Skip to main content

Parse YAML, JSON, and TOML into Python objects with smart environment variable interpolation.

Project description

Overview

Confspec is a configuration loading library for Python that supports smart environment variable interpolation at runtime.

It provides a simple way to load configuration from common formats like YAML, JSON, and TOML, while being easy to add support for any additional formats. Out-of-the-box, confspec also supports parsing application configuration into Pydantic models, or Msgspec Structs.

The environment variable interpolation syntax allows you to keep your defaults in the configuration file, while still being able to override them from a deployment environment.

Confspec also supports environment-based configuration overrides, allowing you to maintain separate configuration files for different environments (akin to spring's application.properties) without duplicating your entire configuration.

Installation

Confspec is available on PyPI and can be installed using your package manager of choice.

pip install confspec
uv add confspec

Usage

config.toml

[server]
port = "${PORT:8080}"
debug = "${DEBUG:false}"

[database]
url = "postgres://${DB_URL:postgres:postgres@localhost:5432}/postgres"

[logging]
level = "${LOG_LEVEL~:INFO}"
handlers = "${LOG_HANDLERS[,]:console,file}"

[features]
enabled = "${FEATURE_FLAGS[,]:auth,metrics,caching}"

The above configuration file can easily be loaded, with any environment substitution happening automatically.

>>> import confspec
>>> confspec.load("config.toml")
{
    "server": {
        "port": "8080",
        "debug": "false"
    },
    "database": {
        "url": "postgres://postgres:postgres@localhost:5432/postgres"
    },
    "logging": {
        "level": "INFO",
        "handlers": ["console", "file"]
    },
    "features": {
        "enabled": ["auth", "metrics", "caching"]
    }
}

Or if you wanted it to be loaded into a msgspec Struct object. Note that msgspec will coerce some fields into the requested types automatically, unless you pass strict=True.

>>> import confspec
>>> import msgspec

>>> class Server(msgspec.Struct):
...    port: int
...    debug: bool

>>> class Database(msgspec.Struct):
...    url: str

>>> class Logging(msgspec.Struct):
...    level: str
...    handlers: list[str]

>>> class Features(msgspec.Struct):
...    enabled: list[str]

>>> class Config(msgspec.Struct):
...    server: Server
...    database: Database
...    logging: Logging
...    features: Features

>>> confspec.load("config.toml", cls=Config)
Config(
    server=Server(port=8080, debug=False),
    database=Database(url='postgres://postgres:postgres@localhost:5432/postgres'),
    logging=Logging(level='INFO', handlers=['console', 'file']),
    features=Features(enabled=['auth', 'metrics', 'caching'])
)

Interpolation Syntax

  • ${VAR}
    • replaced with the value of the VAR environment variable
    • if VAR is unset during interpolation, a KeyError will be raised
  • ${VAR:default}
    • replaced with the value of the VAR environment variable
    • if VAR is unset during interpolation, the default value will be used instead
  • ${VAR[,]}
    • performs list expansion on the value of the VAR environment variable, splitting on the delimiter (in this case ,)
    • the delimiter can be any combination of one or more characters, excluding ] and }
    • this operator cannot be applied to patterns within a longer string
  • ${VAR~}
    • strips whitespace from the value of the VAR environment variable (or the default if one is specified)
    • if combined with list expansion, each individual element will have its whitespace stripped
  • ${VAR?}
    • replaced with the value of the VAR environment variable
    • if VAR is unset during interpolation, it will be replaced with Python's None instead
    • this operator cannot be applied to patterns within a longer string

Most of these interpolation rules can be combined, except for a default value and the "None as default" flag.

For example:

  • ${VAR[,]~:default} - valid
  • ${VAR[,]~?} - valid
  • ${VAR[,]~?:default} - invalid

[!NOTE] The order that the flags are written is important, as the interpolation syntax is parsed using regex. You should always specify the flags in the same order as the valid expressions shown above.

Environment-specific Configurations

When an environment is specified, Confspec automatically looks for an additional file matching the base configuration name with the environment suffix and merges it with the base configuration.

For example:

config.yaml
config.prod.yaml
config.dev.yaml

You can activate an additional environment in two ways:

import confspec

# Explicitly pass the environment
config = confspec.load("config.yaml", env="prod")

# Or use the CONFSPEC_ENV environment variable
# export CONFSPEC_ENV=prod
config = confspec.load("config.yaml")

In both cases, config.prod.yaml will be loaded and merged with config.yaml, allowing you to override specific values for that environment while keeping shared defaults in one place. Any values specified in the environment-specific config file will override those specified within the root configuration file.

Environment variable interpolations are performed after the configurations have been merged.

Issues

If you find any bugs, issues, or unexpected behaviour while using the library, you should open an issue with details of the problem and how to reproduce if possible. Please also open an issue for any new features you would like to see added.

Contributing

Pull requests are welcome. For major changes, please open an issue/discussion first to discuss what you would like to change.

Please try to ensure that documentation is updated if you add any features accessible through the public API.

If you use this library and like it, feel free to sign up to GitHub and star the project, it is greatly appreciated and lets me know that I'm going in the right direction!

Links

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

confspec-0.0.4.tar.gz (65.0 kB view details)

Uploaded Source

Built Distribution

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

confspec-0.0.4-py3-none-any.whl (16.9 kB view details)

Uploaded Python 3

File details

Details for the file confspec-0.0.4.tar.gz.

File metadata

  • Download URL: confspec-0.0.4.tar.gz
  • Upload date:
  • Size: 65.0 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: python-requests/2.32.3

File hashes

Hashes for confspec-0.0.4.tar.gz
Algorithm Hash digest
SHA256 f004813c5f413eb16ea0d9f38f2c5fa2058902bcb0be6f5d87e4bb94041fba6e
MD5 70e881fb6f0a02acb03388e420f2a74a
BLAKE2b-256 4e95296f5e39c967ee6c054211d55608e70362461b04dceda148c8cacb4cac8b

See more details on using hashes here.

File details

Details for the file confspec-0.0.4-py3-none-any.whl.

File metadata

  • Download URL: confspec-0.0.4-py3-none-any.whl
  • Upload date:
  • Size: 16.9 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? No
  • Uploaded via: python-requests/2.32.3

File hashes

Hashes for confspec-0.0.4-py3-none-any.whl
Algorithm Hash digest
SHA256 a4d97ccf604c783420f6633e3a284efcedce2295a85e261c5a259b638756893f
MD5 f7ed9aec5badf2083af438c6269c1fba
BLAKE2b-256 430ce1f77417a3cf98c5b475442e04bc729d1e6fbc92d6e5c0dad18828d9c474

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