labthings_fastapi.server.config_model
Pydantic models to enable server configuration to be loaded from file.
The models in this module allow ThingConfig dataclasses to be constructed
from dictionaries or JSON files. They also describe the full server configuration
with ServerConfigModel. These models are used by the cli module to
start servers based on configuration files or strings.
Attributes
Exceptions
Failed to import Thing. Raise with import traceback. |
Functions
|
Prevent errors during import from causing odd validation errors. |
Ensure every Thing is defined by a |
Module Contents
- exception labthings_fastapi.server.config_model.ThingImportFailure
Bases:
BaseExceptionFailed to import Thing. Raise with import traceback.
Initialize self. See help(type(self)) for accurate signature.
- labthings_fastapi.server.config_model.contain_import_errors(value: Any, handler: pydantic.ValidatorFunctionWrapHandler) Any
Prevent errors during import from causing odd validation errors.
This is used to wrap the pydantic ImportString validator, and ensures that any module that won’t import shows up with a single clear error.
- Parameters:
value – The value being validated.
handler – The validator handler.
- Returns:
The validated value.
- Raises:
ThingImportFailure – if an import error occurs, with the stack trace from retrying the import.
Exception – In the unlikely event that the import error cannot be reproduced
- labthings_fastapi.server.config_model.ThingImportString
- labthings_fastapi.server.config_model.ThingName
- type labthings_fastapi.server.config_model.ThingsConfig = Mapping[ThingName, ThingConfig | ThingImportString]
- labthings_fastapi.server.config_model.normalise_things_config(things: ThingsConfig) collections.abc.Mapping[ThingName, ThingConfig]
Ensure every Thing is defined by a
ThingConfigobject.Things may be specified either using a
ThingConfigobject, or just a bareThingsubclass, if the other parameters are not needed. To simplify code that uses the configuration, this function wraps bare classes in aThingConfigso the values are uniformly typed.- Parameters:
things – A mapping of names to Things, either classes or
ThingConfigobjects.- Returns:
A mapping of names to
ThingConfigobjects.- Raises:
ValueError – if a Python object is passed that’s neither a
typenor adict.