Skip to main content

Runch: refined munch. Attribute-style access for python dicts, with generic typing support. Help handle your configs, data, and more.

Project description

Runch

Refined munch. Provides basic munch functionality (attribute-style access for python dicts) with additional generic typing support and runtime validation.

Also provides a config reader that reads config files into predefined runch models. Say goodbye to config["key"], config.get("key") and runtime errors caused by missing keys!

Installation

pip install runch

If you find any bugs, please submit an issue or a pull request at GitHub.

Usage

Example Config Model Definition:

$ python3 -m runch ./etc/base.yaml
# Generated from base{.example,}.yaml by runch
# Please be aware that `float` fields might be annotated as `int` due to the lack of type info in the config.

from __future__ import annotations

from typing import List

from pydantic import Field
from runch import RunchModel, RunchConfigReader

class BaseConfigModel(RunchModel):
    host: str
    port: str
    user: str
    password: str


base_reader = RunchConfigReader[BaseConfigModel]("base.yaml", config_dir="./etc", config_type="yaml")
base = base_reader.read()

# example:
print(base.config.host, base.config.port)  # with awesome intellicode support & runtime validation!

# uncomment the following line to enable the watch_update feature
# _base_reader.enable_feature("watch_update", {"update_interval": 10})

Model Definition Generator

$ python -m runch <config_path> [config_ext]

Usage:

Usage: python -m runch <config_path> [config_name [config_ext]]
    Generate a model definition from a config file.

    config_path: path to your config file.
    config_name: controls generated variable name and class name.
    config_type: content type of your config file. Default is `yaml`.

    Example:
        python -m runch path/to/my_config.foo
        python -m runch path/to/my_config.foo chat_config
        python -m runch path/to/my_config.foo chat_config yaml

RunchConfigReader Arguments

# Read config from file.                 ↓ square brackets
example_config_reader = RunchConfigReader[ExampleConfig](
                                        # ^^^^^^^^^^^^^ Config model class name
    config_name="config_file_name",     # with file extension, but don't include the ".example" part
    config_dir="config_dir",            # default is os.environ.get("RUNCH_CONFIG_DIR", "./etc")
    config_type="yaml",                 # default is "yaml"
    config_encoding="utf-8",            # default is "utf-8"
    config_logger=getLogger("example"), # default is None
)
example_config = example_config_reader.read()  # Or .read_lazy() for lazy loading

print(example_config.config.db_host)
$ touch example_config_dir/example_config_file.yaml
db_host: localhost
db_port: 5432
db_user: user
db_password: password
db_name: database

Supported File Formats

  • YAML
  • JSON
  • TOML
  • arbitrary file formats with custom parser, specified via the custom_config_parser param of RunchConfigReader.__init__(). The custom parser should be a function that takes a str-type file content as its first argument, and returns a parsed dictionary.

Use Remote Config Source

from httpx import AsyncClient
from runch import (
    RunchModel,
    RunchAsyncCustomConfigReader,
)


class TestConfig(RunchModel):
    status: str
    method: str


async def example_config_loader(config_name: str) -> TestConfig:
    """Load config from a remote source."""

    print(f"Loading config from remote source for {config_name=}...")

    # Simulate a network request to fetch the config.
    async with AsyncClient() as client:
        response = await client.get(
            f"https://dummyjson.com/test",
            headers=headers,
        )
        response.raise_for_status()

    return TestConfig(**response.json())


test_reader = RunchAsyncCustomConfigReader[TestConfig](
    config_name="example",
    config_loader=example_config_loader,
)

async def main():
    test_config = await test_reader.read()
    print(test_config.config.status)

    # read_cached() does not require `await`, but it raises RunchLookupError if there's no prior sucessful call to `read()`
    print(test_config, test_reader.read_cached())

Note that you can raise runch.exceptions.RunchConfigUnchanged inside custom config loaders to prevent unnecessary config reloads / updates.

Other Features

  • configurable auto sync & update.
  • optional lazy load & evaluate. Useful for optional configs that may not exist.
  • configurable example merging for fast local development.
  • read arbitrary file formats from any places (e.g. disk, network, db) with sync reader + custom config parser / async reader + custom config loader.

Auto Update

test_reader.enable_feature("watch_update", {"update_interval": 2})  # update every 2s
test_reader.enable_feature(
    "watch_update",
    {
        "update_interval": 2.5,  # update every 2.5s
        "on_update_error": "ignore"  # ignore or raise. default=ignore
    }
)

Merge Example Yaml for Local Dev

test_reader = RunchConfigReader[TestConfig](
    config_name="test.yaml",
    config_type="yaml",
)
if os.environ.get("SERVE_MODE") != "prod":
    test_reader.enable_feature("merge_example", {})

User-Defined Custom Config Validation Logic

Note that this is different from the custom config loader feature.

from runch import RunchModel, RunchConfigReader

class CustomConfigModel(RunchModel):
    host: str
    port: int

    def __init__(self, *args: Any, **kw: Any):
        super().__init__(*args, **kw)

        if self.port % 2 == 0:
            raise ValueError("Only odd port numbers are accepted")

Logging

from logging import getLogger
from typing import Any
from runch import (
    RunchModel,
    RunchConfigReader,
    RunchLogLevel,
)

class RunchLogAdapter:
    def log(self, level: RunchLogLevel, msg: str, /, *, exc_info: BaseException | None = None, **kwargs: Any):
        getLogger("runch").log(level, msg, exc_info=exc_info, **kwargs)

class TestConfig(RunchModel):
    x: int

test_reader = RunchConfigReader[TestConfig](
    config_name="test.yaml", config_dir="runch/test", logger=RunchLogAdapter()
)
test_reader.enable_feature("watch_update", {"update_interval": 1})
test_config = test_reader.read()  # will call RunchLogAdapter().log with level RunchLogLevel.ERROR if config is invalid.

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

runch-1.9.3.tar.gz (23.2 kB view details)

Uploaded Source

Built Distribution

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

runch-1.9.3-py3-none-any.whl (27.6 kB view details)

Uploaded Python 3

File details

Details for the file runch-1.9.3.tar.gz.

File metadata

  • Download URL: runch-1.9.3.tar.gz
  • Upload date:
  • Size: 23.2 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: poetry/2.1.3 CPython/3.12.6 Darwin/22.6.0

File hashes

Hashes for runch-1.9.3.tar.gz
Algorithm Hash digest
SHA256 98ae499ccaf3cc3cddcd5c29d9e6de9a1fb87516af334febc1b8b9988c3bffb4
MD5 b62555d9303484a03896fc20ce4d0a12
BLAKE2b-256 9bffc87baa306fc036f2f6ad58efdfd5da2c948b491c6e8a87f415a118f5f6ea

See more details on using hashes here.

File details

Details for the file runch-1.9.3-py3-none-any.whl.

File metadata

  • Download URL: runch-1.9.3-py3-none-any.whl
  • Upload date:
  • Size: 27.6 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? No
  • Uploaded via: poetry/2.1.3 CPython/3.12.6 Darwin/22.6.0

File hashes

Hashes for runch-1.9.3-py3-none-any.whl
Algorithm Hash digest
SHA256 89218d89cf309bde4c02db2fc47acff37185740f5cf5e3758df982fcddeaa531
MD5 a181a9e73801a0e66051de4326a7764f
BLAKE2b-256 2acb1aff2bbacf0c06e94b67719a5ffd46bb2d9af2fb268fcc344d07189ee9f8

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