Rendering

Overview

Since all models inherit from Pydantic they gain access to useful methods for rendering the contained data in different formats. Namely, the ability to generate JSON:

import bdantic

from beancount import loader

bfile = bdantic.parse_loader(*loader.load_file("ledger.beancount"))
js = bfile.json()
print(js) # Look ma, my beancount data in JSON!

This allows, for example, processing some Beancount data and then exporting it to another language or tool which can ingest JSON. The resulting JSON can be parsed back into a model:

from bdantic.models import BeancountFile

bfile = BeancountFile.parse_raw(js)

This behavior allows one to write or generate Beancount constructs in JSON and then parse them back into their native types in Python.

Syntax

In addition to the support provided by Pydantic, directive models also provide a useful method for generating raw Beancount syntax:

from bdantic.models import Amount, Posting, Transaction
from datetime import date
from decimal import Decimal

txn = Transaction(
    date=date.today(),
    meta={},
    flag="*",
    payee="Home Depot",
    narration="Tools n stuff",
    tags=None,
    links=None,
    postings=[
        Posting(
            account="Assets:Bank:Cash",
            units=Amount(number=Decimal(-142.32), currency="USD"),
            cost=None,
            CostSpec=None,
            flag=None,
            meta={},
        ),
        Posting(
            account="Expenses:HomeDepot",
            units=Amount(number=Decimal(142.32), currency="USD"),
            cost=None,
            CostSpec=None,
            flag=None,
            meta={},
        ),
    ],
)

print(txn.syntax())

Combined with the above, it's possible to go all the way from JSON to valid Beancount syntax with the use of bdantic.