Normalization

Normalization standardizes argument names before digestion runs.

Use it when a function accepts aliases (name vs atom_name) or when argument meaning depends on another argument (element + name).

Standardizer hook

ArgDigest supports an optional standardizer callable:

from argdigest import arg_digest

def argument_names_standardization(caller, kwargs):
    if caller.endswith(".get") and "name" in kwargs:
        kwargs = dict(kwargs)
        kwargs["element_name"] = kwargs.pop("name")
    return kwargs

@arg_digest(standardizer=argument_names_standardization, strictness="warn")
def get(element_name=None, **kwargs):
    return element_name

Contract:

• input: (caller, kwargs),

• output: normalized kwargs dictionary.

Normalization runs before any argument digester.

Recommended modular layout

mylib/_private/digestion/normalization/
  aliases.py
  caller_rules.py
  dynamic_rules.py
  standardizer.py

Suggested responsibilities:

• aliases.py: global and local static aliases.

• caller_rules.py: caller-specific rename tables.

• dynamic_rules.py: conditional rules using runtime context.

• standardizer.py: orchestration entry point used by ArgDigest.

Dynamic rule example (name + element)

def normalize_get(caller, kwargs):
    if "name" in kwargs and "element" in kwargs:
        kwargs = dict(kwargs)
        kwargs[f"{kwargs['element']}_name"] = kwargs.pop("name")
    return kwargs

This pattern covers cases like:

• if element="atom" and name="CA", produce atom_name="CA".

Next

Continue with skip_digestion Behavior.