From f888b13c56b0b2a6bb92ffbcd624c6dad002d83d 2017-10-22 18:05:56
From: Brett Smith <brettcsmith@brettcsmith.org>
Date: 2017-10-22 18:05:56
Subject: [PATCH] hooks.add_entity: Better handle common name prefix parts.

Keep these with the name they're attached to, rather than breaking the name
in the middle.

---

diff --git a/import2ledger/hooks/add_entity.py b/import2ledger/hooks/add_entity.py
index d0f4791453e7fe79c19d2e630a12fda4fb667294..d4f6191055304e7fdbafffabb9d797d75edaeee1 100644
--- a/import2ledger/hooks/add_entity.py
+++ b/import2ledger/hooks/add_entity.py
@@ -2,6 +2,11 @@ import re
 import unicodedata
 
 class AddEntityHook:
+    NAME_PREFIXES = frozenset([
+        'da',
+        'de',
+        'van',
+    ])
     NONASCII_RE = re.compile(r'[^-A-Za-z0-9]')
     NONALNUM_RE = re.compile(r'[^-\w]')
     OPEN_PARENS = ['\\(', '\\[', '\\{']
@@ -41,6 +46,8 @@ class AddEntityHook:
         if not parts:
             return ''
         parts.insert(0, parts.pop())
+        if parts[-1].lower() in self.NAME_PREFIXES:
+            parts.insert(0, parts.pop())
         return '-'.join(parts)
 
     def run(self, data):
diff --git a/tests/test_hooks.py b/tests/test_hooks.py
index d9f68a3d0d4c1b2a2fce5c059c1443318bcc85e3..270209f5f1b525ae2ec00ae781ca9b8555c547d8 100644
--- a/tests/test_hooks.py
+++ b/tests/test_hooks.py
@@ -19,6 +19,9 @@ def test_load_all():
     ('Alex(Nickname) Smith', 'Smith-Alex'),
     ('稲荷', '稲荷'),
     ('Pøweł', 'Powel'),
+    ('Elyse Jan Smith', 'Smith-Elyse-Jan'),
+    ('Jan van Smith', 'van-Smith-Jan'),
+    ('Francis da Silva', 'da-Silva-Francis'),
 ])
 def test_add_entity(payee, expected):
     data = {'payee': payee}