Metadata-Version: 2.4
Name: alchemy-config
Version: 1.2.0
Summary: Configuration framework in Python for general configuration use.
Project-URL: Homepage, https://github.com/IBM/alchemy-config
Author-email: Gabe Goodhart <ghart@us.ibm.com>
License-Expression: MIT
License-File: LICENSE
Keywords: config
Requires-Python: >=3.9
Requires-Dist: pyyaml>=5.3.1
Provides-Extra: test
Requires-Dist: pytest-cov>=2.9.0; extra == 'test'
Requires-Dist: pytest>=5.3.2; extra == 'test'
Description-Content-Type: text/markdown

# Alchemy Config
The `aconfig` library provides simple `yaml` configuration in `python` with environment-based overrides.

## Library advantages
1. Easily reads environment variables with error handling, e.g. for absence of variables or boolean capitalization
1. Allows users to easily access and traverse nested configuration without checking for key presence on every access
1. Automatically handles overriding logic - environment variables take precedence over local configuration files

## Installation

To install, simply use `pip`

```sh
pip install alchemy-config
```

## Quick Start

**config.yaml**
```yaml
foo: 1
bar:
    baz: "bat"
```

**main.py**
```py
import aconfig

if __name__ == "__main__":
    config = aconfig.Config.from_yaml("config.yaml")
    print(config.foo)
    print(config.bar.baz)
```

```sh
export BAR_BAZ="buz"
python3 main.py
```

## Corner-case Behavior

You CAN set builtin method names as attributes on the `config`. However, you should only access/delete them via dictionary access methods.

For example:

```py
import aconfig
cfg = {"update": True}

config = aconfig.Config(cfg)

# DO NOT DO THIS:
config.update

# DO THIS INSTEAD:
config["update"]
```

This is because there is no way in Python to tell whether you want the method or the attribute `"update"` when "getting" it from the object.
