From 51ae2e8e47ed1af9df3a74b2c49ddc53f47d840b Mon Sep 17 00:00:00 2001
From: padreug <padreug@aiolabs.dev>
Date: Mon, 10 Nov 2025 15:04:27 +0100
Subject: [PATCH] Sanitizes reference links for Beancount

Ensures that user-provided reference strings for expense,
receivable, and revenue entries are sanitized before being
included as Beancount links. This prevents issues caused by
invalid characters in the links, improving compatibility with
Beancount's link format. A new utility function is introduced
to handle the sanitization process.
---
 beancount_format.py | 25 +++++++++++++++++++++++++
 views_api.py        | 18 +++++++++---------
 2 files changed, 34 insertions(+), 9 deletions(-)

diff --git a/beancount_format.py b/beancount_format.py
index 06f0afb..2819e67 100644
--- a/beancount_format.py
+++ b/beancount_format.py
@@ -14,6 +14,31 @@ Key concepts:
 from datetime import date, datetime
 from decimal import Decimal
 from typing import Any, Dict, List, Optional
+import re
+
+
+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("castle-abc123")
+        'castle-abc123'
+    """
+    # 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 format_transaction(
diff --git a/views_api.py b/views_api.py
index 783309c..b6e95dc 100644
--- a/views_api.py
+++ b/views_api.py
@@ -802,7 +802,7 @@ async def api_create_expense_entry(
 
     # Format as Beancount entry and submit to Fava
     from .fava_client import get_fava_client
-    from .beancount_format import format_expense_entry
+    from .beancount_format import format_expense_entry, sanitize_link
 
     fava = get_fava_client()
 
@@ -814,10 +814,10 @@ async def api_create_expense_entry(
     import uuid
     entry_id = str(uuid.uuid4()).replace("-", "")[:16]
 
-    # Add castle ID as reference/link
+    # Add castle ID as reference/link (sanitized for Beancount)
     castle_reference = f"castle-{entry_id}"
     if data.reference:
-        castle_reference = f"{data.reference}-{entry_id}"
+        castle_reference = f"{sanitize_link(data.reference)}-{entry_id}"
 
     # Format Beancount entry
     entry = format_expense_entry(
@@ -930,7 +930,7 @@ async def api_create_receivable_entry(
 
     # Format as Beancount entry and submit to Fava
     from .fava_client import get_fava_client
-    from .beancount_format import format_receivable_entry
+    from .beancount_format import format_receivable_entry, sanitize_link
 
     fava = get_fava_client()
 
@@ -942,10 +942,10 @@ async def api_create_receivable_entry(
     import uuid
     entry_id = str(uuid.uuid4()).replace("-", "")[:16]
 
-    # Add castle ID as reference/link
+    # Add castle ID as reference/link (sanitized for Beancount)
     castle_reference = f"castle-{entry_id}"
     if data.reference:
-        castle_reference = f"{data.reference}-{entry_id}"
+        castle_reference = f"{sanitize_link(data.reference)}-{entry_id}"
 
     # Format Beancount entry
     entry = format_receivable_entry(
@@ -1007,7 +1007,7 @@ async def api_create_revenue_entry(
     Submits entry to Fava/Beancount.
     """
     from .fava_client import get_fava_client
-    from .beancount_format import format_revenue_entry
+    from .beancount_format import format_revenue_entry, sanitize_link
 
     # Get revenue account
     revenue_account = await get_account_by_name(data.revenue_account)
@@ -1057,10 +1057,10 @@ async def api_create_revenue_entry(
     import uuid
     entry_id = str(uuid.uuid4()).replace("-", "")[:16]
 
-    # Add castle ID as reference/link
+    # Add castle ID as reference/link (sanitized for Beancount)
     castle_reference = f"castle-{entry_id}"
     if data.reference:
-        castle_reference = f"{data.reference}-{entry_id}"
+        castle_reference = f"{sanitize_link(data.reference)}-{entry_id}"
 
     entry = format_revenue_entry(
         payment_account=payment_account.name,