A lot of improvements (#28)

.pre-commit-config.yaml

@@ -30,31 +30,13 @@ repos:
         files: 'cloud\/.*\.py'
 
   # Static type and code checkers below
-
-  - repo: https://github.com/PyCQA/flake8
-    rev: 6.0.0
+  - repo: https://github.com/charliermarsh/ruff-pre-commit
+    rev: 'v0.6.0'
     hooks:
-      - id: flake8
-        additional_dependencies:
-          - flake8-comprehensions
-          - flake8-print
-          - flake8-mutable
-          - flake8-pytest-style
-          - flake8-printf-formatting
-          - 'flake8-simplify==0.19.2'
-          - 'flake8-type-checking==2.1.3'
-        args: [ '--enable-extensions=G' ]
+      - id: ruff
+        args: [ --fix ]
 
   - repo: https://github.com/python-poetry/poetry
     rev: 1.5.0
     hooks:
       - id: poetry-check
-
-
-  - repo: https://github.com/pre-commit/mirrors-mypy
-    rev: v0.942
-    hooks:
-      - id: mypy
-        files: 'aladdin/.*|tests/.*'
-        additional_dependencies:
-          - types-freezegun

Dockerfile

@@ -7,14 +7,11 @@ RUN poetry config virtualenvs.create false
 RUN pip install pip --upgrade
 
 COPY ./pyproject.toml /opt/app/pyproject.toml
-# COPY ./poetry.lock /opt/app/poetry.lock
+COPY ./poetry.lock /opt/app/poetry.lock
 
 RUN mkdir /opt/app/aligned
 RUN poetry install --no-dev --no-root --extras "redis psql server aws"
 
 COPY ./aligned /opt/app/aligned
 
-# COPY /. opt/app/aligned
-
 ENTRYPOINT ["python", "-m", "aligned.cli"]
-# RUN pip install -U 'opt/app/aligned[redis,aws,psql,server,text]'

Dockerfile.test

@@ -7,9 +7,10 @@ RUN poetry config virtualenvs.create false
 RUN pip install pip --upgrade
 
 COPY ./pyproject.toml /opt/app/pyproject.toml
+COPY ./poetry.lock /opt/app/poetry.lock
 
 RUN mkdir /opt/app/aligned
-RUN poetry install --no-root --all-extras
+RUN poetry install --no-root --extras "redis pandera kafka ollama mlflow lancedb"
 
 COPY ./aligned /opt/app/aligned
 COPY ./conftest.py /opt/app/conftest.py

aligned/__init__.py

@@ -24,7 +24,7 @@
 from aligned.data_source.stream_data_source import HttpStreamSource
 from aligned.data_source.batch_data_source import CustomMethodDataSource
 from aligned.feature_store import ContractStore, FeatureStore
-from aligned.feature_view import feature_view, combined_feature_view, check_schema
+from aligned.feature_view import feature_view, check_schema
 from aligned.schemas.text_vectoriser import EmbeddingModel
 from aligned.sources.kafka import KafkaConfig
 from aligned.sources.local import FileSource, Directory, ParquetConfig, CsvConfig
@@ -77,7 +77,6 @@
     'Json',
     'EmbeddingModel',
     'feature_view',
-    'combined_feature_view',
     'model_contract',
     # Aggregation
     'CustomAggregation',

aligned/active_learning/job.py

@@ -1,9 +1,11 @@
+from __future__ import annotations
+
 import logging
 from dataclasses import dataclass
 
-import pandas as pd
 import polars as pl
 
+from aligned.lazy_imports import pandas as pd
 from aligned.active_learning.selection import ActiveLearningMetric, ActiveLearningSelection
 from aligned.active_learning.write_policy import ActiveLearningWritePolicy
 from aligned.retrival_job import RetrivalJob

aligned/checks.py

@@ -37,7 +37,7 @@ def as_markdown(self) -> str:
 def feature_exist(feature: FeatureReference, store: ContractStore) -> bool:
 
     loc = feature.location
-    if loc.location == 'model':
+    if loc.location_type == 'model':
         model = store.model(loc.name).model
         all_features = model.predictions_view.full_schema
     else:

aligned/cli.py

@@ -152,7 +152,12 @@ async def compile(repo_path: str, reference_file: str, env_file: str, ignore_fil
 
         repo_def = await RepoReader.definition_from_path(dir, excludes)
 
-        await file.write(repo_def.to_json(omit_none=True).encode('utf-8'))
+        data = repo_def.to_json(omit_none=True)
+        if isinstance(data, str):
+            data_bytes = data.encode('utf-8')
+        else:
+            data_bytes = data
+        await file.write(data_bytes)
     else:
         click.echo(f'No repo file found at {dir}')