Validaciones sobre generación de dumps

Atacan casos borde en donde no existen series de tiempo sobre los nodos, o no se pudieron indexar correctamente
datosgobar · Dec 11, 2018 · 8676360 · 8676360
1 parent a65e554
commit 8676360
Show file tree

Hide file tree

Showing 4 changed files with 32 additions and 1 deletion.
diff --git a/series_tiempo_ar_api/apps/dump/generator/dta.py b/series_tiempo_ar_api/apps/dump/generator/dta.py
@@ -24,6 +24,11 @@ def generate_dump(self, dump_name):
         dump_file = DumpFile.objects.filter(file_type=DumpFile.TYPE_CSV,
                                             file_name=dump_name,
                                             node=self.node).last()
+
+        if dump_file is None:
+            GenerateDumpTask.info(self.task, f"No hay dumps CSV generados para el nodo {self.node.catalog_id}")
+            return
+
         df = pd.read_csv(dump_file.file)
         if dump_name == DumpFile.FILENAME_VALUES:
             df = df[constants.STATA_VALUES_COLS]

diff --git a/series_tiempo_ar_api/apps/dump/generator/generator.py b/series_tiempo_ar_api/apps/dump/generator/generator.py
@@ -34,8 +34,13 @@ def init_data(self):
         fields = Field.objects.all()
 
         if self.catalog:
+            try:
+                catalog = Catalog.objects.get(identifier=self.catalog)
+            except Catalog.DoesNotExist:
+                return
+
             fields = Field.objects.filter(
-                distribution__dataset__catalog=Catalog.objects.get(identifier=self.catalog)
+                distribution__dataset__catalog=catalog
             )
 
         fields = fields.filter(
@@ -72,6 +77,10 @@ def init_data(self):
             }
 
     def generate(self):
+        if not self.fields:
+            GenerateDumpTask.info(self.task, f"No hay series cargadas para el catálogo {self.catalog}")
+            return
+
         FullCsvGenerator(self.task, self.fields, self.catalog).generate()
         ValuesCsvGenerator(self.task, self.fields, self.catalog).generate()
         SourcesCsvGenerator(self.task, self.fields, self.catalog).generate()

diff --git a/series_tiempo_ar_api/apps/dump/tests/csv_generator_tests.py b/series_tiempo_ar_api/apps/dump/tests/csv_generator_tests.py
@@ -37,6 +37,16 @@ def setUpClass(cls):
         gen = DumpGenerator(cls.task)
         gen.generate()
 
+    def test_invalid_catalog(self):
+        task = GenerateDumpTask()
+        task.save()
+
+        gen = DumpGenerator(task, "no_catalog")
+        gen.generate()
+        task.refresh_from_db()
+
+        self.assertFalse(DumpFile.objects.filter(file_type=DumpFile.TYPE_CSV, task=task))
+
     def test_values_dump(self):
         file = self.task.dumpfile_set.get(file_name=DumpFile.FILENAME_VALUES).file
         reader = read_file_as_csv(file)

diff --git a/series_tiempo_ar_api/apps/dump/tests/dta_generator_tests.py b/series_tiempo_ar_api/apps/dump/tests/dta_generator_tests.py
@@ -3,6 +3,7 @@
 import faker
 import pandas as pd
 from django.test import TestCase
+from django_datajsonar.models import Node
 
 from series_tiempo_ar_api.apps.dump.generator import constants
 from series_tiempo_ar_api.apps.dump.generator.dta import DtaGenerator
@@ -25,6 +26,12 @@ def setUpClass(cls):
         enqueue_write_csv_task()
         super(DtaGeneratorTests, cls).setUpClass()
 
+    def test_generate_dta_no_csv_loaded(self):
+        node = Node.objects.create(catalog_id="empty_catalog", catalog_url="test.com", indexable=True)
+        task = GenerateDumpTask.objects.create()
+        DtaGenerator(task.id).generate()
+        self.assertFalse(DumpFile.objects.filter(node=node, file_type=DumpFile.TYPE_DTA))
+
     def test_generate_dta(self):
         task = GenerateDumpTask.objects.create()
         DtaGenerator(task.id).generate()