From 52917352389427facfc0717f187b6586dce68417 Mon Sep 17 00:00:00 2001
From: Claude <noreply@anthropic.com>
Date: Mon, 13 Apr 2026 12:22:37 +0000
Subject: [PATCH] fix: preserve trailing punctuation when extracting PIB words
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

parse_hybrid_line() was appending words with trailing commas
(e.g. "Володимировича,") to pib_words, causing the punctuation
to be swallowed during masking replacement. Now strips trailing
punctuation (,.!?;:) from each PIB word before adding it.

https://claude.ai/code/session_01KownKVoEBudRVp7npQ1idz
---
 data_masking.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/data_masking.py b/data_masking.py
index 8ecdce2..88fe398 100644
--- a/data_masking.py
+++ b/data_masking.py
@@ -738,7 +738,7 @@ def parse_hybrid_line(line: str) -> Tuple[Optional[str], Optional[str], Optional
     pib_words = []
     for word in parts[pib_start_index:pib_start_index + 3]:
         if looks_like_name(word):
-            pib_words.append(word)
+            pib_words.append(word.rstrip(',.!?;:'))
         else:
             break
     pib = " ".join(pib_words) if pib_words else None