From 0c9a79f0838991ca3e15fd51e759030d3b61f661 Mon Sep 17 00:00:00 2001
From: zekiye <111131603+zekiyemeral@users.noreply.github.com>
Date: Wed, 3 Jun 2026 17:50:55 +0300
Subject: [PATCH 1/2] fix: escape pipe characters in CSV table cells

---
 .../markitdown/converters/_csv_converter.py   | 20 +++++++++++++++++--
 1 file changed, 18 insertions(+), 2 deletions(-)

diff --git a/packages/markitdown/src/markitdown/converters/_csv_converter.py b/packages/markitdown/src/markitdown/converters/_csv_converter.py
index 7e9631e1b..21183050a 100644
--- a/packages/markitdown/src/markitdown/converters/_csv_converter.py
+++ b/packages/markitdown/src/markitdown/converters/_csv_converter.py
@@ -1,17 +1,25 @@
 import csv
 import io
 from typing import BinaryIO, Any
+
 from charset_normalizer import from_bytes
+
 from .._base_converter import DocumentConverter, DocumentConverterResult
 from .._stream_info import StreamInfo
 
+
 ACCEPTED_MIME_TYPE_PREFIXES = [
     "text/csv",
     "application/csv",
 ]
+
 ACCEPTED_FILE_EXTENSIONS = [".csv"]
 
 
+def _escape_markdown_table_cell(cell: str) -> str:
+    return cell.replace("|", r"\|")
+
+
 class CsvConverter(DocumentConverter):
     """
     Converts CSV files to Markdown tables.
@@ -28,11 +36,14 @@ def accepts(
     ) -> bool:
         mimetype = (stream_info.mimetype or "").lower()
         extension = (stream_info.extension or "").lower()
+
         if extension in ACCEPTED_FILE_EXTENSIONS:
             return True
+
         for prefix in ACCEPTED_MIME_TYPE_PREFIXES:
             if mimetype.startswith(prefix):
                 return True
+
         return False
 
     def convert(
@@ -58,7 +69,8 @@ def convert(
         markdown_table = []
 
         # Add header row
-        markdown_table.append("| " + " | ".join(rows[0]) + " |")
+        header_row = [_escape_markdown_table_cell(cell) for cell in rows[0]]
+        markdown_table.append("| " + " | ".join(header_row) + " |")
 
         # Add separator row
         markdown_table.append("| " + " | ".join(["---"] * len(rows[0])) + " |")
@@ -68,10 +80,14 @@ def convert(
             # Make sure row has the same number of columns as header
             while len(row) < len(rows[0]):
                 row.append("")
+
             # Truncate if row has more columns than header
             row = row[: len(rows[0])]
-            markdown_table.append("| " + " | ".join(row) + " |")
+
+            escaped_row = [_escape_markdown_table_cell(cell) for cell in row]
+            markdown_table.append("| " + " | ".join(escaped_row) + " |")
 
         result = "\n".join(markdown_table)
 
         return DocumentConverterResult(markdown=result)
+    
\ No newline at end of file

From 226668c05b2bba95b586403241bdff016bcc3896 Mon Sep 17 00:00:00 2001
From: zekiye <111131603+zekiyemeral@users.noreply.github.com>
Date: Wed, 3 Jun 2026 18:46:26 +0300
Subject: [PATCH 2/2] test: add regression coverage for CSV pipe escaping

---
 packages/markitdown/tests/_test_vectors.py      | 17 +++++++++++++++++
 .../tests/test_files/test_pipe_chars.csv        |  4 ++++
 2 files changed, 21 insertions(+)
 create mode 100644 packages/markitdown/tests/test_files/test_pipe_chars.csv

diff --git a/packages/markitdown/tests/_test_vectors.py b/packages/markitdown/tests/_test_vectors.py
index 74fa9bd0a..a30f84d35 100644
--- a/packages/markitdown/tests/_test_vectors.py
+++ b/packages/markitdown/tests/_test_vectors.py
@@ -151,6 +151,23 @@ class FileTestVector(object):
             "| 髙橋淳 | 35 | 名古屋 |",
         ],
         must_not_include=[],
+    ),
+        FileTestVector(
+        filename="test_pipe_chars.csv",
+        mimetype="text/csv",
+        charset="ascii",
+        url=None,
+        must_include=[
+            "| Name | Formula | Notes |",
+            "| --- | --- | --- |",
+            "| OR Gate | A \\| B | Pipe character should be escaped |",
+            "| AND Gate | A & B | Regular value |",
+            "| Regex | foo\\|bar | Pipe in regex-like value |",
+        ],
+        must_not_include=[
+            "| OR Gate | A | B | Pipe character should be escaped |",
+            "| Regex | foo|bar | Pipe in regex-like value |",
+        ],
     ),
     FileTestVector(
         filename="test.json",
diff --git a/packages/markitdown/tests/test_files/test_pipe_chars.csv b/packages/markitdown/tests/test_files/test_pipe_chars.csv
new file mode 100644
index 000000000..769679291
--- /dev/null
+++ b/packages/markitdown/tests/test_files/test_pipe_chars.csv
@@ -0,0 +1,4 @@
+Name,Formula,Notes
+OR Gate,A | B,Pipe character should be escaped
+AND Gate,A & B,Regular value
+Regex,foo|bar,Pipe in regex-like value
\ No newline at end of file