Fix #228 - Add name field to `DataSet`

distributionviewer/api/migrations/0007_dataset_name.py

@@ -0,0 +1,35 @@
+# -*- coding: utf-8 -*-
+# Generated by Django 1.10.6 on 2017-03-29 18:24
+from __future__ import unicode_literals
+
+from django.db import migrations, models
+
+
+def update_name_from_date(apps, schema_editor):
+    DataSet = apps.get_model('api', 'DataSet')
+    for ds in DataSet.objects.all():
+        ds.name = ds.date.isoformat()
+        ds.save()
+
+
+class Migration(migrations.Migration):
+
+    dependencies = [
+        ('api', '0006_on_delete_cascade_dataset'),
+    ]
+
+    operations = [
+        migrations.AddField(
+            model_name='dataset',
+            name='name',
+            field=models.CharField(default='update-me', max_length=50),
+            preserve_default=False,
+        ),
+        migrations.RunPython(update_name_from_date, migrations.RunPython.noop),
+        migrations.AlterField(
+            model_name='dataset',
+            name='name',
+            field=models.CharField(max_length=50, unique=True),
+            preserve_default=False,
+        ),
+    ]

distributionviewer/api/models.py

@@ -3,6 +3,7 @@ from django.db import models
 
 
 class DataSet(models.Model):
+    name = models.CharField(max_length=50, unique=True)
     date = models.DateField()
     display = models.BooleanField(default=False)
     import_start = models.DateTimeField(null=True)

distributionviewer/api/tests/test_views.py

@@ -33,7 +33,8 @@ class TestMetric(TestCase):
             id=2, name='Searches Per Active Day', description='Searches descr',
             type='N')[0]
 
-        dataset, _ = DataSet.objects.get_or_create(date=date, display=True)
+        dataset, _ = DataSet.objects.get_or_create(name=date.isoformat(),
+                                                   date=date, display=True)
 
         cat_data = [
             ('x86', 0.95, 1),
@@ -125,7 +126,9 @@ class TestMetric(TestCase):
         # Test that a newer dataset with display=False isn't returned in
         # the API.
         self.create_data()
-        DataSet.objects.create(date=datetime.date(2016, 2, 2), display=False)
+        newdate = datetime.date(2016, 2, 2)
+        DataSet.objects.create(name=newdate.isoformat(), date=newdate,
+                               display=False)
         response = self.client.get(reverse('metric', args=['1']),
                                    data={'date': '2016-02-05'})
         self.assertEqual(response.status_code, 200)

notebooks/aggregate-and-import.py

@@ -176,8 +176,8 @@ conn, cursor = get_database_connection()
 process_date = datetime.datetime.strptime(environ.get('date'), '%Y%m%d').date()
 
 # Ensure we haven't already imported a dataset for this process_date.
-sql = "SELECT 1 FROM api_dataset WHERE date=%s"
-params = [process_date]
+sql = "SELECT 1 FROM api_dataset WHERE name=%s"
+params = [process_date.isoformat()]
 if DEBUG_SQL:
     print sql, params
 else:
@@ -200,9 +200,9 @@ df = df.filter("application_name_mode='Firefox'")
 columns = df.columns
 
 # Create a new dataset for this import.
-sql = ("INSERT INTO api_dataset (date, display, import_start) "
-       "VALUES (%s, %s, %s) RETURNING id")
-params = [process_date, False, datetime.datetime.now()]
+sql = ("INSERT INTO api_dataset (name, date, display, import_start) "
+       "VALUES (%s, %s, %s, %s) RETURNING id")
+params = [process_date.isoformat(), process_date, False, datetime.datetime.now()]
 if DEBUG_SQL:
     dataset_id = 0
     print sql, params