Merge pull request #415 from Duncan-Hunter/main

db optional dependency

Author: nnansters

.github/workflows/dev.yml

@@ -35,7 +35,7 @@ jobs:
 
       - name: Install dependencies
         run: |
-          python -m pip install --upgrade pip
+          python -m pip install -e '.[db]' --upgrade pip
           pip install poetry tox tox-gh-actions
 
       - name: test with tox

.github/workflows/preview.yml

@@ -11,6 +11,10 @@ on:
   # Allows you to run this workflow manually from the Actions tab
   workflow_dispatch:
 
+permissions:
+  contents: write
+  id-token: write
+
 # A workflow run is made up of one or more jobs that can run sequentially or in parallel
 jobs:
   publish_dev_build:
@@ -50,9 +54,9 @@ jobs:
           poetry build
 
       - name: publish to Test PyPI
-        uses: pypa/gh-action-pypi-publish@master
+        uses: pypa/gh-action-pypi-publish@release/v1
         with:
           user: __token__
           password: ${{ secrets.TEST_PYPI_API_TOKEN}}
           repository_url: https://test.pypi.org/legacy/
-          skip_existing: true
+          skip-existing: true

Dockerfile

@@ -55,7 +55,7 @@ ENV POETRY_INSTALLER_MAX_WORKERS=10
 
 # Install project in editable mode and with development dependencies
 WORKDIR $APP_PATH
-RUN poetry install
+RUN poetry install --all-extras
 
 ENTRYPOINT ["poetry", "run"]
 CMD ["nml"]

README.md

@@ -149,6 +149,21 @@ docker -v /local/config/dir/:/config/ run nannyml/nannyml nml run
 python -m pip install git+https://github.com/NannyML/nannyml
 ```
 
+#### Extras
+
+If you're using database connections to read model inputs/outputs or you're exporting monitoring results to a database,
+you'll need to include the optional `db` dependency. For example using `pip`:
+
+```bash
+pip install nannyml[db]
+```
+
+or using `poetry`
+
+```bash
+poetry install nannyml --all-extras
+```
+
 ### Quick Start
 
 _The following snippet is based on our [latest release](https://github.com/NannyML/nannyml/releases/latest)_.

docs/installing_nannyml.rst

@@ -32,6 +32,21 @@ or
 
 Any issues with installation? `Let us know`_ so we can help you.
 
+Extras
+------
+
+If you're using database connections to read model inputs/outputs or you're exporting monitoring results to a database,
+you'll need to include the optional `db` dependency. For example using `pip`:
+
+.. code-block:: bash
+
+    $ pip install nannyml[db]
+
+or using `poetry`
+
+.. code-block:: bash
+
+    $ poetry install nannyml --all-extras
 
 .. _`Let us know`: https://github.com/NannyML/nannyml/issues
 .. _`LightGBM`: https://github.com/microsoft/LightGBM

docs/tutorials/working_with_results.rst

@@ -202,6 +202,9 @@ results to disk using a :class:`~nannyml.io.raw_files_writer.RawFilesWriter`, se
 :class:`~nannyml.io.db.database_writer.DatabaseWriter`. This example will show how to use the
 :class:`~nannyml.io.db.database_writer.DatabaseWriter`.
 
+In order to get the dependencies required for database access, please ensure you've installed the
+optional `db` dependency. Check the :ref:`installation instructions<installing_nannyml>` for more information.
+
 We construct the :class:`~nannyml.io.db.database_writer.DatabaseWriter` by providing a database connection string.
 Upon calling the :meth:`~nannyml.io.base.Writer.write` method, all results will be written into
 the database, in this case, an `SQLite` database.

nannyml/__init__.py

@@ -31,15 +31,17 @@
 # Dev branch marker is: 'X.Y.dev' or 'X.Y.devN' where N is an integer.
 # 'X.Y.dev0' is the canonical version of 'X.Y.dev'
 #
-__version__ = '0.11.0'
+__version__ = '0.12.0'
 
 import logging
 
 from dotenv import load_dotenv
+from importlib import import_module
+
 
 from .calibration import Calibrator, IsotonicCalibrator, needs_calibration
 from .chunk import Chunk, Chunker, CountBasedChunker, DefaultChunker, PeriodBasedChunker, SizeBasedChunker
-from .data_quality import MissingValuesCalculator, UnseenValuesCalculator, NumericalRangeCalculator
+from .data_quality import MissingValuesCalculator, NumericalRangeCalculator, UnseenValuesCalculator
 from .datasets import (
     load_modified_california_housing_dataset,
     load_synthetic_binary_classification_dataset,
@@ -59,7 +61,7 @@
     UnivariateDriftCalculator,
 )
 from .exceptions import ChunkerException, InvalidArgumentsException, MissingMetadataException
-from .io import DatabaseWriter, PickleFileWriter, RawFilesWriter
+from .io import PickleFileWriter, RawFilesWriter
 from .performance_calculation import PerformanceCalculator
 from .performance_estimation import CBPE, DLE
 from .stats import (
@@ -71,6 +73,20 @@
 )
 from .usage_logging import UsageEvent, disable_usage_logging, enable_usage_logging, log_usage
 
+
+_optional_dependencies = {
+    'DatabaseWriter': '.io.db',
+}
+
+
+def __getattr__(name: str):
+    optional_module_path = _optional_dependencies.get(name)
+    if optional_module_path is not None:
+        module = import_module(optional_module_path, package=__name__)
+        return getattr(module, name)
+    raise AttributeError(f"module {__name__} has no attribute {name}")
+
+
 try:
     import nannyml_premium
 

nannyml/io/__init__.py

@@ -12,13 +12,27 @@
 
 The `store` module implements an object cache, meant to cache fitted calculators in between runs.
 """
+from importlib import import_module
 
 from .base import Reader, Writer, WriterFactory
-from .db import DatabaseWriter
 from .file_reader import FileReader
 from .file_writer import FileWriter
 from .pickle_file_writer import PickleFileWriter
 from .raw_files_writer import RawFilesWriter
 from .store import FilesystemStore, JoblibPickleSerializer, Serializer, Store
 
+
+_optional_dependencies = {
+    'DatabaseWriter': '.db'
+}
+
+
+def __getattr__(name):
+    optional_module_path = _optional_dependencies.get(name)
+    if optional_module_path is not None:
+        module = import_module(optional_module_path, package=__name__)
+        return getattr(module, name)
+    raise AttributeError(f"module {__name__} has no attribute {name}")
+
+
 DEFAULT_WRITER = RawFilesWriter(path='out')

nannyml/io/db/database_writer.py

@@ -1,6 +1,11 @@
 from typing import Any, Dict, Optional
 
-from sqlmodel import Session, SQLModel, create_engine, select
+try:
+    from sqlmodel import Session, SQLModel, create_engine, select
+except ImportError:
+    raise ImportError(
+        "`sqlmodel` module is not available. Please install the `nannyml[db]` extra to use this functionality."
+    )
 
 from nannyml._typing import Result
 from nannyml.exceptions import WriterException

nannyml/io/db/entities.py

@@ -9,12 +9,17 @@
 """
 
 import sys
-
 from datetime import datetime
 from typing import List, Optional
 
 from pydantic import ConfigDict
-from sqlmodel import Field, Relationship, SQLModel
+
+try:
+    from sqlmodel import Field, Relationship, SQLModel
+except ImportError:
+    raise ImportError(
+        "`sqlmodel` module is not available. Please install the `nannyml[db]` extra to use this functionality."
+    )
 
 
 class Model(SQLModel, table=True):  # type: ignore[call-arg]