View on GitHub
Star —
Used by —

Internals

This page documents internal implementation details for contributors. Public users should rely on smartfaker.Faker and the helpers in smartfaker.iban instead.

Module layout

  • smartfaker/__init__.py — package metadata and re-exports.

  • smartfaker/fake.py — Faker and address generation.

  • smartfaker/iban.py — bank code data, country format specs, per-country BBAN generators, MOD-97 helpers and async wrappers.

  • smartfaker/data/ — bundled per-country address JSON files.

  • compiler/docs/compiler.py — auto-generates per-method / per-function RST pages under docs/source/api/.

Async wrapper strategy

Async methods on Faker (aaddress, abatch_addresses, aiban) all dispatch their synchronous counterparts to asyncio.to_thread(). This guarantees:

  • the event loop never blocks, even for very large amount values;

  • semantics, return shape and exceptions are bit-for-bit identical to the sync API — there is no second implementation to keep in sync.

The IBAN helpers (acalculate_check_digits, avalidate_iban, agenerate_*) are thin coroutines that call their sync siblings directly. They exist so async code can await them without ceremony.

Determinism

SmartFaker uses the global random module under the hood. Seed it yourself if you need reproducible output:

import random
from smartfaker import Faker

random.seed(1234)
faker = Faker()
print(faker.iban("DE")["iban"])