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

[PR #4354] When the Terminal is resized, don't remove lines from scrollback #25718

New Issue
Open
opened 2026-01-31 09:11:20 +00:00 by claunia · 0 comments
claunia commented 2026-01-31 09:11:20 +00:00
Owner
Copy Link

Original Pull Request: https://github.com/microsoft/terminal/pull/4354

State: closed
Merged: No

Summary of the Pull Request

This PR fixes a bug in the Windows Terminal, where decreasing the height of the window would cause some lines from the scrollback to be lost. While a seemingly trivial bug, this actually involved some elaborate manipulation of how conpty decides to render a resize operation.

References

PR Checklist

Detailed Description of the Pull Request / Additional comments

Conpty was not emitting the correct thing when you'd decrease the height of the viewport. When you increase the height of the window, we'd both invalidate the entire buffer, and insert new lines to the bottom of the buffer. What this meant is that on every height-only resize, we'd re-emit the entire viewport.

  • If you resized down twice (From R to R-1 to R-2), once during the frame being emitted by conpty, the Terminal would try to process the R-1 lines emitted by conpty when there are only R-2 available rows in the terminal buffer. This would lead to duplicated lines.

This involved a bunch of changes throughout the codebase to make sure we weren't just calling TriggerRedraw on the entire viewport when we resize the buffer. Magically, conpty was pretty good about knowing what actually changed, though that did need some updating as well (in VtEngine::UpdateViewport).

  • Importantly, there's a minor change to ScreenInfo to cause it to only invalidate wrapped lines during a ResizeWithReflow. Previously, we'd be invalidating the entire viewport, but that caused issues terminal-side. It's better to only tell the terminal what actually changed during the reflow.

With those changes, the Terminal was changed to more intelligently stick to the right location in the buffer. When increasing the height of conpty, new lines always get inserted at the bottom of the viewport, so the terminal needs to stick to roughly the old top's position to re-create this stateWhen resizing down, empty lines are clipped from the conpty if there are any, then lines from the top are removed.

This is obviously all very complicated, so I've added some tests to validate all this. ConptyRoundtripTests::TestResizeHeight is elaborate and it does the right thing here, validating the contents of both conpty and the Terminal after emitting various numbers of lines into the buffer and doing arbitrary resizes.

As with any conpty change, I'm horrified that this will break in new and more horrifying ways. I'm hoping that a long bake time will help us identify bugs in this changelist (if there are any).

Validation Steps Performed

In VtPipeTerm and Windows Terminal:

  1. wsl seq 100
  2. resize down - scroll to make sure buffer has the right contents
  3. resize up - scroll to make sure buffer has the right contents
  4. resize down quickly - scroll to make sure buffer has the right contents
  5. resize up quickly - scroll to make sure buffer has the right contents

Additionally, resize quickly from the corner, and ensure ll the lines are right.

Aditionally, perform all the above with less than a full viewport of output.

Aditidionally, after the above tests, perform a sleep 2 ; printf "\e[6n" and quickly hit esc to query the cursors position in conpty, and make sure it lines up with the Terminal's.

Also play with resizing and PSReadline, and all the lovely things PSR does to the cursor.

This was the original PR body I had for this change. It's included here for future code archeologists. It's not really accurate, but it might guide people with the text that's in #3490 to help recreaate how I got here.

This PR encompasses a pair of fixes that are needed simultaneously to make the Windows Terminal work correctly.

  • Conpty was not emitting the correct thing when you'd decrease the height of the viewport. When you increase the height of the window, we'd both invalidate the entire buffer, and insert new lines to the bottom of the buffer. What this meant is that on every height-only resize, we'd re-emit the entire viewport.
    • If you resized down twice (From R to R-1 to R-2), once during the frame being emitted by conpty, the Terminal would try to process the R-1 lines emitted by conpty when there are only R-2 available rows in the terminal buffer. This would tlead to duplicated lines.
  • This buggy conpty behavior was causing us to correctly believe that the terminal viewport should remain "stuck" to the top position. That was what was causing us to apparently erase lines from the scrollback as we'd resize down. So this also changes the terminal to stick to the bottom when the viewport height changes.

This involved a bunch of changes throughout the codebase to make sure we weren't just calling TriggerRedraw on the entire viewport when we resize the buffer. Magically, conpty was pretty good about knowing what actually changed, though that did need some updating as well (in VtEngine::UpdateViewport).

As with any conpty change, I'm horrified that this will break in new and more horrifying ways. I'm hoping that a long bake time will help us identify bugs in this changelist (if there are any).

Validation Steps Performed

In VtPipeTerm and Windows Terminal:

  1. wsl seq 100
  2. resize down - scroll to make sure buffer has the right contents
  3. resize up - scroll to make sure buffer has the right contents
  4. resize down quickly - scroll to make sure buffer has the right contents
  5. resize up quickly - scroll to make sure buffer has the right contents
**Original Pull Request:** https://github.com/microsoft/terminal/pull/4354 **State:** closed **Merged:** No --- ## Summary of the Pull Request This PR fixes a bug in the Windows Terminal, where decreasing the height of the window would cause some lines from the scrollback to be lost. While a seemingly trivial bug, this actually involved some elaborate manipulation of how conpty decides to render a resize operation. ## References ## PR Checklist * [x] Closes #3490 * [x] I work here * [x] Tests added/passed - you don't even know * [n/a] Requires documentation to be updated ## Detailed Description of the Pull Request / Additional comments Conpty was not emitting the correct thing when you'd decrease the height of the viewport. When you increase the height of the window, we'd both invalidate the entire buffer, and insert new lines to the bottom of the buffer. What this meant is that on every height-only resize, we'd re-emit the entire viewport. - If you resized down twice (From `R` to `R-1` to `R-2`), once _during the frame being emitted by conpty_, the Terminal would try to process the `R-1` lines emitted by conpty when there are only `R-2` available rows in the terminal buffer. This would lead to duplicated lines. This involved a bunch of changes throughout the codebase to make sure we weren't just calling `TriggerRedraw` on the entire viewport when we resize the buffer. Magically, conpty was pretty good about knowing what actually changed, though that did need some updating as well (in `VtEngine::UpdateViewport`). - Importantly, there's a minor change to `ScreenInfo` to cause it to only invalidate wrapped lines during a ResizeWithReflow. Previously, we'd be invalidating the entire viewport, but that caused issues terminal-side. It's better to only tell the terminal what actually changed during the reflow. With those changes, the Terminal was changed to more intelligently stick to the right location in the buffer. When increasing the height of conpty, new lines always get inserted at the bottom of the viewport, so the terminal needs to stick to _roughly_ the old top's position to re-create this stateWhen resizing down, _empty_ lines are clipped from the conpty if there are any, then lines from the top are removed. This is obviously all very complicated, so I've added some tests to validate all this. `ConptyRoundtripTests::TestResizeHeight` is _elaborate_ and it does the right thing here, validating the contents of both conpty and the Terminal after emitting various numbers of lines into the buffer and doing arbitrary resizes. As with any conpty change, I'm horrified that this will break in new and more horrifying ways. I'm hoping that a long bake time will help us identify bugs in this changelist (if there are any). ## Validation Steps Performed In VtPipeTerm and Windows Terminal: 1. `wsl seq 100` 2. resize down - scroll to make sure buffer has the right contents 3. resize up - scroll to make sure buffer has the right contents 4. resize down _quickly_ - scroll to make sure buffer has the right contents 5. resize up _quickly_ - scroll to make sure buffer has the right contents Additionally, resize quickly from the corner, and ensure ll the lines are right. Aditionally, perform all the above with less than a full viewport of output. Aditidionally, after the above tests, perform a `sleep 2 ; printf "\e[6n"` and quickly hit <kbd>esc</kbd> to query the cursors position in conpty, and make sure it lines up with the Terminal's. Also play with resizing and PSReadline, and all the lovely things PSR does to the cursor. <details> <summary> This was the original PR body I had for this change. It's included here for future code archeologists. It's not really accurate, but it might guide people with the text that's in #3490 to help recreaate how I got here.</summary> This PR encompasses a pair of fixes that are needed simultaneously to make the Windows Terminal work correctly. * Conpty was not emitting the correct thing when you'd decrease the height of the viewport. When you increase the height of the window, we'd both invalidate the entire buffer, and insert new lines to the bottom of the buffer. What this meant is that on every height-only resize, we'd re-emit the entire viewport. - If you resized down twice (From `R` to `R-1` to `R-2`), once _during the frame being emitted by conpty_, the Terminal would try to process the `R-1` lines emitted by conpty when there are only `R-2` available rows in the terminal buffer. This would tlead to duplicated lines. * This buggy conpty behavior was causing us to correctly believe that the terminal viewport should remain "stuck" to the top position. That was what was causing us to apparently erase lines from the scrollback as we'd resize down. So this also changes the terminal to stick to the bottom when the viewport height changes. This involved a bunch of changes throughout the codebase to make sure we weren't just calling `TriggerRedraw` on the entire viewport when we resize the buffer. Magically, conpty was pretty good about knowing what actually changed, though that did need some updating as well (in `VtEngine::UpdateViewport`). As with any conpty change, I'm horrified that this will break in new and more horrifying ways. I'm hoping that a long bake time will help us identify bugs in this changelist (if there are any). ## Validation Steps Performed In VtPipeTerm and Windows Terminal: 1. `wsl seq 100` 2. resize down - scroll to make sure buffer has the right contents 3. resize up - scroll to make sure buffer has the right contents 4. resize down _quickly_ - scroll to make sure buffer has the right contents 5. resize up _quickly_ - scroll to make sure buffer has the right contents </details>
claunia added the pull-request label 2026-01-31 09:11:20 +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 pull-request
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#25718
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?