Merge branch 'master' into release/1.4.3

.checklist.yaml

@@ -0,0 +1,30 @@
+apiVersion: quintoandar.com.br/checklist/v2
+kind: ServiceChecklist
+metadata:
+  name: butterfree
+spec:
+  description: >-
+    A solution for Feature Stores.
+
+  costCenter: C055
+  department: engineering
+  lifecycle: production
+  docs: true
+
+  ownership:
+    team: data_products_mlops
+    line: tech_platform
+    owner: otavio.cals@quintoandar.com.br
+
+  libraries:
+    - name: butterfree
+      type: common-usage
+      path: https://quintoandar.github.io/python-package-server/
+      description: A lib to build Feature Stores.
+      registries:
+      - github-packages
+      tier: T0
+
+  channels:
+    squad: 'mlops'
+    alerts: 'data-products-reports'

.github/workflows/skip_lint.yml

@@ -0,0 +1,17 @@
+# This step is used only because we want to mark the runner-linter check as required
+# for PRs to develop, but not for the merge queue to merge into develop,
+# github does not have this functionality yet
+
+name: 'Skip github-actions/runner-linter check at merge queue'
+
+on:
+  merge_group:
+
+jobs:
+  empty_job:
+    name: 'github-actions/runner-linter'
+    runs-on: github-actions-developers-runner
+    steps:
+      - name: Skip github-actions/runner-linter check at merge queue
+        run: |
+          echo "Done"

butterfree/migrations/database_migration/cassandra_migration.py

@@ -89,6 +89,7 @@ def _get_alter_column_type_query(self, column: Diff, table_name: str) -> str:
             Alter column type query.
 
         """
+
         temp_column_name = f"{column.column}_temp"
 
         add_temp_column_query = (

butterfree/transform/aggregated_feature_set.py

@@ -605,6 +605,7 @@ def construct(
 
             # keeping this logic to maintain the same behavior for already implemented
             # feature sets
+
             if self._windows[0].slide == "1 day":
                 base_df = self._get_base_dataframe(
                     client=client, dataframe=output_df, end_date=end_date
@@ -636,6 +637,10 @@ def construct(
         else:
             output_df = self._aggregate(output_df, features=self.features)
 
+        output_df = self.incremental_strategy.filter_with_incremental_strategy(
+            dataframe=output_df, start_date=start_date, end_date=end_date
+        )
+
         output_df = output_df.select(*self.columns).replace(  # type: ignore
             float("nan"), None
         )