Staging to release 0.8.0

.github/workflows/black.yml

@@ -6,6 +6,10 @@ jobs:
   lint:
     runs-on: ubuntu-latest
     steps:
-      - uses: actions/checkout@v2
-      - uses: actions/setup-python@v2
-      - uses: psf/black@20.8b1
+      - uses: actions/checkout@v4
+      - uses: actions/setup-python@v5
+        with:
+          python-version: "3.12"
+      - run: pip install ruff
+      - run: ruff check .
+      - run: ruff format --check .

.github/workflows/python-publish.yml

@@ -1,4 +1,4 @@
-# This workflows will upload a Python Package using Twine when a release is created
+# This workflow will upload a Python Package using Twine when a release is created
 # For more information see: https://help.github.com/en/actions/language-and-framework-guides/using-python-with-github-actions#publishing-to-package-registries
 
 name: Upload Python Package
@@ -9,23 +9,22 @@ on:
 
 jobs:
   deploy:
-
     runs-on: ubuntu-latest
 
     steps:
-    - uses: actions/checkout@v2
+    - uses: actions/checkout@v4
     - name: Set up Python
-      uses: actions/setup-python@v2
+      uses: actions/setup-python@v5
       with:
         python-version: '3.x'
     - name: Install dependencies
       run: |
         python -m pip install --upgrade pip
-        pip install setuptools wheel twine
+        pip install build twine
     - name: Build and publish
       env:
         TWINE_USERNAME: ${{ secrets.PYPI_USERNAME }}
         TWINE_PASSWORD: ${{ secrets.PYPI_PASSWORD }}
       run: |
-        python setup.py sdist bdist_wheel
+        python -m build
         twine upload dist/*

.pre-commit-config.yaml

@@ -1,20 +1,15 @@
 repos:
   - repo: https://github.com/pre-commit/pre-commit-hooks
-    rev: v3.4.0
+    rev: v5.0.0
     hooks:
       - id: trailing-whitespace
       - id: check-yaml
       - id: end-of-file-fixer
-      - id: requirements-txt-fixer
-      - id: trailing-whitespace
-
-  - repo: https://github.com/PyCQA/isort
-    rev: 5.7.0
-    hooks:
-      - id: isort
-        args: ["--profile", "black"]
+      - id: check-ast
 
-  - repo: https://github.com/psf/black
-    rev: 20.8b1
+  - repo: https://github.com/astral-sh/ruff-pre-commit
+    rev: v0.9.0
     hooks:
-      - id: black
+      - id: ruff
+        args: [--fix]
+      - id: ruff-format

Dockerfile

@@ -1,5 +1,7 @@
-FROM tiangolo/uvicorn-gunicorn:python3.7-alpine3.8
+FROM python:3.12-slim
 LABEL authors="Nathan Sheffield, Michal Stolarczyk"
 
 COPY . /app
+WORKDIR /app
 RUN pip install .
+CMD ["uvicorn", "refgenieserver.main:app", "--host", "0.0.0.0", "--port", "80"]

changelog.md

@@ -2,6 +2,22 @@
 
 This project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html) and [Keep a Changelog](https://keepachangelog.com/en/1.0.0/) format.
 
+## [0.8.0] -- 2026-02-25
+
+### Changed
+- Updated yacman and refgenconf dependency requirements
+- Modernized packaging to use pyproject.toml with hatchling
+- Updated FastAPI route definitions to use `pattern` instead of deprecated `regex`
+- Updated GitHub Actions to modern versions
+
+### Added
+- `app_factory.create_app()` function for programmatic server creation
+
+### Fixed
+- Compatibility with yacman v1 (`with rgc as r:` context manager removed)
+- Compatibility with refgenconf 0.13.0
+- Various modernization and small bugfix improvements
+
 ## [0.7.0] -- 2021-04-27
 ### Added
 - `remotes` section in the refgenieserver config, which supersedes `remote_url_base`. It can be used to define multiple remote data providers.

pyproject.toml

@@ -0,0 +1,67 @@
+[project]
+name = "refgenieserver"
+version = "0.8.0"
+description = "A web interface and RESTful API for reference genome assets"
+readme = "README.md"
+license = {text = "BSD-2-Clause"}
+requires-python = ">=3.10"
+authors = [
+    { name = "Michal Stolarczyk" },
+    { name = "Vince Reuter" },
+    { name = "Nathan Sheffield" },
+]
+keywords = ["bioinformatics", "sequencing", "ngs", "genomes", "server"]
+classifiers = [
+    "Development Status :: 4 - Beta",
+    "Programming Language :: Python :: 3.10",
+    "Programming Language :: Python :: 3.11",
+    "Programming Language :: Python :: 3.12",
+    "Programming Language :: Python :: 3.13",
+    "Topic :: Scientific/Engineering :: Bio-Informatics",
+]
+dependencies = [
+    "aiofiles",
+    "fastapi",
+    "jinja2",
+    "logmuse>=0.2",
+    "refgenconf>=0.13.0",
+    "ubiquerg>=0.6.1",
+    "uvicorn>=0.7.1",
+    "yacman>=0.9.5",
+]
+
+[project.urls]
+Homepage = "https://refgenie.databio.org"
+
+[project.scripts]
+refgenieserver = "refgenieserver.__main__:main"
+
+[build-system]
+requires = ["setuptools>=61"]
+build-backend = "setuptools.build_meta"
+
+[tool.setuptools.packages.find]
+include = ["refgenieserver*"]
+
+[tool.setuptools.package-data]
+refgenieserver = ["templates/**", "static/*"]
+
+[project.optional-dependencies]
+test = [
+    "pytest",
+    "httpx",
+]
+
+[tool.pytest.ini_options]
+addopts = "-rfE"
+testpaths = ["tests"]
+
+[tool.ruff]
+line-length = 88
+
+[tool.ruff.lint]
+select = ["E", "F", "I"]
+ignore = ["F403", "F405", "E501"]
+
+[tool.ruff.lint.isort]
+known-first-party = ["refgenieserver"]

refgenieserver/__init__.py

@@ -1,3 +1,5 @@
+from __future__ import annotations
+
 from .const import *
 from .helpers import *
 from .main import *

refgenieserver/_version.py

@@ -1 +1,6 @@
-__version__ = "0.7.0"
+from __future__ import annotations
+
+# Version is defined in pyproject.toml
+from importlib.metadata import version
+
+__version__ = version("refgenieserver")

refgenieserver/app_factory.py

@@ -0,0 +1,81 @@
+"""Factory function to create a configured refgenieserver FastAPI app."""
+
+from __future__ import annotations
+
+import logging
+import sys
+
+from fastapi import FastAPI
+from refgenconf import RefGenConf
+
+from .const import PKG_NAME, PRIVATE_API, TAGS_METADATA
+from .helpers import purge_nonservable
+
+_LOGGER = logging.getLogger(PKG_NAME)
+
+
+def create_app(config_path: str, archive_base_dir: str | None = None) -> FastAPI:
+    """Create a configured FastAPI app for refgenieserver.
+
+    This builds a fresh FastAPI app with the real refgenieserver routers,
+    configured from a given YAML config file. Used both for production
+    (as an alternative to the CLI entry point) and for integration tests.
+
+    Args:
+        config_path: Path to the refgenie server config YAML.
+        archive_base_dir: Override for BASE_DIR (default: /genomes).
+            Used in tests to point at a temp directory.
+
+    Returns:
+        Configured FastAPI app ready to serve.
+    """
+    # Use sys.modules to get the actual module objects. Using
+    # `import refgenieserver.main as m` can return a different object
+    # than what's in sys.modules (due to __init__.py's `from .main import *`),
+    # which means attribute modifications won't be visible to other modules.
+    import refgenieserver.const  # noqa: F401 ensure loaded
+    import refgenieserver.helpers  # noqa: F401 ensure loaded
+    import refgenieserver.main  # noqa: F401 ensure loaded
+
+    main_module = sys.modules["refgenieserver.main"]
+    const_module = sys.modules["refgenieserver.const"]
+    helpers_module = sys.modules["refgenieserver.helpers"]
+
+    # Load config and purge non-servable entries
+    rgc = RefGenConf.from_yaml_file(config_path)
+    purge_nonservable(rgc)
+
+    # Override the module-level globals that the routers import.
+    # The routers do `from ..main import _LOGGER, rgc, app, templates`
+    # which reads from main's module dict at import time.
+    main_module.rgc = rgc
+    main_module._LOGGER = _LOGGER
+
+    if archive_base_dir is not None:
+        # Must override BASE_DIR in both const and helpers modules,
+        # because helpers.py uses `from .const import *` which copies
+        # BASE_DIR into helpers' own namespace.
+        const_module.BASE_DIR = archive_base_dir
+        helpers_module.BASE_DIR = archive_base_dir
+
+    from ._version import __version__ as server_v
+
+    app = FastAPI(
+        title=PKG_NAME,
+        description="a web interface and RESTful API for reference genome assets",
+        version=server_v,
+        openapi_tags=TAGS_METADATA,
+    )
+
+    # Set the app on main_module so routers that import `app` from main
+    # can access it (needed for openapi spec introspection)
+    main_module.app = app
+
+    # Import routers AFTER rgc is set (they read rgc at import time)
+    from .routers import private, version3
+
+    app.include_router(version3.router)
+    app.include_router(version3.router, prefix="/v3")
+    app.include_router(private.router, prefix=f"/{PRIVATE_API}")
+
+    return app

refgenieserver/const.py

@@ -1,4 +1,7 @@
-""" Package constants """
+"""Package constants"""
+
+from __future__ import annotations
+
 import os
 from platform import python_version
 
@@ -7,39 +10,41 @@
 
 from ._version import __version__ as server_v
 
-ALL_VERSIONS = {
+ALL_VERSIONS: dict[str, str] = {
     "server_version": server_v,
     "rgc_version": rgc_v,
     "python_version": python_version(),
 }
-PKG_NAME = "refgenieserver"
-DEFAULT_PORT = 80
-BASE_DIR = "/genomes"
+PKG_NAME: str = "refgenieserver"
+DEFAULT_PORT: int = 80
+BASE_DIR: str = "/genomes"
 # if running outside of the Docker container 'BASE_DIR' can be replaced with rgc[CFG_ARCHIVE_KEY]
-TEMPLATES_DIRNAME = "templates"
-TEMPLATES_PATH = os.path.join(
+TEMPLATES_DIRNAME: str = "templates"
+TEMPLATES_PATH: str = os.path.join(
     os.path.dirname(os.path.abspath(__file__)), TEMPLATES_DIRNAME
 )
-STATIC_DIRNAME = "static"
-STATIC_PATH = os.path.join(os.path.dirname(os.path.abspath(__file__)), STATIC_DIRNAME)
-LOG_FORMAT = "%(levelname)s in %(funcName)s: %(message)s"
-MSG_404 = "No such {} on server"
-DESC_PLACEHOLDER = "No description"
-CHECKSUM_PLACEHOLDER = "No digest"
+STATIC_DIRNAME: str = "static"
+STATIC_PATH: str = os.path.join(
+    os.path.dirname(os.path.abspath(__file__)), STATIC_DIRNAME
+)
+LOG_FORMAT: str = "%(levelname)s in %(funcName)s: %(message)s"
+MSG_404: str = "No such {} on server"
+DESC_PLACEHOLDER: str = "No description"
+CHECKSUM_PLACEHOLDER: str = "No digest"
 # Here we define the key name changes; format: {"new_key": "old_key"}
 # This dict is then used to pre-process the attributes dict before serving to the old versions of the client
-CHANGED_KEYS = {CFG_ASSET_PATH_KEY: "path"}
+CHANGED_KEYS: dict[str, str] = {CFG_ASSET_PATH_KEY: "path"}
 
 # TODO: to be removed in the future
-CFG_LEGACY_ARCHIVE_CHECKSUM_KEY = "legacy_archive_digest"
+CFG_LEGACY_ARCHIVE_CHECKSUM_KEY: str = "legacy_archive_digest"
 
-API1_ID = "APIv1"
-API2_ID = "APIv2"
-API3_ID = "APIv3"
-PRIV_API_ID = "PRIVATE_API"
+API1_ID: str = "APIv1"
+API2_ID: str = "APIv2"
+API3_ID: str = "APIv3"
+PRIV_API_ID: str = "PRIVATE_API"
 
 
-TAGS_METADATA = [
+TAGS_METADATA: list[dict[str, str]] = [
     {
         "name": API3_ID,
         "description": "These are the most recent API endpoints. "

refgenieserver/data_models.py

@@ -1,12 +1,12 @@
+from __future__ import annotations
+
 from typing import Dict, List
 
 from pydantic import BaseModel
 
 
 class Tag(BaseModel):
-    """
-    Tag data model
-    """
+    """Tag data model."""
 
     asset_path: str
     asset_digest: str
@@ -19,19 +19,15 @@ class Tag(BaseModel):
 
 
 class Asset(BaseModel):
-    """
-    Asset data model
-    """
+    """Asset data model."""
 
     asset_description: str
     tags: Dict[str, Tag]
     default_tag: str
 
 
 class Genome(BaseModel):
-    """
-    Genome data model
-    """
+    """Genome data model."""
 
     genome_description: str
     assets: Dict[str, Asset]