Update to Atoti 0.9.0

.github/actions/setup/action.yaml

@@ -1,13 +1,15 @@
-name: Setup the repository
-description: Set up Python, install Poetry and the dependencies
+name: Set up the project
+description: Install uv, Python, and the project dependencies
 runs:
   using: "composite"
   steps:
-    - run: pipx install poetry==1.7.1
-      shell: bash
-    - uses: actions/setup-python@v5
+    - uses: astral-sh/setup-uv@v3
       with:
-        cache: poetry
-        python-version: "3.9.18"
-    - run: poetry install
+        enable-cache: true
+        version: "0.4.10"
+
+    - run: uv python install 3.10
+      shell: bash
+
+    - run: uv sync --locked
       shell: bash

.github/workflows/test.yml

@@ -10,8 +10,7 @@ jobs:
     steps:
       - uses: actions/checkout@v4
       - uses: ./.github/actions/setup
-      - run: poetry check --lock
-      - run: poetry run ruff format --check
-      - run: poetry run ruff check
-      - run: poetry run mypy
-      - run: poetry run pytest
+      - run: uv run ruff format --check
+      - run: uv run ruff check
+      - run: uv run mypy
+      - run: uv run pytest

.gitignore

@@ -1,2 +1,3 @@
 __pycache__/
+.venv/
 content/

CONTRIBUTING.md

@@ -4,4 +4,4 @@
 
 This project is not intended to become a showcase of all Atoti's features.
 It should have just enough structure to show how to start a project on sound and modular foundations, not more.
-https://docs.atoti.io is the right place to discover and learn what Atoti can do.
+The right place to teach everything Atoti can do is https://docs.atoti.io.

Dockerfile

@@ -1,22 +1,28 @@
-# syntax=docker/dockerfile:1.2
+# Inspired from https://github.com/astral-sh/uv-docker-example/blob/dee88a8c43be3b16b0ad58f0daee5eaee7e2157a/multistage.Dockerfile.
 
-# `--platform=linux/am64` is required to build this image on macOS with Apple Silicon until https://github.com/activeviam/jdk4py/issues/73 is done.
-FROM --platform=linux/amd64 python:3.9.18-slim AS builder
+FROM ghcr.io/astral-sh/uv:0.4.10-python3.10-bookworm-slim AS builder
 
-RUN pip install poetry==1.7.1
+ENV UV_COMPILE_BYTECODE=1 UV_LINK_MODE=copy
 
-COPY poetry.lock pyproject.toml ./
+WORKDIR /venv
 
-RUN POETRY_VIRTUALENVS_CREATE=false poetry install --no-cache --no-root --only main --sync
+RUN --mount=type=cache,target=/root/.cache/uv \
+    --mount=type=bind,source=uv.lock,target=uv.lock \
+    --mount=type=bind,source=pyproject.toml,target=pyproject.toml \
+    uv sync --frozen --no-dev --no-install-project
 
-FROM --platform=linux/amd64 python:3.9.18-slim AS runner
+# Keep this synced with the builder image.
+FROM python:3.10-slim-bookworm
 
-ENV ATOTI_HIDE_EULA_MESSAGE=true
-ENV PORT=80
+COPY --from=builder /venv app
+
+ENV PATH="/app/.venv/bin:$PATH"
 
-COPY --from=builder /usr/local/lib/python3.9/site-packages /usr/local/lib/python3.9/site-packages
 COPY app app
 
-ENTRYPOINT ["python", "-u", "-m", "app"]
+ENV ATOTI_HIDE_EULA_MESSAGE=true
+ENV PORT=80
 
 EXPOSE $PORT
+
+CMD ["python", "-u", "-m", "app"]

README.md

@@ -4,7 +4,7 @@ This template can be used to start Atoti projects where the goal is to [go into
 
 On top of the `atoti` package, it comes with:
 
-- Dependency management with [Poetry](https://python-poetry.org)
+- Dependency management with [uv](https://docs.astral.sh/uv)
 - Config management with [Pydantic Settings](https://docs.pydantic.dev/2.6/concepts/pydantic_settings)
 - Testing with [pytest](https://docs.pytest.org)
 - Type checking with [mypy](http://mypy-lang.org)
@@ -15,26 +15,24 @@ On top of the `atoti` package, it comes with:
 
 ### Installation
 
-- [Install `poetry`](https://python-poetry.org/docs/#installation)
+- [Install `uv`](https://docs.astral.sh/uv/getting-started/installation)
 - Install the dependencies:
 
   ```bash
-  poetry install
+  uv sync
   ```
 
 ### Commands
 
 To start the app:
 
 ```bash
-poetry run python -m main
+uv run python -m app
 ```
 
 Other useful commands can be found in [`test.yml`](.github/workflows/test.yml).
 
-## Variants
+## Deployment
 
-This repository has the following long-lived branches showcasing different aspects:
-
-- [`deploy-to-aws`](https://github.com/atoti/project-template/tree/deploy-to-aws) for deploying on AWS ECS.
-- [`deploy-to-heroku`](https://github.com/atoti/project-template/tree/deploy-to-heroku) for a one-click deploy to Heroku.
+This repository is configured to continuously deploy to AWS ECS.
+To deploy somewhere else, delete [`task-definition.json`][task-definition.json] and adapt [`deploy.yml`](.github/workflows/deploy.yml).

app/__init__.py

@@ -1,5 +1,3 @@
-from __future__ import annotations
-
-from .config import *
-from .constants import *
-from .start_app import *
+from .config import Config as Config
+from .constants import *  # noqa: F403
+from .start_app import start_app as start_app

app/__main__.py

@@ -1,7 +1,8 @@
-from __future__ import annotations
+from urllib.parse import urlparse
 
 from . import Config, start_app
 
 with start_app(config=Config()) as session:
-    print(f"Session listening on port {session.port}")  # noqa: T201
+    port = urlparse(session.url) or 80
+    print(f"Session listening on port {port}")  # noqa: T201
     session.wait()

app/config.py

@@ -1,11 +1,6 @@
-# ruff: noqa: UP007
-# Pydantic evaluates type annotations at runtime which does not support `|`.
-
-from __future__ import annotations
-
 from datetime import timedelta
 from pathlib import Path
-from typing import Annotated, Optional, Union
+from typing import Annotated
 
 from pydantic import (
     AliasChoices,
@@ -19,7 +14,7 @@
 )
 from pydantic_settings import BaseSettings, SettingsConfigDict
 
-from .util import normalize_postgres_dsn_for_atoti_sql
+from .util import normalize_postgres_dsn_for_atoti_jdbc
 
 
 class Config(BaseSettings):
@@ -30,27 +25,25 @@ class Config(BaseSettings):
 
     model_config = SettingsConfigDict(frozen=True)
 
-    data_refresh_period: Optional[timedelta] = timedelta(minutes=1)
+    data_refresh_period: timedelta | None = timedelta(minutes=1)
 
     # The $PORT environment variable is used by most PaaS to indicate the port the app server should bind to.
     port: int = 9090
 
-    requests_timeout: timedelta = timedelta(seconds=30)
-
-    reverse_geocoding_path: Union[HttpUrl, FilePath] = TypeAdapter(
-        HttpUrl
-    ).validate_python("https://api-adresse.data.gouv.fr/reverse/csv/")
+    reverse_geocoding_path: HttpUrl | FilePath = TypeAdapter(HttpUrl).validate_python(
+        "https://api-adresse.data.gouv.fr/reverse/csv/"
+    )
 
     user_content_storage: Annotated[
-        Optional[Union[PostgresDsn, Path]],
+        PostgresDsn | Path | None,
         Field(
             # $DATABASE_URL is used by some PaaS such to designate the URL of the app's primary database.
             # For instance: https://devcenter.heroku.com/articles/heroku-postgresql#designating-a-primary-database.
             validation_alias=AliasChoices("user_content_storage", "database_url")
         ),
     ] = Path("content")
 
-    velib_data_base_path: Union[HttpUrl, DirectoryPath] = TypeAdapter(
+    velib_data_base_path: HttpUrl | DirectoryPath = TypeAdapter(
         HttpUrl
     ).validate_python(
         "https://velib-metropole-opendata.smovengo.cloud/opendata/Velib_Metropole"
@@ -61,6 +54,6 @@ class Config(BaseSettings):
     def normalize_postgres_dsn(cls, value: object) -> object:
         try:
             postgres_dsn: PostgresDsn = TypeAdapter(PostgresDsn).validate_python(value)
-            return normalize_postgres_dsn_for_atoti_sql(postgres_dsn)
+            return normalize_postgres_dsn_for_atoti_jdbc(postgres_dsn)
         except ValueError:
             return value

app/constants.py

@@ -1,5 +1,3 @@
-from __future__ import annotations
-
 from enum import Enum
 
 

app/create_and_join_tables.py

@@ -1,5 +1,3 @@
-from __future__ import annotations
-
 import atoti as tt
 
 from .constants import StationDetailsTableColumn, StationStatusTableColumn, Table

app/create_cubes.py

@@ -1,5 +1,3 @@
-from __future__ import annotations
-
 import atoti as tt
 
 from .constants import (
@@ -57,19 +55,16 @@ def create_station_cube(session: tt.Session, /) -> None:
         }
     )
 
-    m.update(
-        {
-            StationCubeMeasure.BIKES.value: tt.agg.sum(
-                station_status_table[StationStatusTableColumn.BIKES.value]
-            ),
-            StationCubeMeasure.CAPACITY.value: tt.agg.sum(
-                tt.agg.single_value(
-                    station_details_table[StationDetailsTableColumn.CAPACITY.value]
-                ),
-                scope=tt.OriginScope(l[StationCubeStationLevel.ID.value]),
+    with session.data_model_transaction():
+        m[StationCubeMeasure.BIKES.value] = tt.agg.sum(
+            station_status_table[StationStatusTableColumn.BIKES.value]
+        )
+        m[StationCubeMeasure.CAPACITY.value] = tt.agg.sum(
+            tt.agg.single_value(
+                station_details_table[StationDetailsTableColumn.CAPACITY.value]
             ),
-        }
-    )
+            scope=tt.OriginScope({l[StationCubeStationLevel.ID.value]}),
+        )
 
 
 def create_cubes(session: tt.Session, /) -> None:

app/load_tables.py

@@ -1,7 +1,4 @@
-from __future__ import annotations
-
 from collections.abc import Iterable, Mapping
-from datetime import timedelta
 from pathlib import Path
 from typing import Any, cast
 
@@ -17,14 +14,11 @@
 def read_station_details(
     *,
     reverse_geocoding_path: HttpUrl | Path,
-    timeout: timedelta,
     velib_data_base_path: HttpUrl | Path,
 ) -> pd.DataFrame:
     stations_data: Any = cast(
         Any,
-        read_json(
-            velib_data_base_path, Path("station_information.json"), timeout=timeout
-        ),
+        read_json(velib_data_base_path, Path("station_information.json")),
     )["data"]["stations"]
     station_information_df = pd.DataFrame(stations_data)[
         ["station_id", "name", "capacity", "lat", "lon"]
@@ -48,7 +42,7 @@ def read_station_details(
     )
 
     reverse_geocoded_df = reverse_geocode(
-        coordinates, reverse_geocoding_path=reverse_geocoding_path, timeout=timeout
+        coordinates, reverse_geocoding_path=reverse_geocoding_path
     ).rename(
         columns={
             "department": StationDetailsTableColumn.DEPARTMENT.value,
@@ -64,15 +58,10 @@ def read_station_details(
     ).drop(columns=coordinates_column_names)
 
 
-def read_station_status(
-    velib_data_base_path: HttpUrl | Path,
-    /,
-    *,
-    timeout: timedelta,
-) -> pd.DataFrame:
+def read_station_status(velib_data_base_path: HttpUrl | Path, /) -> pd.DataFrame:
     stations_data = cast(
         Any,
-        read_json(velib_data_base_path, Path("station_status.json"), timeout=timeout),
+        read_json(velib_data_base_path, Path("station_status.json")),
     )["data"]["stations"]
     station_statuses: list[Mapping[str, Any]] = []
     for station_status in stations_data:
@@ -97,14 +86,12 @@ def read_station_status(
 def load_tables(session: tt.Session, /, *, config: Config) -> None:
     station_details_df = read_station_details(
         reverse_geocoding_path=config.reverse_geocoding_path,
-        timeout=config.requests_timeout,
         velib_data_base_path=config.velib_data_base_path,
     )
     station_status_df = read_station_status(
         config.velib_data_base_path,
-        timeout=config.requests_timeout,
     )
 
-    with session.start_transaction():
+    with session.tables.data_transaction():
         session.tables[Table.STATION_DETAILS.value].load_pandas(station_details_df)
         session.tables[Table.STATION_STATUS.value].load_pandas(station_status_df)

app/start_app.py

@@ -1,5 +1,3 @@
-from __future__ import annotations
-
 from collections.abc import Generator
 from contextlib import contextmanager, nullcontext
 
@@ -13,8 +11,13 @@
 
 @contextmanager
 def start_app(*, config: Config) -> Generator[tt.Session, None, None]:
-    with start_session(config=config) as session, run_periodically(
-        lambda: load_tables(session, config=config),
-        period=config.data_refresh_period,
-    ) if config.data_refresh_period else nullcontext():
+    with (
+        start_session(config=config) as session,
+        run_periodically(
+            lambda: load_tables(session, config=config),
+            period=config.data_refresh_period,
+        )
+        if config.data_refresh_period
+        else nullcontext(),
+    ):
         yield session