toml-init 0.1.5

A library to auto-create and validate TOML configuration files from templates

TOML-Init

A lightweight Python library for auto-creating and validating TOML-based configuration files from one or more default templates.

For use where you use multiple relatively independent modules that need config values and you want a standardized way to handle, implement, and validate.

---

Defaults

Variable	Value
DEFAULT_CONFIG_FOLDER_PATH	<Current Working Directory>/config
DEFAULT_CONFIG_DEFAULT_FOLDER_PATH	<Current Working Directory>/defaults
DEFAULT_CONFIG_FILE	"config.toml"

Features

• Automatic directory setup

  • Ensures your primary configs/ directory and its defaults/ subdirectory exist (creates them if missing).

• Defaults merging

  • Scans all *.toml files under configs/defaults/. Each top-level TOML table is treated as a “block” of settings.
  • Merges multiple default files into a single master config.toml, with conflict detection.

• Schema-driven validation

  • Each setting may be declared in two ways:

    1. Simple: KEY = <defaultValue>
    2. Full: KEY = { defaultValue = <...>, type = "<int|float|bool|str>", min = <...>, max = <...>, allowedValues = [...], validator = "<name>" }

  • Supports numeric range checks (min/max), enumerations (allowedValues), and custom validation hooks.

• Custom validators

  • Implement by subclassing Validator and registering via register_validator("name", instance).
  • Allows arbitrary user‑defined checks (e.g. path existence, regex match).

• Preserves user overrides

  • If users manually edit config.toml, extra keys remain intact.
  • Invalid values are reset to defaults with a logged warning.

• CLI & programmatic API

  • Installable console script: toml-init for quick command-line setup.
  • ConfigManager class for embedding in Python code.

• Dry‑run & logging

  • --dry-run to validate without writing changes.
  • Verbose mode with --verbose for debug logging.

---

Installation

pip install toml-init

Package Structure

toml_init
├── __init__.py
├── __main__.py
├── exceptions.py
├── validators.py
└── manager.py

Quickstart (Python)

from pathlib import Path
from toml_init import ConfigManager, register_validator, Validator

# Optional: register a custom validator
class PathExistsValidator(Validator):
    def validate(self, value):
        from pathlib import Path as P
        if not (isinstance(value, str) and P(value).exists()):
            raise InvalidConfigValueError(f"Path not found: {value}")
        return value

register_validator("path_exists", PathExistsValidator())

# Initialize and validate
cm = ConfigManager(
    base_path=Path("configs"),
    defaults_path=Path("configs/defaults"),
    master_filename="config.toml"
)
cm.initialize(dry_run=False)

# Access validated settings
settings = cm.get_block("QuickBooks.Invoices.Saver")
print(settings["WINDOW_LOAD_DELAY"])  # e.g. 0.5

Quickstart (CLI)

# Create or validate configs in ./configs (defaults in ./configs/defaults)
toml-init

# Verbose & dry-run (no file writes)
toml-init --verbose --dry-run

# Custom paths
 toml-init --base /etc/myapp/configs \
           --defaults /etc/myapp/configs/defaults \
           --master settings.toml

Example Project

A working sample lives under example_project/. Run python run_demo.py inside that directory to generate config.toml using the defaults in configs/defaults/example_defaults.toml.

Default File Format

Place your default templates in configs/defaults/. Each file may define multiple blocks (tables):

# defaults/app_defaults.toml

[__meta__]
name = "Config - QuickBooks Invoice Saver"
module_version = "0.1.0"

[QuickBooks.Invoices.Saver]
SHOW_TOASTS = true
WINDOW_LOAD_DELAY  = { defaultValue = 0.5,  type = "float", min = 0.0 }
NAVIGATION_DELAY   = { defaultValue = 0.15, type = "float", min = 0.0 }

[MyApp.Logging]
LEVEL       = { defaultValue = "INFO", type = "str", allowedValues = ["DEBUG","INFO","WARN","ERROR"] }
ENABLE_LOGS = { defaultValue = true,   type = "bool" }

• Block names = the table names (e.g. QuickBooks.Invoices.Saver).

• Settings under each block may be:

  • A primitive default: KEY = 123 (shorthand for a setting with only defaultValue).
  • A full schema object: specifying type, optional min/max, allowedValues, and validator.

• An optional [__meta__] table may be present and is ignored by the library.

Supported Types

• int, float, bool, str, list, dict, datetime, date, time
• Numeric range checks with min/max
• Enumerations via allowedValues
• Custom hooks via validator (must match a registered Validator)

---

Exception Hierarchy

• TomlInitError (base)

  • MultipleConfigFilesError
  • InvalidDefaultSchemaError
  • InvalidConfigValueError
  • BlockConflictError

Catch these to handle specific error cases gracefully.

---

Potential Future Features

Feature	Description
Multiple config files	Will support having separate config files for organization purposes?
Comment injection	Allow comments to be specified in the defaults?