Skip to main content

File-backed pydantic configuration with migration support

Project description

fluxconf

Tests PyPI version Supported versions license

File-backed Pydantic configuration with migration support.

Installation

pip install fluxconf

or with uv:

uv add fluxconf

Quick start

ConfigIO is a generic base class for reading and writing YAML-backed Pydantic models. Subclass it, set file_name and config_type, and you get type-safe read/write out of the box.

from pydantic import BaseModel
from fluxconf import ConfigIO

class AppConfig(BaseModel):
    name: str = "my-app"
    debug: bool = False

class AppConfigIO(ConfigIO[AppConfig]):
    file_name = "app.yml"
    config_type = AppConfig

io = AppConfigIO("~/.config/my-app")
io.write(AppConfig(name="my-app", debug=True))
config = io.read()  # AppConfig(name='my-app', debug=True)

Migrations

Migrations transform stored configuration data when your schema evolves. Instead of breaking existing config files, you define migration steps that update old data to match the new schema.

To use migrations, inherit from VersionedBaseModel instead of Pydantic's BaseModel. This adds a version field that tracks which migrations have been applied.

Migration keys follow the "N_description" format — the integer prefix determines execution order and is stored as the version in the config file. On read(), any pending migrations run automatically and the file is updated on disk.

JSON Patch migrations

The simplest approach: declare JSON Patch (RFC 6902) operations directly. No Python functions needed.

from fluxconf import ConfigIO, VersionedBaseModel

class ServerConfig(VersionedBaseModel):
    host: str = "localhost"
    port: int = 8080

class ServerConfigIO(ConfigIO[ServerConfig]):
    file_name = "server.yml"
    config_type = ServerConfig
    migrations = {
        "1_rename_host": [
            {"op": "move", "from": "/hostname", "path": "/host"},
        ],
        "2_add_port": [
            {"op": "add", "path": "/port", "value": 8080},
        ],
    }

Supported operations: add, remove, replace, move, copy, and test.

Python function migrations

When you need conditional logic or complex transforms, use a Python function. Each function receives the raw config dict and must return the updated dict.

from fluxconf import ConfigIO, VersionedBaseModel

class UserConfig(VersionedBaseModel):
    full_name: str = ""
    email: str = ""

def merge_name_fields(data: dict) -> dict:
    first = data.pop("first_name", "")
    last = data.pop("last_name", "")
    if first or last:
        data["full_name"] = f"{first} {last}".strip()
    return data

class UserConfigIO(ConfigIO[UserConfig]):
    file_name = "user.yml"
    config_type = UserConfig
    migrations = {
        "1_merge_name": merge_name_fields,
    }

Python functions and JSON Patches can be mixed freely in the same migrations dict.

Directory-based migrations

For projects with many migrations, store each one as a separate file in a directory instead of inlining them all in the class definition.

myapp/migrations/
    1_rename_host.json
    2_merge_name.py
    3_add_defaults.py
    _helpers.py           # skipped (starts with _)

Only files with an integer prefix are loaded. Files starting with _ or without an integer prefix are silently skipped, so helper modules can live alongside migration files.

.json files contain a JSON array of patch operations:

myapp/migrations/1_rename_host.json

[
    {"op": "move", "from": "/hostname", "path": "/host"}
]

.py files with a patch attribute are equivalent to .json files but written in Python:

myapp/migrations/3_add_defaults.py

patch = [
    {"op": "add", "path": "/port", "value": 8080},
    {"op": "add", "path": "/retries", "value": 3},
]

.py files with a migrate function offer full flexibility:

myapp/migrations/2_merge_name.py

def migrate(data: dict) -> dict:
    first = data.pop("first_name", "")
    last = data.pop("last_name", "")
    if first or last:
        data["full_name"] = f"{first} {last}".strip()
    return data

If a .py file defines both migrate and patch, the migrate function takes precedence.

Point migrations_dir at the directory to load them:

from pathlib import Path
from fluxconf import ConfigIO, VersionedBaseModel

class AppConfigIO(ConfigIO[AppConfig]):
    file_name = "app.yml"
    config_type = AppConfig
    migrations_dir = Path(__file__).parent / "migrations"

migrations and migrations_dir can be used together — fluxconf merges them, raising ValueError on key collisions.

Error handling

MigrationError is raised when a migration function or patch fails. It carries two attributes:

  • last_successful_migration — the version of the last migration that completed successfully (or the stored version if none succeeded)
  • original_error — the underlying exception

ValueError is raised when the stored version is ahead of the latest known migration. This typically means the config file was written by a newer version of the software than the one currently running.

License

MIT License

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

fluxconf-0.0.2.tar.gz (56.6 kB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

fluxconf-0.0.2-py3-none-any.whl (11.8 kB view details)

Uploaded Python 3

File details

Details for the file fluxconf-0.0.2.tar.gz.

File metadata

  • Download URL: fluxconf-0.0.2.tar.gz
  • Upload date:
  • Size: 56.6 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.2.0 CPython/3.12.12

File hashes

Hashes for fluxconf-0.0.2.tar.gz
Algorithm Hash digest
SHA256 4a2fd40708a7c7a115957494acb37dba67216eb21ae8bce06ca198c76508c594
MD5 03c676ab6529f152d39bc18faca985b1
BLAKE2b-256 2482ad78f86c6eb6df9e8b2bc53aa1499fb30112f8111768325ddbeb601ed959

See more details on using hashes here.

File details

Details for the file fluxconf-0.0.2-py3-none-any.whl.

File metadata

  • Download URL: fluxconf-0.0.2-py3-none-any.whl
  • Upload date:
  • Size: 11.8 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.2.0 CPython/3.12.12

File hashes

Hashes for fluxconf-0.0.2-py3-none-any.whl
Algorithm Hash digest
SHA256 889a78652be1abbb652ff430e6ba4eae6d779811de0aa2dfdb33348fb5af1803
MD5 15d3ce930a3cc73d3da4095be06732c7
BLAKE2b-256 ca8c6cf32e33ce3216637d19d54ca9e439b8c037a0507e4d91bbb1b1cfa1c550

See more details on using hashes here.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page