Messing up \t \r \n in utf8 files
Resolved 💬 3 comments Opened Jan 9, 2026 by ToolTech Closed Jan 9, 2026
💡 Likely answer: A maintainer (etraut-openai, contributor)
responded on this thread — see the highlighted reply below.
What version of Codex is running?
latest
What subscription do you have?
plus + extra credits
Which model were you using?
5.2 codex xhigh
What platform is your computer?
Win64
What issue are you seeing?
I spend a lot of credits all the time on Windows working with utf8 and swedish. codex messes up my files, adds blank lines and control characters, and then removes all my code formatting sometimes. quite annoying and expensive
What steps can reproduce the bug?
Uploaded thread: 019b9c71-d17c-7523-8151-dee89d175af6
What is the expected behavior?
_No response_
Additional information
_No response_
3 Comments
Thanks for the bug report. This appears to be the same issue as #7582. We have implemented an experimental fix for this. Please refer to instructions in the linked bug to enable this, and let us know if it addresses your problem. Once we receive confirmation that the fix works, we'll enable it by default.
Some errors relate to utf 8 double conversion. Some errors relate to cr lf problems. Python adds extra ctrl signs and some merges accidently write out control chars like tab as text \t instead of actual tab controlBest regards!/Anders Modén9 jan. 2026 kl. 20:53 skrev Eric Traut @.***>:etraut-openai left a comment (openai/codex#8993)
Thanks for the bug report. This appears to be the same issue as #7582. We have implemented an experimental fix for this. Please refer to instructions in the linked bug to enable this, and let us know if it addresses your problem. Once we receive confirmation that the fix works, we'll enable it by default.
—Reply to this email directly, view it on GitHub, or unsubscribe.You are receiving this because you authored the thread.Message ID: @.***>
this is still an issue