Source code for bag.settings

"""Facilitate materialization of resources indicated in configuration."""

from importlib import import_module
from types import ModuleType

[docs]def read_ini_files(*config_files, encoding='utf-8'): """Get a settings object (dict-like) by reading some ``config_files``.""" from configparser import ConfigParser settings = ConfigParser(), encoding=encoding) return settings
[docs]def resolve(resource_spec): """Return the variable referred to in the ``resource_spec`` string. Example resource_spec: ``"my.python.module:some_callable"``. """ if isinstance(resource_spec, ModuleType # arg is a python module ) or callable(resource_spec): # arg is a callable return resource_spec parts = resource_spec.split(':') # arg is assumed to be a string if len(parts) == 1: return import_module(parts[0]) elif len(parts) == 2: module = import_module(parts[0]) return getattr(module, parts[1]) else: raise ValueError( '":" may appear only once in a resource spec, but I received "{}"' .format(resource_spec))
[docs]def resolve_path(resource_spec): """Return a pathlib.Path corresponding to the ``resource_spec`` string. Example argument: ``"my.python.module:some/subdirectory"`` Similar: ``from pyramid.resource import abspath_from_asset_spec`` """ from pathlib import Path module, var = resource_spec.split(':') # arg is assumed to be a string module = import_module(module) return Path(module.__path__[0], var)
[docs]def asbool(s): """Convert the argument to a boolean. Return the boolean value ``True`` if the case-lowered value of string input ``s`` is a truthy string. If ``s`` is already one of the boolean values ``True`` or ``False``, return it. """ if s is None: return False if isinstance(s, bool): return s val = _boolean_states.get(str(s).strip().lower()) if val is None: raise ValueError('Not a boolean: "%s"' % s) return val
_boolean_states = {'1': True, 'yes': True, 'true': True, 'on': True, '0': False, 'no': False, 'false': False, 'off': False}
[docs]class SettingsReader(object): """Convenient for reading configuration settings in an app. However, with Python 3.6+ I have used the **pydantic** library with better results. """ def __init__(self, adict): """``adict`` should be a settings dictionary.""" self.settings = adict
[docs] def read(self, key, default=None, required=False): """Return setting value, or ``default`` if missing. Raise RuntimeError if ``required`` is true and value is empty. """ value = self.settings.get(key, default) if required and value in (None, ''): raise RuntimeError( 'Settings are missing a "{}" entry.'.format(key)) return value
[docs] def bool(self, key, default=None, required=False): """Return a boolean setting value.""" value =, default=default, required=required) return asbool(value)
[docs] def resolve(self, key, default=None, required=False): """Return the variable or module indicated in the setting value. Therefore the setting value should be a resource specification such as ``some.module:SomeClass``. """ resource_spec = key, default=default, required=required) return None if resource_spec is None else resolve(resource_spec)
[docs] def resolve_path(self, key, default=None, required=False): """Return a pathlib.Path corresponding to the setting value. Example argument: ``"my.python.module:some/subdirectory"`` """ resource_spec = key, default=default, required=required) return None if resource_spec is None else resolve_path(resource_spec)