From 730de9cc7f521a4a444daf42675e5e7b5260a377 Mon Sep 17 00:00:00 2001
From: Yash Raj Pandey <55940078+devYRPauli@users.noreply.github.com>
Date: Tue, 28 Jul 2026 05:04:11 -0400
Subject: [PATCH] Fix tag CSV parser splicing wrong text after a multi-line
 quoted field (#17131)

---
 rag/app/tag.py | 8 ++++++--
 1 file changed, 6 insertions(+), 2 deletions(-)

diff --git a/rag/app/tag.py b/rag/app/tag.py
index 2c3a27785e..656226a792 100644
--- a/rag/app/tag.py
+++ b/rag/app/tag.py
@@ -96,12 +96,16 @@ def chunk(filename, binary=None, lang="Chinese", callback=None, **kwargs):
         fails = []
         content = ""
         res = []
-        reader = csv.reader(lines)
+        reader = csv.reader((line + "\n" for line in lines))
+        prev_line_num = 0
 
+        # line_num tracks the physical span when quoted fields cross lines.
         for i, row in enumerate(reader):
+            raw = "\n".join(lines[prev_line_num : reader.line_num])
+            prev_line_num = reader.line_num
             row = [r.strip() for r in row if r.strip()]
             if len(row) != 2:
-                content += "\n" + lines[i]
+                content += "\n" + raw
             elif len(row) == 2:
                 content += "\n" + row[0]
                 res.append(beAdoc(deepcopy(doc), content, row[1], eng, i))