Improves Beancount entry generation and sanitization

Adds a function to sanitize strings for use as Beancount links,
ensuring compatibility with Beancount's link restrictions.

Refactors the journal entry creation process to use EUR-based
postings when fiat currency is provided, improving accuracy
and consistency. The legacy SATS-based fallback is retained for
cases without fiat currency information.

Adjusts reference generation for Beancount entries using the
sanitized description.
This commit is contained in:
padreug 2025-11-10 11:35:41 +01:00
parent 0e93fc5ffc
commit a6b67b7416
2 changed files with 57 additions and 11 deletions

View file

@ -238,6 +238,30 @@ class AccountLookup:
# ===== CONVERSION FUNCTIONS =====
def sanitize_link(text: str) -> str:
"""
Sanitize a string to make it valid for Beancount links.
Beancount links can only contain: A-Z, a-z, 0-9, -, _, /, .
All other characters are replaced with hyphens.
Examples:
>>> sanitize_link("Test (pending)")
'Test-pending'
>>> sanitize_link("Invoice #123")
'Invoice-123'
>>> sanitize_link("import-20250623-Action Ressourcerie")
'import-20250623-Action-Ressourcerie'
"""
import re
# Replace any character that's not alphanumeric, dash, underscore, slash, or period with a hyphen
sanitized = re.sub(r'[^A-Za-z0-9\-_/.]', '-', text)
# Remove consecutive hyphens
sanitized = re.sub(r'-+', '-', sanitized)
# Remove leading/trailing hyphens
sanitized = sanitized.strip('-')
return sanitized
def eur_to_sats(eur_amount: Decimal, btc_eur_rate: float) -> int:
"""Convert EUR to satoshis using BTC/EUR rate"""
btc_amount = eur_amount / Decimal(str(btc_eur_rate))
@ -454,10 +478,13 @@ def convert_to_castle_entry(parsed: dict, btc_eur_rate: float, account_lookup: A
"metadata": metadata
})
# Create sanitized reference link
desc_part = sanitize_link(parsed['description'][:30])
return {
"description": parsed['description'],
"entry_date": parsed['date'].isoformat(),
"reference": f"import-{parsed['date'].strftime('%Y%m%d')}-{parsed['description'][:20].replace(' ', '-')}",
"reference": f"import-{parsed['date'].strftime('%Y%m%d')}-{desc_part}",
"flag": "*",
"meta": {
"source": "beancount_import",