From c487aa1949bd725a9f825451bbfc9bd5160be737 Mon Sep 17 00:00:00 2001
From: Ghraven <115199279+Ghraven@users.noreply.github.com>
Date: Wed, 20 May 2026 18:48:39 +0000
Subject: [PATCH] fix: add explicit encoding="utf-8" to .txt read in
 _validators.py

The text-mode open() for user-supplied .txt fine-tuning files relied on
the platform default encoding (cp1252 on Windows), which raises
UnicodeDecodeError or corrupts non-ASCII content on valid UTF-8 files.
Pinning encoding="utf-8" makes the read consistent across platforms.
---
 src/openai/lib/_validators.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/src/openai/lib/_validators.py b/src/openai/lib/_validators.py
index cf24cd2294..a947283641 100644
--- a/src/openai/lib/_validators.py
+++ b/src/openai/lib/_validators.py
@@ -482,7 +482,7 @@ def read_any_format(
             elif fname.lower().endswith(".txt"):
                 immediate_msg = "\n- Based on your file extension, you provided a text file"
                 necessary_msg = "Your format `TXT` will be converted to `JSONL`"
-                with open(fname, "r") as f:
+                with open(fname, "r", encoding="utf-8") as f:
                     content = f.read()
                     df = pd.DataFrame(
                         [["", line] for line in content.split("\n")],