WIP: test windows PTY #7742
WIP: test windows PTY #7742
Conversation
![chatgpt-codex-connector[bot]](https://avatars.githubusercontent.com/in/1144995?s=60&v=4){kind=small}
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
ℹ️ About Codex in GitHub
Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you
- Open a pull request for review
- Mark a draft as ready
- Comment "@codex review".
If Codex has suggestions, it will comment; otherwise it will react with 👍.
Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".
|
@codex review |
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
ℹ️ About Codex in GitHub
Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you
- Open a pull request for review
- Mark a draft as ready
- Comment "@codex review".
If Codex has suggestions, it will comment; otherwise it will react with 👍.
Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".
| if cfg!(target_os = "windows") { | ||
| let mut parser = vt100::Parser::new(40, 120, 1_024); | ||
| parser.process(raw.as_bytes()); | ||
| parser.screen_mut().set_scrollback(usize::MAX); | ||
| return parser.screen().contents(); |
There was a problem hiding this comment.
On Windows we now pass unified-exec output through a fresh vt100::Parser with a scrollback of only 1 024 lines (Parser::new(40, 120, 1_024)) and only set scrollback to usize::MAX after processing. Any command that emits more than 1 024 lines will therefore have all earlier lines discarded before token truncation and before being sent back to the user, whereas other platforms still preserve everything up to UNIFIED_EXEC_OUTPUT_MAX_BYTES/max_output_tokens. This is a regression for long-running commands on Windows because their output is silently chopped to the last ~1 024 lines even when higher limits are configured.
Useful? React with 👍 / 👎.
No description provided.