starred/terminal
1
0
Fork 0
mirror of https://github.com/microsoft/terminal.git synced 2026-02-04 05:35:20 +00:00
Code Issues 19.0k Packages Projects Releases 20 Wiki Activity

Properly support 2J sequence for clearing the screen #5971

New Issue
Open
opened 2026-01-31 00:26:49 +00:00 by claunia · 0 comments
claunia commented 2026-01-31 00:26:49 +00:00
Owner
Copy Link

Originally created by @zadjii-msft on GitHub (Jan 15, 2020).

This was taken from MSFT:21096414, which was found here in the code:

23d1bcbd94/src/renderer/vt/XtermEngine.cpp (L69-L78)

Apparently this bug never made the hop from ADO to Github. Full text below:

Currently, we only ever send one CSI 2 J from conpty - when we first start up.
Otherwise, the other call to ClearScreen is behind a conditional with !resized - see XtermEngine::StartPaint:

auto dirtyRect = GetDirtyRectInChars();
const auto dirtyView = Viewport::FromInclusive(dirtyRect);
if (!_resized && dirtyView == _lastViewport)
{
    RETURN_IF_FAILED(_ClearScreen());
    _clearedAllThisFrame = true;
}

Unfortunately, !resized will never be true - in state.cpp, in VtEngine::UpdateViewport, we always set _resized to true, and we always call UpdateViewport at the start of as frame.

What that means currently is that the terminal will never get cleared when the conpty is cleared.
Removing the _resized check, or changing the assignment of _resized to only when the dimensions change is NOT correct however.

There's no way to tell the difference between a frame that was all invalid was because the screen was cleared or because it contains all new text.

For example, if you make the simple fix described above, and run help or ping in vtpipeterm, then scroll into history, you might see the output duplicated. This would be because the pty emitted an entire screen of output, then on the next frame, determined everything was invalid, then cleared the terminal screen, then re-emitted the same screen of text.

So we're might be doing something funky in the full screen writing case, where we on the next frame still think everything is invalid. Maybe the scroll delta is causing us to have invalid text up near the top of the pty, causing us to think the screen should be cleared after a full screen of text.

Alternatively, as a half measure, we could have the adapter handle 2J, then in pty mode return false, causing the state machine to write the 2J to the renderer as well. That's probably not the right solution - we might want a specific call for the adapter to tell the renderer that it's requested an EraseAll, so the renderer can update it's state accordingly. Alternatively, the IRenderTarget could have a TriggerClearAll that does specifically that, though that would be relatively unused by the other engines, so I'm not a fan of that.
This half measure isn't great, because it wouldn't work for cmd.exe's cls - it would only work for things emitting a 2J (wsl)

Originally created by @zadjii-msft on GitHub (Jan 15, 2020). This was taken from MSFT:21096414, which was found here in the code: https://github.com/microsoft/terminal/blob/23d1bcbd94bbc8992bb9a30c8d31e3c7878528da/src/renderer/vt/XtermEngine.cpp#L69-L78 Apparently this bug never made the hop from ADO to Github. Full text below: Currently, we only ever send one `CSI 2 J` from conpty - when we first start up. Otherwise, the other call to `ClearScreen` is behind a conditional with `!resized` - see `XtermEngine::StartPaint`: ```c++ auto dirtyRect = GetDirtyRectInChars(); const auto dirtyView = Viewport::FromInclusive(dirtyRect); if (!_resized && dirtyView == _lastViewport) { RETURN_IF_FAILED(_ClearScreen()); _clearedAllThisFrame = true; } ``` Unfortunately, `!resized` will *never* be true - in `state.cpp`, in `VtEngine::UpdateViewport`, we always set `_resized` to true, and we always call `UpdateViewport` at the start of as frame. What that means currently is that the terminal will never get cleared when the conpty is cleared. Removing the `_resized` check, or changing the assignment of `_resized` to only when the dimensions change is NOT correct however. There's no way to tell the difference between a frame that was all invalid was because the screen was cleared or because it contains all new text. For example, if you make the simple fix described above, and run `help` or `ping` in `vtpipeterm`, then scroll into history, you might see the output duplicated. This would be because the pty emitted an entire screen of output, then on the next frame, determined everything was invalid, then cleared the terminal screen, then re-emitted the same screen of text. So we're might be doing something funky in the full screen writing case, where we on the next frame still think everything is invalid. Maybe the scroll delta is causing us to have invalid text up near the top of the pty, causing us to think the screen should be cleared after a full screen of text. Alternatively, as a half measure, we could have the adapter handle 2J, then in pty mode return false, causing the state machine to write the 2J to the renderer as well. That's probably not the right solution - we might want a specific call for the adapter to tell the renderer that it's requested an `EraseAll`, so the renderer can update it's state accordingly. Alternatively, the `IRenderTarget` could have a `TriggerClearAll` that does specifically that, though that would be relatively unused by the other engines, so I'm not a fan of that. This half measure isn't great, because it wouldn't work for cmd.exe's `cls` - it would only work for things emitting a 2J (wsl)
claunia added the Area-RenderingIssue-TaskNeeds-Tag-FixProduct-ConptyPriority-2 labels 2026-01-31 00:26:49 +00:00
Sign in to join this conversation.
No Branch/Tag Specified
Branches Tags
Labels
Clear labels
⛺ Reserved
A11yCO
A11yMAS
A11ySev1
A11ySev2
A11ySev3
A11yTTValidated
A11yUsable
A11yVoiceAccess
A11yWCAG
Area-Accessibility
Area-AtlasEngine
Area-AzureShell
Area-Build
Area-Build
Area-Chat
Area-CmdPal
Area-CodeHealth
Area-Commandline
Area-CookedRead
Area-DefApp
Area-Extensibility
Area-Fonts
Area-GroupPolicy
Area-i18n
Area-Input
Area-Interaction
Area-Interop
Area-Localization
Area-Output
Area-Performance
Area-Portable
Area-Quality
Area-Remoting
Area-Rendering
Area-Schema
Area-Server
Area-Settings
Area-SettingsUI
Area-ShellExtension
Area-ShellExtension
Area-ShellExtension
Area-Suggestions
Area-Suggestions
Area-TerminalConnection
Area-TerminalControl
Area-Theming
Area-UserInterface
Area-VT
Area-Windowing
Area-WPFControl
AutoMerge
Blocking-Ingestion
Culprit-Centennial
Culprit-WinUI
Disability-All
Disability-Blind
Disability-LowVision
Disability-Mobility
External-Blocked-WinUI3
Fixed
Gathering-Data
good first issue
HCL-E+D
HCL-WindowsTerminal
Help Wanted
Impact-Compatibility
Impact-Compliance
Impact-Correctness
Impact-Visual
In-PR
InclusionBacklog
InclusionBacklog-Windows TerminalWin32
InclusionCommitted-202206
Issue-Bug
Issue-Docs
Issue-Feature
Issue-Feature
Issue-Question
Issue-Samples
Issue-Scenario
Issue-Task
Needs-Attention
Needs-Author-Feedback
Needs-Bisect
Needs-Discussion
Needs-Repro
Needs-Tag-Fix
Needs-Tag-Fix
Needs-Triage
No-Recent-Activity
Priority-0
Priority-1
Priority-2
Priority-3
Product-Cmd.exe
Product-Colortool
Product-Colortool
Product-Colortool
Product-Conhost
Product-Conpty
Product-Meta
Product-Powershell
Product-Terminal
Product-WSL
pull-request
Mirrored from GitHub Pull Request
 Resolution-Answered
Resolution-By-Design
Resolution-Duplicate
Resolution-External
Resolution-Fix-Available
Resolution-Fix-Committed
Resolution-No-Repro
Resolution-Won't-Fix
Severity-Blocking
Severity-Crash
Severity-DataLoss
spam
this-will-be-a-breaking-change
Tracking-External
WindowsTerminal_Win32
Work-Item
zAskModeBug
zInbox-Bug
No Label Area-Rendering Issue-Task Needs-Tag-Fix Priority-2 Product-Conpty
Milestone
No items
No Milestone
Projects
Clear projects
No project
Assignees
Clear assignees
claunia
No Assignees
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: starred/terminal#5971
Delete Branch "%!s()"

Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?