticsearch response
        :rtype: dict
        """
        return self.es.search(index=self.index, body=query)

    def search_collections(self, query):
        """
        Search the collections index

        :param query: Elasticsearch collection query
        :type query: dict

        :return: Elasticsearch response
        :rtype: dict
        """
        return self.es.search(index=self.collection_index, body=query)

    def count(self, query):
        """
        Return the hit count from the current file query

        :param query: Elasticsearch file query
        :type query: dict

        :return: Elasticsearch count response
        :rtype: dict
        """
        return self.es.count(index=self.index, body=query)

    def count_collections(self, query):
        """
        Return the hit count from the current collection query

        :param query: Elasticsearch collection query
        :type query: dict

        :return: Elasticsearch count response
        :rtype: dict
        """
        return self.es.count(index=self.collection_index, body=query)