def setUp(self):
self.elastic = ElasticInstance.get()
self.task = ReadDataJsonTask.objects.create()
self.meta_task = IndexMetadataTask.objects.create()
fake_index.doc_type(self.FakeField)
fake_index.create()
self.FakeField.init(using=self.elastic)
class MetadataConfig(AppConfig):
name = 'series_tiempo_ar_api.apps.metadata'
es_configurations = settings.ES_CONFIGURATION
es_urls = es_configurations["ES_URLS"]
client_options = es_configurations["CONNECTIONS"]["default"]
ElasticInstance.init(es_urls, client_options)
def init_index(index_name):
elastic_instance = ElasticInstance.get()
index = Index(index_name, using=elastic_instance)
add_analyzer(index)
if not index.exists():
index.create()
Metadata.init(using=elastic_instance, index=index_name)
return index
def index_catalog(catalog_id, catalog_path, index, node=None):
"""Indexa un catálogo. Útil para tests"""
if not node:
node = Node(catalog_id=catalog_id,
catalog_url=catalog_path,
indexable=True)
catalog = DataJson(node.catalog_url)
node.catalog = json.dumps(catalog)
node.save()
task = ReadDataJsonTask()
task.save()
read_datajson(task, read_local=True, whitelist=True)
for distribution in Distribution.objects.filter(
dataset__catalog__identifier=catalog_id):
DistributionIndexer(index=index).run(distribution)
ElasticInstance.get().indices.forcemerge(index=index)
def __init__(self, index, series_id, rep_mode, args, collapse_agg=None):
self.index = index
self.elastic = ElasticInstance.get()
self.series_id = series_id
self.rep_mode = rep_mode
self.args = args.copy()
self.collapse_agg = collapse_agg or constants.API_DEFAULT_VALUES[
constants.PARAM_COLLAPSE_AGG]
self.search = self.init_search()
def __init__(self,
task,
index=settings.TS_INDEX,
output_directory=constants.DUMP_DIR):
self.task = task
self.index = index
self.output_directory = output_directory
self.fields = {}
self.catalog_themes = {}
self.elastic = ElasticInstance.get()
self.init_fields_dict()
def scheduler():
task = ReadDataJsonTask.objects.last()
if task.status == task.FINISHED:
return
if not get_queue('indexing').jobs:
ReportGenerator(task).generate()
elastic = ElasticInstance.get()
if elastic.indices.exists(index=settings.TS_INDEX):
elastic.indices.forcemerge(index=settings.TS_INDEX)
def __init__(self, node: Node, task: IndexMetadataTask, doc_type):
self.node = node
self.task = task
self.doc_type = doc_type
self.elastic = ElasticInstance.get()
self.fields_meta = {}
self.init_fields_meta_cache()
try:
data_json = DataJson(node.catalog_url)
themes = data_json['themeTaxonomy']
self.themes = self.get_themes(themes)
except Exception:
raise ValueError("Error de lectura de los themes del catálogo")
def setUpClass(cls):
super(ViewTests, cls).setUpClass()
es_client = ElasticInstance.get()
if es_client.indices.exists(cls.index):
es_client.indices.delete(cls.index)
es_client.indices.create(cls.index, body=INDEX_CREATION_BODY)
cls.catalog_id = 'csv_dump_test_catalog'
path = os.path.join(samples_dir, 'distribution_daily_periodicity.json')
index_catalog(cls.catalog_id, path, cls.index)
cls.task = CSVDumpTask()
cls.task.save()
gen = CSVDumpGenerator(cls.task,
index=cls.index,
output_directory=cls.directory)
gen.generate()
def __init__(self, node: Node, task: IndexMetadataTask, index: str):
self.node = node
self.task = task
self.index_name = index
self.elastic: Elasticsearch = ElasticInstance.get()
if not self.elastic.indices.exists(self.index_name):
init_index(self.index_name)
self.fields_meta = {}
self.init_fields_meta_cache()
try:
data_json = DataJson(node.catalog_url)
themes = data_json['themeTaxonomy']
self.themes = self.get_themes(themes)
except Exception:
raise ValueError("Error de lectura de los themes del catálogo")
def __init__(self, index):
"""
args:
index (str): Índice de Elasticsearch a ejecutar las queries.
"""
self.index = index
self.series = []
self.elastic = ElasticInstance()
self.data = None
self.count = None
# Parámetros que deben ser guardados y accedidos varias veces
self.args = {
constants.PARAM_START:
constants.API_DEFAULT_VALUES[constants.PARAM_START],
constants.PARAM_LIMIT:
constants.API_DEFAULT_VALUES[constants.PARAM_LIMIT],
constants.PARAM_SORT:
constants.API_DEFAULT_VALUES[constants.PARAM_SORT]
}
def tseries_index(name: str) -> Index:
index = Index(name, using=ElasticInstance.get())
# Fija el límite superior de valores en una respuesta. Si filtramos por serie, sería
# la cantidad de valores máximas que puede tener una única serie temporal.
index.settings(max_result_window=settings.MAX_SERIES_VALUES)
if not index.exists():
index.create()
index.put_mapping(doc_type=settings.TS_DOC_TYPE,
body=constants.MAPPING)
index.save()
# Actualizo el mapping
mapping = index.get_mapping(doc_type=settings.TS_DOC_TYPE)
doc_properties = mapping[name]['mappings'][
settings.TS_DOC_TYPE]['properties']
if not doc_properties.get('raw_value'):
index.put_mapping(doc_type=settings.TS_DOC_TYPE,
body=constants.MAPPING)
return index
def get_fields_meta_index():
fields_meta = Index(constants.FIELDS_INDEX, using=ElasticInstance.get())
add_analyzer(fields_meta)
return fields_meta
def tearDownClass(cls):
super(CSVDumpCommandTests, cls).tearDownClass()
ElasticInstance.get().indices.delete(cls.index)
Catalog.objects.all().delete()
Node.objects.all().delete()
def tearDownClass(cls):
ElasticInstance.get().indices.delete(cls.index)
Node.objects.all().delete()
class Meta:
dynamic = MetaField('strict')
doc_type = constants.METADATA_DOC_TYPE
using = ElasticInstance.get()
def setUpClass(cls):
Catalog.objects.all().delete()
cls.elastic = ElasticInstance()
super(IndexerTests, cls).setUpClass()
def tearDownClass(cls):
super(ViewTests, cls).tearDownClass()
ElasticInstance.get().indices.delete(cls.index)
DumpFile.objects.all().delete()
Node.objects.all().delete()
#! coding: utf-8
from django.conf import settings
from series_tiempo_ar_api.libs.indexing.elastic import ElasticInstance
from .helpers import setup_database
from .support.generate_data import get_generator
elastic = ElasticInstance.get()
def setup():
if not elastic.indices.exists(settings.TEST_INDEX):
generator = get_generator()
generator.run()
setup_database()
def teardown():
elastic.indices.delete(settings.TEST_INDEX)
def execute(self):
"""Ejecuta la query. Devuelve un diccionario con el siguiente formato
{
"limit": 20,
"offset": 0,
"count": 1,
"data": [
{
"title": "foo",
"description": "bar",
"id": "if-foo",
}
]
}
"""
self.validate()
if self.errors:
self.response['errors'] = self.errors
return self.response
es_client = ElasticInstance.get()
search = Metadata.search(using=es_client, index=constants.METADATA_ALIAS)
querystring = self.args.get(constants.PARAM_QUERYSTRING)
if querystring is not None:
search = search.query('match', all=querystring)
offset = self.args[constants.PARAM_OFFSET]
limit = self.args[constants.PARAM_LIMIT]
search = search[offset:limit + offset]
for arg, field in constants.FILTER_ARGS.items():
search = self.add_filters(search, arg, field)
response = search.execute()
self.response = {
'data': [],
'count': response.hits.total
}
for hit in response:
start_date = getattr(hit, 'start_date', None)
if start_date:
start_date = start_date.date()
end_date = getattr(hit, 'end_date', None)
if end_date:
end_date = end_date.date()
self.response['data'].append({
'field': {
'id': getattr(hit, 'id', None),
'description': getattr(hit, 'description', None),
'title': getattr(hit, 'title', None),
'frequency': getattr(hit, 'periodicity', None),
'time_index_start': start_date,
'time_index_end': end_date,
'units': getattr(hit, 'units', None),
},
'dataset': {
'title': getattr(hit, 'dataset_title', None),
'publisher': {
'name': getattr(hit, 'dataset_publisher_name', None),
},
'source': getattr(hit, 'dataset_source', None),
'theme': getattr(hit, 'dataset_theme', None),
}
})
self.response[constants.PARAM_LIMIT] = self.args[constants.PARAM_LIMIT]
self.response[constants.PARAM_OFFSET] = self.args[constants.PARAM_OFFSET]
return self.response
def __init__(self, task):
self.elastic: Elasticsearch = ElasticInstance.get()
self.task = task
from elasticsearch_dsl import Index
from django.test import TestCase
from django_datajsonar.tasks import read_datajson
from django_datajsonar.models import ReadDataJsonTask, Node, Field as datajsonar_Field
from series_tiempo_ar_api.apps.metadata.indexer.catalog_meta_indexer import CatalogMetadataIndexer
from series_tiempo_ar_api.apps.metadata.indexer.doc_types import Field
from series_tiempo_ar_api.apps.metadata.indexer.index import add_analyzer
from series_tiempo_ar_api.apps.metadata.models import IndexMetadataTask
from series_tiempo_ar_api.libs.indexing.elastic import ElasticInstance
from series_tiempo_ar_api.apps.management import meta_keys
SAMPLES_DIR = os.path.join(os.path.dirname(__file__), 'samples')
fake = faker.Faker()
fake_index = Index(fake.word(), using=ElasticInstance.get())
add_analyzer(fake_index)
class IndexerTests(TestCase):
class FakeField(Field):
class Meta:
index = fake_index._name
def setUp(self):
self.elastic = ElasticInstance.get()
self.task = ReadDataJsonTask.objects.create()
self.meta_task = IndexMetadataTask.objects.create()
fake_index.doc_type(self.FakeField)
fake_index.create()
self.FakeField.init(using=self.elastic)
def delete_metadata(fields: list):
es_instance = ElasticInstance.get()
search = Search(using=es_instance, index=constants.METADATA_ALIAS)
return search.filter('terms',
id=[field.identifier for field in fields]).delete()
import faker
from elasticsearch_dsl import Index, Search
from django.test import TestCase
from django_datajsonar.tasks import read_datajson
from django_datajsonar.models import ReadDataJsonTask, Node, Field as datajsonar_Field
from series_tiempo_ar_api.apps.metadata.indexer.catalog_meta_indexer import CatalogMetadataIndexer
from series_tiempo_ar_api.apps.metadata.indexer.index import add_analyzer
from series_tiempo_ar_api.apps.metadata.models import IndexMetadataTask
from series_tiempo_ar_api.libs.indexing.elastic import ElasticInstance
from series_tiempo_ar_api.apps.management import meta_keys
SAMPLES_DIR = os.path.join(os.path.dirname(__file__), 'samples')
fake = faker.Faker()
fake_index = Index(fake.pystr(max_chars=50).lower(), using=ElasticInstance.get())
add_analyzer(fake_index)
class IndexerTests(TestCase):
def setUp(self):
self.elastic = ElasticInstance.get()
self.task = ReadDataJsonTask.objects.create()
self.meta_task = IndexMetadataTask.objects.create()
def test_index(self):
index_ok = self._index(catalog_id='test_catalog', catalog_url='single_distribution.json')
search = Search(
index=fake_index._name,
using=self.elastic
def setUp(self):
self.elastic = ElasticInstance.get()
self.task = ReadDataJsonTask.objects.create()
self.meta_task = IndexMetadataTask.objects.create()
def tearDownClass(cls):
super(CSVTest, cls).tearDownClass()
ElasticInstance.get().indices.delete(cls.index)
Node.objects.all().delete()
def setUpClass(cls):
cls.elastic = ElasticInstance()
super(IndexerTests, cls).setUpClass()
def __init__(self, index: str):
self.elastic: Elasticsearch = ElasticInstance.get()
self.index_name = index
self.index = tseries_index(index)
def tearDown(self):
ElasticInstance.get().indices.delete(self.index)
Catalog.objects.all().delete()
DumpFile.objects.all().delete()
Node.objects.all().delete()
def tearDownClass(cls):
super(SQLGeneratorTests, cls).tearDownClass()
ElasticInstance.get().indices.delete(cls.index)
