DS-3791: Groupid aggregation

.circleci/config.yml

@@ -26,15 +26,12 @@ test_settings: &test_settings
           chmod +x codecov
           ./codecov -F "$(basename $PWD | sed s/[^a-z]/_/g)"
 
-
-
 ####################
 # Jobs: see https://circleci.com/docs/2.0/jobs-steps/
 ####################
 
 version: 2
 jobs:
-
   py310:
     <<: *test_settings
     docker:
@@ -44,12 +41,12 @@ jobs:
     docker:
       - image: cimg/python:3.10
     steps:
-    - checkout
-    - run:
-        name: Run linting
-        command: |
-          pip install ruff
-          ruff check src/ tests/
+      - checkout
+      - run:
+          name: Run linting
+          command: |
+            pip install -r requirements.txt
+            ruff check src/ tests/
 
   # Runs when the repository is tagged for release; see the workflows section
   # below for trigger logic.
@@ -133,7 +130,7 @@ workflows:
       - deploy:
           filters:
             tags:
-              only: /[0-9]{4}.[0-9]{1,2}.[0-9]+/  # Calver: YYYY.M.MINOR
+              only: /[0-9]{4}.[0-9]{1,2}.[0-9]+/ # Calver: YYYY.M.MINOR
             branches:
               # Ignore all branches; this workflow should only run for tags.
               ignore: /.*/

requirements-dev.in

@@ -226,7 +226,7 @@ more-itertools==10.4.0
     #   -c requirements.txt
     #   jaraco-classes
     #   jaraco-functools
-mozilla-metric-config-parser==2024.7.1
+mozilla-metric-config-parser==2024.8.1
     # via
     #   -c requirements.txt
     #   mozanalysis

requirements-dev.txt

@@ -953,9 +953,9 @@ more-itertools==10.4.0 \
     #   -r requirements-dev.in
     #   jaraco-classes
     #   jaraco-functools
-mozilla-metric-config-parser==2024.7.1 \
-    --hash=sha256:07ba32624cb9a38662bdff259dd4ec385bdc8ca4500b461d8a9025a736a25c6b \
-    --hash=sha256:ad5169b4cec7b0fa013b3136584b7ee3d9b8a6a807d2eb71e5d9f269517f9d22
+mozilla-metric-config-parser==2024.8.1 \
+    --hash=sha256:49fb6e67367809e3750108246e50dc1e68778564c2c158ba26bd4835b0b9c89c \
+    --hash=sha256:cadc4ba9fb8399be0b857abb4bdd1f12ff1943159463fec0d128b71b2e72b554
     # via -r requirements-dev.in
 mozilla-nimbus-schemas==2023.10.3 \
     --hash=sha256:8771344a63b0d197dbebd8d7955ce8034d0f5063a13899f4da7d8a99803a76da \

requirements.in

@@ -124,7 +124,7 @@ more-itertools==10.4.0
     # via
     #   jaraco-classes
     #   jaraco-functools
-mozilla-metric-config-parser==2024.7.1
+mozilla-metric-config-parser==2024.8.1
     # via mozanalysis
 mozilla-nimbus-schemas==2023.10.3
     # via mozilla-metric-config-parser

requirements.txt

@@ -924,9 +924,9 @@ more-itertools==10.4.0 \
     #   -r requirements.in
     #   jaraco-classes
     #   jaraco-functools
-mozilla-metric-config-parser==2024.7.1 \
-    --hash=sha256:07ba32624cb9a38662bdff259dd4ec385bdc8ca4500b461d8a9025a736a25c6b \
-    --hash=sha256:ad5169b4cec7b0fa013b3136584b7ee3d9b8a6a807d2eb71e5d9f269517f9d22
+mozilla-metric-config-parser==2024.8.1 \
+    --hash=sha256:49fb6e67367809e3750108246e50dc1e68778564c2c158ba26bd4835b0b9c89c \
+    --hash=sha256:cadc4ba9fb8399be0b857abb4bdd1f12ff1943159463fec0d128b71b2e72b554
     # via -r requirements.in
 mozilla-nimbus-schemas==2023.10.3 \
     --hash=sha256:8771344a63b0d197dbebd8d7955ce8034d0f5063a13899f4da7d8a99803a76da \

src/mozanalysis/config.py

@@ -6,6 +6,9 @@
 
 from metric_config_parser.config import ConfigCollection
 
+from mozanalysis.metrics import DataSource, Metric
+from mozanalysis.segments import Segment, SegmentDataSource
+
 METRIC_HUB_JETSTREAM_REPO = "https://github.com/mozilla/metric-hub/tree/main/jetstream"
 
 
@@ -68,12 +71,11 @@ def check_configs_for_app(self, app_name: str) -> bool:
                 return True
         return False
 
-    def get_metric(self, metric_slug: str, app_name: str):
+    def get_metric(self, metric_slug: str, app_name: str) -> Metric:
         """Load a metric definition for the given app.
 
         Returns a :class:`mozanalysis.metrics.Metric` instance.
         """
-        from mozanalysis.metrics import Metric
 
         metric_definition = self.configs.get_metric_definition(metric_slug, app_name)
         if metric_definition is None:
@@ -100,12 +102,11 @@ def get_metric(self, metric_slug: str, app_name: str):
             app_name=app_name,
         )
 
-    def get_data_source(self, data_source_slug: str, app_name: str):
+    def get_data_source(self, data_source_slug: str, app_name: str) -> DataSource:
         """Load a data source definition for the given app.
 
         Returns a :class:`mozanalysis.metrics.DataSource` instance.
         """
-        from mozanalysis.metrics import DataSource
 
         data_source_definition = self.configs.get_data_source_definition(
             data_source_slug, app_name
@@ -120,26 +121,13 @@ def get_data_source(self, data_source_slug: str, app_name: str):
                     f"Could not find application {app_name}, so data source {data_source_slug} could not be resolved"  # noqa:E501
                 )
 
-        return DataSource(
-            name=data_source_definition.name,
-            from_expr=data_source_definition.from_expression,
-            client_id_column=data_source_definition.client_id_column,
-            submission_date_column=data_source_definition.submission_date_column,
-            experiments_column_type=(
-                None
-                if data_source_definition.experiments_column_type == "none"
-                else data_source_definition.experiments_column_type
-            ),
-            default_dataset=data_source_definition.default_dataset,
-            app_name=app_name,
-        )
+        return DataSource.from_mcp_data_source(data_source_definition, app_name)
 
-    def get_segment(self, segment_slug: str, app_name: str):
+    def get_segment(self, segment_slug: str, app_name: str) -> Segment:
         """Load a segment definition for the given app.
 
         Returns a :class:`mozanalysis.segments.Segment` instance.
         """
-        from mozanalysis.segments import Segment
 
         segment_definition = self.configs.get_segment_definition(segment_slug, app_name)
         if segment_definition is None:
@@ -165,12 +153,13 @@ def get_segment(self, segment_slug: str, app_name: str):
             app_name=app_name,
         )
 
-    def get_segment_data_source(self, data_source_slug: str, app_name: str):
+    def get_segment_data_source(
+        self, data_source_slug: str, app_name: str
+    ) -> SegmentDataSource:
         """Load a segment data source definition for the given app.
 
         Returns a :class:`mozanalysis.segments.SegmentDataSource` instance.
         """
-        from mozanalysis.segments import SegmentDataSource
 
         data_source_definition = self.configs.get_segment_data_source_definition(
             data_source_slug, app_name
@@ -196,7 +185,9 @@ def get_segment_data_source(self, data_source_slug: str, app_name: str):
             app_name=app_name,
         )
 
-    def get_outcome_metric(self, metric_slug: str, outcome_slug: str, app_name: str):
+    def get_outcome_metric(
+        self, metric_slug: str, outcome_slug: str, app_name: str
+    ) -> Metric:
         """Load a metric definition from an outcome defined for the given app.
 
         Parametrized metrics are not supported, since they may not be defined outside
@@ -230,18 +221,21 @@ class MinimalConfiguration:
         summaries = metric_definition.resolve(outcome_spec, conf, self.configs)
         metric = summaries[0].metric
 
+        if metric.data_source is None:
+            raise ValueError(f"Unable to resolve DataSource for Metric {metric.name}")
+
         return Metric(
             name=metric.name,
             select_expr=metric.select_expression,
             friendly_name=metric.friendly_name,
             description=metric.description,
-            data_source=metric.data_source,
+            data_source=DataSource.from_mcp_data_source(metric.data_source, app_name),
             bigger_is_better=metric.bigger_is_better,
         )
 
     def get_outcome_data_source(
         self, data_source_slug: str, outcome_slug: str, app_name: str
-    ):
+    ) -> DataSource:
         """Load a data source definition from an outcome defined for the given app.
 
         Returns a :class:`mozanalysis.metrics.DataSource` instance.
@@ -266,17 +260,8 @@ def get_outcome_data_source(
                 + f" in outcome {outcome_slug}"
             )
 
-        return DataSource(
-            name=data_source_definition.name,
-            from_expr=data_source_definition.from_expression,
-            client_id_column=data_source_definition.client_id_column,
-            submission_date_column=data_source_definition.submission_date_column,
-            experiments_column_type=(
-                None
-                if data_source_definition.experiments_column_type == "none"
-                else data_source_definition.experiments_column_type
-            ),
-            default_dataset=data_source_definition.default_dataset,
+        return DataSource.from_mcp_data_source(
+            data_source_definition, app_name=app_name
         )