Antares Study Version

Table of Contents

Overview
Installation
Usage
Development
- Development tasks
- Building the package
License
Changelog

Overview

The antares-study-version package defines StudyVersion and SolverVersion classes to manage version numbers. It can be used to manage the version of a study, but also the version of Antares Solver. It supports the semver format ("major.minor.patch") and the integer format (major×100 + minor×10 + patch), which is specific to Antares.

This module harmonizes the management of versions in Antares:

at the level of Antares studies (configuration files, database, etc.)
at the level of Antares applications, in particular AntaREST.

In the data of a study and in the programs, we encounter several version formats:

dotted string (ex. "8.7" or "8.7.2"),
compact string (ex. "870"),
integer (ex. 870).
tuples or lists (ex. (8, 7) or [8, 7, 2]).
dictionaries (ex. {"major": 8, "minor": 7, "patch": 2}).

For instance, since version 9.0 of Antares Solver, versions are stored as dotted strings; the compact format is now obsolete (backward compatibility is ensured for versions prior to 9.0);

For instance, the study.antares configuration file now uses the "X.Y" format for the study version instead of the "XYZ" format.

[antares]
version = 9.1
caption = My Study
created = 1618413128
lastsave = 1625583204
author = John DOE

This module allows to convert these formats to each other, and to compare versions.

Installation

pip install antares-study-version

Usage

This package provides a StudyVersion class to manage study version numbers, and a SolverVersion class to manage Antares Solver version numbers.

The difference between StudyVersion and SolverVersion is that Solver versions are generally on 3 digits (with the major, minor and patch components), while study versions are on 2 digits (with the major and minor components only), and the patch component is not used (always 0).

StudyVersion

Using the antares-study-version module is straightforward:

from antares.study.version import StudyVersion

version = StudyVersion(8, 7, 2)  # patch component is not used
print(version)  # 8.7

You can also create a version object from a dotted string:

from antares.study.version import StudyVersion

version = StudyVersion.parse("8.7")
print(version)  # 8.7

You can create a version object from a compact string:

from antares.study.version import StudyVersion

version = StudyVersion.parse("870")
print(version)  # 8.7

You can create a version object from an integer:

from antares.study.version import StudyVersion

version = StudyVersion.parse(870)
print(version)  # 8.7

You can compare versions:

from antares.study.version import StudyVersion

version1 = StudyVersion(8, 6)
version2 = StudyVersion(8, 7)
print(version1 < version2)  # True

You can convert a version to string using format specifiers:

from antares.study.version import StudyVersion

version = StudyVersion(8, 7)
print(f"{version}")  # 8.7
print(f"{version:02d}")  # 08.07
print(f"{version:03d}")  # 08.07.00
print(f"{version:ddd}")  # 870

You can convert a version to an integer:

from antares.study.version import StudyVersion

version = StudyVersion(8, 7)
print(int(version))  # 870

SolverVersion

Of course, the same operations can be done with SolverVersion objects, but with 3 digits:

from antares.study.version import SolverVersion

version = SolverVersion(8, 7, 2)
print(version)  # 8.7.2

Objects of the StudyVersion and SolverVersion classes can be compared to each other:

from antares.study.version import StudyVersion, SolverVersion

study_version = StudyVersion(8, 7)
solver_version = SolverVersion(8, 7, 2)
print(study_version <= solver_version)  # True

Pydantic model

You can even use the StudyVersion or SolverVersion classes in a Pydantic model:

import datetime
import typing as t
import uuid

from pydantic import BaseModel, Field, validator

from antares.study.version import StudyVersion


class StudyDTO(BaseModel):
    id: uuid.UUID = Field(default_factory=lambda: uuid.uuid4())
    name: str
    version: StudyVersion
    created: datetime.datetime = Field(default_factory=lambda: datetime.datetime.now(tz=datetime.timezone.utc))

    @validator("version", pre=True)
    def _validate_version(cls, v: t.Any) -> StudyVersion:
        return StudyVersion.parse(v)


study = StudyDTO(name="foo", version=StudyVersion(4, 5))
obj = study.json()
print(obj)
# {
#     "created": "2024-12-31T12:30:00+00:00",
#     "id": "4930a577-63d2-4ea9-b0b9-581110d97475",
#     "name": "foo",
#     "version": {"major": 4, "minor": 5, "patch": 0}
# }

Development

This projet uses Hach to manage the development environment.

Project setup

➢ To install the development environment, run:

hatch env create

See hatch env create documentation

This command will create a virtual environment and install the development dependencies.

NOTE: hatch creates a virtual environment in the ~/.local/share/hatch/env/virtual/antares-study-version directory.

➢ To activate the virtual environment, run:

hatch shell

See hatch shell documentation

This command will spawn a new shell with the virtual environment activated. Use Ctrl+D to exit the shell.

NOTE: this command will display the full path to the virtual environment. You can use it to configure PyCharm or Visual Studio Code to use this virtual environment.

Development tasks

➢ To format and lint the source code with ruff, run:

hatch fmt

See hatch fmt documentation

➢ To run the tests on the current Python version, run:

hatch run test

See hatch run documentation

➢ To run the tests on Python 3.12, for example, run:

hatch run all.py3.12:test

➢ To generate the test coverage report, run:

hatch run cov

This command will run the unit tests and generate a coverage report in the htmlcov directory.

➢ To check the typing with mypy, run:

hatch run types:check

➢ To check the typing on unit tests, run:

hatch run types:check-tests

Building the package

➢ To build the package, run:

hatch build

This command will create a dist directory with the built package.

➢ To build the package and upload it to PyPI, run:

hatch publish

➢ To clean the project, run:

hatch clean

This command will remove the dist directory.

License

antares-study-version is distributed under the terms of the MIT license.

AntaresSimulatorTeam/antares-study-version