predicthq · jthi3rry · Jun 20, 2016 · Jun 16, 2016
diff --git a/predicthq/endpoints/schemas.py b/predicthq/endpoints/schemas.py
@@ -129,6 +129,20 @@ class Options:
     exact = ListType(StringType)
 
 
+class Signal(Model):
+
+    class Options:
+        serialize_when_none = False
+
+    id = StringType(required=True)
+    analysis_from = DateTimeType()
+    analysis_to = DateTimeType()
+    analysis_tz = StringType(choices=pytz.all_timezones)
+    significance = FloatType(min_value=0, max_value=100)
+    metric = StringType(choices=('demand', 'lead', 'span'))
+    explain = DateType()
+
+
 class DateTimeRange(Model):
 
     class Options:

diff --git a/predicthq/endpoints/v1/events/schemas.py b/predicthq/endpoints/v1/events/schemas.py
@@ -4,7 +4,7 @@
 from predicthq.endpoints.schemas import PaginatedMixin, SortableMixin, Model, ResultSet, \
     ListType, StringType, GeoJSONPointType, StringListType, StringModelType, Area, \
     ModelType, IntRange, IntType, DateTimeRange, DateTimeType, FloatType, ResultType, \
-    DictType, DateType, Place
+    DictType, DateType, Place, Signal
 
 
 class SearchParams(PaginatedMixin, SortableMixin, Model):
@@ -24,6 +24,7 @@ class Options:
     country = ListType(StringType)
     within = StringListType(StringModelType(Area), separator="+")
     place = ModelType(Place)
+    signal = ModelType(Signal)
 
 
 class Event(Model):

diff --git a/predicthq/endpoints/v1/signals/schemas.py b/predicthq/endpoints/v1/signals/schemas.py
@@ -154,34 +154,27 @@ class Dimensions(Model):
     # @todo: Support custom dimensions from signal
 
 
-class DailyAnalysisDetailsComponent(Model):
+class MeanAnalysisComponent(Model):
 
-    count = IntType()
-    min = IntType()
-    max = IntType()
-    avg = FloatType()
-    sum = IntType()
-    sum_of_squares = IntType()
-    variance = FloatType()
+    mean = FloatType()
     std_deviation = FloatType()
-    std_deviation_bounds = DictType(FloatType)
-    percentiles = DictType(FloatType)
+    expected = FloatType()
+    excess = FloatType()
 
 
-class DailyAnalysisDetails(Model):
+class CountAnalysisComponent(Model):
 
-    lead = ModelType(DailyAnalysisDetailsComponent)
-    span = ModelType(DailyAnalysisDetailsComponent)
+    count = IntType()
+    expected = FloatType()
+    excess = FloatType()
 
 
 class DailyAnalysis(Model):
 
     date = DateType()
-    trend = FloatType()
-    actual = FloatType()
-    expected = FloatType()
-    excess = FloatType()
-    details = ModelType(DailyAnalysisDetails)
+    demand = ModelType(CountAnalysisComponent)
+    lead = ModelType(MeanAnalysisComponent)
+    span = ModelType(MeanAnalysisComponent)
 
 
 class AnalysisResultSet(ResultSet):
@@ -199,9 +192,7 @@ class Options:
     initiated = ModelType(DateTimeRange)
     completed = ModelType(DateTimeRange)
     within = StringListType(StringModelType(Area), separator="+")
-    significance = IntType(min_value=0, max_value=100)
-    lead = BooleanType(default=False)
-    span = BooleanType(default=False)
+    significance = FloatType(min_value=0, max_value=100)
     place = ModelType(Place)
 
     # @todo: Support custom dimensions from signal
diff --git a/setup.py b/setup.py
@@ -40,7 +40,7 @@ def read(*parts):
     ],
     tests_require=[
         "nose==1.3.7",
-        "coverage==4.0.3",
+        "coverage>=4.0.0",
         "responses==0.5.1",
         "mock==1.3.0",
     ],

diff --git a/tests/endpoints/v1/events_tests.py b/tests/endpoints/v1/events_tests.py
@@ -16,12 +16,14 @@ def test_search_params(self, client):
         client.events.search(id="id", q="query", rank_level=[4,5], rank__gt=85, country=["NZ", "AU"],
              within__radius="2km", within__longitude=-71.0432, within__latitude=42.346,
              label=["label1", "label2"], category="category",
-             start__gte="2016-03-01", start__lt=datetime(2016, 4, 1), start__tz="Pacific/Auckland",)
+             start__gte="2016-03-01", start__lt=datetime(2016, 4, 1), start__tz="Pacific/Auckland",
+             signal__id='zVNLr8tHvWQw', signal__explain=datetime(2016, 4, 1))
 
         client.request.assert_called_once_with('get', '/v1/events/', params={
             'id': 'id', 'rank.gt': 85, 'rank_level': '4,5', 'category': 'category', 'country': 'NZ,AU',
             'within': '2km@42.346,-71.0432', 'label': 'label1,label2', 'q': 'query',
-            'start.lt': '2016-04-01T00:00:00.000000', 'start.gte': '2016-03-01T00:00:00.000000', 'start.tz': 'Pacific/Auckland'})
+            'start.lt': '2016-04-01T00:00:00.000000', 'start.gte': '2016-03-01T00:00:00.000000', 'start.tz': 'Pacific/Auckland',
+            'signal.id': 'zVNLr8tHvWQw', 'signal.explain': '2016-04-01'})
 
     @with_mock_client()
     def test_search_for_account(self, client):

diff --git a/tests/endpoints/v1/signals_tests.py b/tests/endpoints/v1/signals_tests.py
@@ -158,6 +158,6 @@ def test_summary(self, client, responses):
     @with_mock_responses()
     def test_analysis(self, client, responses):
         signal = client.signals.get(id="zVNLr8tHvWQw")
-        analysis = signal.analysis(lead=True, span=True)
+        analysis = signal.analysis()
         self.assertEqual(analysis.count, 365)
         # @todo: add tests for custom signal dimensions
diff --git a/tests/fixtures/requests_responses/signals_test/test_analysis.json b/tests/fixtures/requests_responses/signals_test/test_analysis.json
@@ -26,7 +26,7 @@
   {
     "method": "GET",
     "match_querystring": true,
-    "url": "/v1/signals/zVNLr8tHvWQw/analysis/?span=1&lead=1",
+    "url": "/v1/signals/zVNLr8tHvWQw/analysis/",
     "status": 200,
     "content_type": "application/json",
     "body": {
@@ -35,122 +35,64 @@
       "previous": null,
       "results": [
         {
-          "actual": 329,
-          "trend": 326.7875867369319,
-          "expected": 407.6649175062919,
-          "details": {
-            "span": {
-              "count": 329,
-              "min": 0,
-              "sum_of_squares": 11645,
-              "max": 29,
-              "sum": 1575,
-              "std_deviation": 3.532354314042595,
-              "percentiles": {
-                "5.0": 0,
-                "25.0": 3,
-                "1.0": 0,
-                "95.0": 11,
-                "75.0": 6,
-                "99.0": 16,
-                "50.0": 4
-              },
-              "std_deviation_bounds": {
-                "upper": 11.851942670638381,
-                "lower": -2.2774745855319978
-              },
-              "variance": 12.47752699993533,
-              "avg": 4.787234042553192
-            },
-            "lead": {
-              "count": 329,
-              "min": 0,
-              "sum_of_squares": 729639,
-              "max": 281,
-              "sum": 7951,
-              "std_deviation": 40.41899871791781,
-              "percentiles": {
-                "5.0": 0,
-                "25.0": 1.125,
-                "1.0": 0,
-                "95.0": 115.19999999999959,
-                "75.0": 27.999999999999996,
-                "99.0": 189.35999999999888,
-                "50.0": 9
-              },
-              "std_deviation_bounds": {
-                "upper": 105.00517068811526,
-                "lower": -56.67082418355598
-              },
-              "variance": 1633.6954573590413,
-              "avg": 24.167173252279635
-            }
+          "date": "2014-01-01T00:00:00",
+          "demand": {
+            "count": 319,
+            "expected": 319,
+            "excess": 0
           },
-          "excess": -78.66491750629189,
-          "date": "2015-01-01T00:00:00"
+          "lead": {
+            "expected": 29.877742946708462,
+            "std_deviation": 44.753600299118915,
+            "excess": 0,
+            "mean": 29.877742946708462
+          },
+          "span": {
+            "expected": 5.554858934169279,
+            "std_deviation": 4.659836385200691,
+            "mean": 5.554858934169279,
+            "excess": 0
+          }
         },
         {
-          "actual": 342,
-          "trend": 325.63983679188283,
-          "expected": 475.5862667153484,
-          "details": {
-            "span": {
-              "count": 342,
-              "min": 0,
-              "sum_of_squares": 14740,
-              "max": 34,
-              "sum": 1762,
-              "std_deviation": 4.0688854916316,
-              "percentiles": {
-                "5.0": 1,
-                "25.0": 2,
-                "1.0": 0,
-                "95.0": 13,
-                "75.0": 7,
-                "99.0": 19,
-                "50.0": 4.166666666666667
-              },
-              "std_deviation_bounds": {
-                "upper": 13.28981776688893,
-                "lower": -2.9857241996374695
-              },
-              "variance": 16.555829144010126,
-              "avg": 5.152046783625731
-            },
-            "lead": {
-              "count": 342,
-              "min": 0,
-              "sum_of_squares": 1107796,
-              "max": 301,
-              "sum": 10220,
-              "std_deviation": 48.43731469724757,
-              "percentiles": {
-                "5.0": 0,
-                "25.0": 2,
-                "1.0": 0,
-                "95.0": 131,
-                "75.0": 37.5,
-                "99.0": 227.20999999999952,
-                "50.0": 9
-              },
-              "std_deviation_bounds": {
-                "upper": 126.75767033016766,
-                "lower": -66.99158845882263
-              },
-              "variance": 2346.1734550801957,
-              "avg": 29.883040935672515
-            }
+          "date": "2014-01-02T00:00:00",
+          "demand": {
+            "count": 427,
+            "expected": 427,
+            "excess": 0
+          },
+          "lead": {
+            "expected": 33.95316159250586,
+            "std_deviation": 48.977819938991054,
+            "excess": 0,
+            "mean": 33.95316159250586
           },
-          "excess": -133.58626671534842,
-          "date": "2015-01-02T00:00:00"
+          "span": {
+            "expected": 4.948477751756441,
+            "std_deviation": 3.819359814466576,
+            "mean": 4.948477751756441,
+            "excess": 0
+          }
         },
         {
-          "actual": 0,
-          "trend": 4.974674771228433,
-          "expected": 3.9252131844241966,
-          "details": {},
-          "excess": -3.9252131844241966,
-          "date": "2015-12-07T00:00:00"
+          "date": "2014-12-24T00:00:00",
+          "demand": {
+            "count": 621,
+            "expected": 509.0973596475694,
+            "excess": 111.90264035243055
+          },
+          "lead": {
+            "expected": 35.075696145743926,
+            "std_deviation": 49.04110699946923,
+            "excess": 1.9967676223720208,
+            "mean": 37.072463768115945
+          },
+          "span": {
+            "expected": 5.040257648953301,
+            "std_deviation": 3.590556938479606,
+            "mean": 5.040257648953301,
+            "excess": 0
+          }
         }
       ]
     }

diff --git a/tests/fixtures/requests_responses/signals_test/test_summary.json b/tests/fixtures/requests_responses/signals_test/test_summary.json
@@ -57,26 +57,28 @@
           -27.46846
         ]
       },
-      "country": {
-        "type": "category",
-        "options": [
-          {
-            "count": 2445,
-            "name": "NZ"
-          },
-          {
-            "count": 282,
-            "name": "AU"
-          }
-        ]
-      },
-      "revenue": {
-        "count": 2729,
-        "min": 0,
-        "max": 754.1099853515625,
-        "std_deviation": 76.9167076432597,
-        "avg": 51.462359763100366,
-        "type": "number"
+      "custom": {
+        "country": {
+          "type": "category",
+          "options": [
+            {
+              "count": 2445,
+              "name": "NZ"
+            },
+            {
+              "count": 282,
+              "name": "AU"
+            }
+          ]
+        },
+        "revenue": {
+          "count": 2729,
+          "min": 0,
+          "max": 754.1099853515625,
+          "std_deviation": 76.9167076432597,
+          "avg": 51.462359763100366,
+          "type": "number"
+        }
       }
     }
   }