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

Discuss: Should ED 2 (EraseType::All) push text on the screen into the scrollback? #10983

New Issue
Open
opened 2026-01-31 02:35:26 +00:00 by claunia · 8 comments
claunia commented 2026-01-31 02:35:26 +00:00
Owner
Copy Link

Originally created by @skyline75489 on GitHub (Oct 12, 2020).

Description of the new feature/enhancement

So I was investigating the "docker stats" flicking issue(#1680) and found something weird. The sequences generated by docker stats first erase-all then print the stats. The way conhost and WT implements this (ITerminalApi::EraseInDisplay), is to push everything on the screen into the scrollback and print new things:

image

Proposed technical implementation details (optional)

I don't know why this is like this but I'm sure there's a reason, and I feel that changing this might fix the flickering.

Originally created by @skyline75489 on GitHub (Oct 12, 2020). <!-- 🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨 I ACKNOWLEDGE THE FOLLOWING BEFORE PROCEEDING: 1. If I delete this entire template and go my own path, the core team may close my issue without further explanation or engagement. 2. If I list multiple bugs/concerns in this one issue, the core team may close my issue without further explanation or engagement. 3. If I write an issue that has many duplicates, the core team may close my issue without further explanation or engagement (and without necessarily spending time to find the exact duplicate ID number). 4. If I leave the title incomplete when filing the issue, the core team may close my issue without further explanation or engagement. 5. If I file something completely blank in the body, the core team may close my issue without further explanation or engagement. All good? Then proceed! --> # Description of the new feature/enhancement So I was investigating the "docker stats" flicking issue(#1680) and found something weird. The sequences generated by `docker stats` first erase-all then print the stats. The way conhost and WT implements this (ITerminalApi::EraseInDisplay), is to push everything on the screen into the scrollback and print new things: ![image](https://user-images.githubusercontent.com/4710575/95733627-fccef400-0cb4-11eb-8c29-c4f05c8f3915.png) <!-- A clear and concise description of what the problem is that the new feature would solve. Describe why and how a user would use this new functionality (if applicable). --> # Proposed technical implementation details (optional) <!-- A clear and concise description of what you want to happen. --> I don't know why this is like this but I'm sure there's a reason, and I feel that changing this might fix the flickering.
claunia added the Product-ConhostIssue-QuestionArea-VT labels 2026-01-31 02:35:27 +00:00
claunia commented 2026-01-31 02:35:28 +00:00
Author
Owner
Copy Link

@zadjii-msft commented on GitHub (Oct 12, 2020):

I'm almost certain that's how it's supposed to work. IIRC, conhost did this wrong for our first few releases with VT support.

@j4james can put me on blast if I'm wrong about this

@zadjii-msft commented on GitHub (Oct 12, 2020): I'm almost _certain_ that's how it's supposed to work. IIRC, conhost did this wrong for our first few releases with VT support. @j4james can put me on blast if I'm wrong about this
claunia commented 2026-01-31 02:35:28 +00:00
Author
Owner
Copy Link

@skyline75489 commented on GitHub (Oct 12, 2020):

I was seeing a different behavior in wsltty which leads me to think that WT is weird. Then I confirmed that iTerm2 is also doing the same thing. So…guess I’m misunderstanding this. Still, this seems to be the root cause of the flickering, I think somewhere the refresh rate is limited by something.

获取 Outlook for iOShttps://aka.ms/o0ukef

@skyline75489 commented on GitHub (Oct 12, 2020): I was seeing a different behavior in wsltty which leads me to think that WT is weird. Then I confirmed that iTerm2 is also doing the same thing. So…guess I’m misunderstanding this. Still, this seems to be the root cause of the flickering, I think somewhere the refresh rate is limited by something. 获取 Outlook for iOS<https://aka.ms/o0ukef>
claunia commented 2026-01-31 02:35:28 +00:00
Author
Owner
Copy Link

@j4james commented on GitHub (Oct 12, 2020):

I'm honestly not sure where the idea of scrolling to clear the screen came from. The DEC STD 070 manual doesn't say anything about ED 2 scrolling, so I don't think think it's something the original hardware terminals did.

Some terminal emulators do scroll, and some don't. VTE is one popular example that does, while XTerm doesn't (at least by default - it might have an option somewhere for that). If we really want to be compatible with XTerm, we may want to consider disabling this by default, and maybe have an option to reenable it for those users that prefer the current behaviour.

As for the docker stats performance issue, is that a problem in conhost too? Because it occurred to me that it might just be relying on the alt buffer, which WT doesn't yet support (see #381). Clearing the screen while in the alt buffer shouldn't normally scroll anything because there is no scrollback, but that's not the case in WT.

@j4james commented on GitHub (Oct 12, 2020): I'm honestly not sure where the idea of scrolling to clear the screen came from. The DEC STD 070 manual doesn't say anything about ED 2 scrolling, so I don't think think it's something the original hardware terminals did. Some terminal emulators do scroll, and some don't. VTE is one popular example that does, while XTerm doesn't (at least by default - it might have an option somewhere for that). If we really want to be compatible with XTerm, we may want to consider disabling this by default, and maybe have an option to reenable it for those users that prefer the current behaviour. As for the docker stats performance issue, is that a problem in conhost too? Because it occurred to me that it might just be relying on the alt buffer, which WT doesn't yet support (see #381). Clearing the screen while in the alt buffer shouldn't normally scroll anything because there is no scrollback, but that's not the case in WT.
claunia commented 2026-01-31 02:35:29 +00:00
Author
Owner
Copy Link

@j4james commented on GitHub (Oct 12, 2020):

I just remembered there was some previous discussion on this subject in issue https://github.com/microsoft/terminal/issues/2832#issuecomment-533779605.

Also I think VTE was (and possibly still is) considering matching the XTerm behaviour (see here).

@j4james commented on GitHub (Oct 12, 2020): I just remembered there was some previous discussion on this subject in issue https://github.com/microsoft/terminal/issues/2832#issuecomment-533779605. Also I think VTE was (and possibly still is) considering matching the XTerm behaviour (see [here](https://bugzilla.gnome.org/show_bug.cgi?id=506438#c5)).
claunia commented 2026-01-31 02:35:30 +00:00
Author
Owner
Copy Link

@jack775544 commented on GitHub (Oct 12, 2020):

As for the docker stats performance issue, is that a problem in conhost too?

I checked this on both both conhost shipped in windows 2004 and OpenConsole shipped with terminal 1.4 and it still flashes the same as on WT.

@jack775544 commented on GitHub (Oct 12, 2020): >As for the docker stats performance issue, is that a problem in conhost too? I checked this on both both conhost shipped in windows 2004 and OpenConsole shipped with terminal 1.4 and it still flashes the same as on WT.
claunia commented 2026-01-31 02:35:30 +00:00
Author
Owner
Copy Link

@skyline75489 commented on GitHub (Oct 13, 2020):

I can also confirm the flickering in both conhost and WT. And no, it does not use alt buffer to my knowledge. It's just pure ED2 and print.

Because of the current clearing behavior, WT also has the same "Ctrl-L adding empty line" issue, as described in VTE 506438 and Konsole 384218. Also @egmontkob has filed this https://gitlab.gnome.org/GNOME/vte/-/issues/84 .

Interestingly, I can not reproduce the "Ctrl-L" bug in Ubuntu Terminal or kitty. I don't have KDE at hand so I can't check Konsole. And it is indeed an annoying bug if someone is using "Ctrl-L" a lot.

@skyline75489 commented on GitHub (Oct 13, 2020): I can also confirm the flickering in both conhost and WT. And no, it does not use alt buffer to my knowledge. It's just pure ED2 and print. Because of the current clearing behavior, WT also has the same "Ctrl-L adding empty line" issue, as described in [VTE 506438](https://bugzilla.gnome.org/show_bug.cgi?id=506438#c5) and [Konsole 384218](https://bugs.kde.org/show_bug.cgi?id=384218). Also @egmontkob has filed this https://gitlab.gnome.org/GNOME/vte/-/issues/84 . Interestingly, I can not reproduce the "Ctrl-L" bug in Ubuntu Terminal or kitty. I don't have KDE at hand so I can't check Konsole. And it is indeed an annoying bug if someone is using "Ctrl-L" a lot.
claunia commented 2026-01-31 02:35:30 +00:00
Author
Owner
Copy Link

@j4james commented on GitHub (Oct 13, 2020):

For the record, I'd be happy for us to make ED2 just clear the screen normally (i.e. matching the XTerm behaviour), unless there's evidence that the scrolling thing is an actual DEC standard (and I haven't seen anything suggesting that).

@j4james commented on GitHub (Oct 13, 2020): For the record, I'd be happy for us to make ED2 just clear the screen normally (i.e. matching the XTerm behaviour), unless there's evidence that the scrolling thing is an actual DEC standard (and I haven't seen anything suggesting that).
claunia commented 2026-01-31 02:35:31 +00:00
Author
Owner
Copy Link

@DHowett commented on GitHub (Oct 23, 2020):

I'm comfortable with the current scrolling behavior, but I'll keep this in our back pocket in case we want to change it.

If we really want to be compatible with XTerm ...

I'm honestly fifty-fifty on whether this is a compatibility break. There's a line between application-visible compatibility (sequences change the addressable buffer in a known way, colors are treated just so, etc.) and user-visible compatibility (scrollback acts like this, the mouse does this, etc.), and I prefer our xterm compatibility to be to be on the application side.

Xterm doesn't do a lot of the things we do, but if applications can't detect that chicanery I'm less worried about it. If we wanted to be xterm's user interface, we could cut out a whole bunch of our work!

I'm going to keep this one open as a discussion. I see the argument that it's unfortunate from a security perspective, but I know that we'd get a lot of flak for changing it.

@DHowett commented on GitHub (Oct 23, 2020): I'm comfortable with the current scrolling behavior, but I'll keep this in our back pocket in case we want to change it. > If we really want to be compatible with XTerm ... I'm honestly fifty-fifty on whether this is a compatibility break. There's a line between application-visible compatibility (sequences change the addressable buffer in a known way, colors are treated just so, etc.) and user-visible compatibility (scrollback acts like this, the mouse does this, etc.), and I prefer our xterm compatibility to be to be on the _application_ side. Xterm doesn't do a lot of the things we do, but if applications can't detect that chicanery I'm less worried about it. If we wanted to be _xterm's user interface_, we could cut out a whole bunch of our work! I'm going to keep this one open as a discussion. I see the argument that it's unfortunate from a security perspective, but I know that we'd get a lot of flak for changing it.
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-VT Issue-Question Product-Conhost
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#10983
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?