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

ENABLE_VIRTUAL_TERMINAL_INPUT and child processes #20273

New Issue
Closed
opened 2026-01-31 07:08:43 +00:00 by claunia · 6 comments
claunia commented 2026-01-31 07:08:43 +00:00
Owner
Copy Link

Originally created by @amoldeshpande on GitHub (Jul 21, 2023).

Windows Terminal version

1.17.11461.0

Windows build number

10.0.22621.1992]

Other Software

windows tcsh, setting the following:

SetConsoleMode(ghstdin, dwmode | ENABLE_VIRTUAL_TERMINAL_INPUT | ENABLE_WINDOW_INPUT))

Steps to reproduce

This is not directly related to running in Windows Terminal, but IIRC you own the conhost as well, so I hope this question/bug belongs here.

I have a Windows native version of tcsh that sets ENABLE_VIRTUAL_TERMINAL_INPUT to handle escape codes and such.

What I find is that when you use ReadConsoleInput on a console handle with this setting, the Virtual Key code for ESCAPE is not set (i.e., INPUT_RECORD.KeyEvent.wVirtualKeyCode is 0, not 0x1b)

The uChar.AsciiChar is set to 0x1b.

Now, this is not a problem for this shell as it is not looking for VK_xxx, but just looking at ascii chars.

However, this setting seems to be inherited by the child process, so a console application that is too old to know (or does not care) about this behavior will get shafted if it expects VK_xyz to be populated.

In my specific case, the child process is a very old hex editor that seems (by looking at disassembly, since I don't have source) to be expecting a non-zero wVirtualKeyCode (or scan code, not quite sure).

I can work around this issue by disabling virtual key processing every time I call createprocess and then re-enabling it after the call.

The issue/bug is whether this behavior is correct.

Expected Behavior

Child process should be unaware of parent's VT behavior

Actual Behavior

as described above.

Originally created by @amoldeshpande on GitHub (Jul 21, 2023). ### Windows Terminal version 1.17.11461.0 ### Windows build number 10.0.22621.1992] ### Other Software windows tcsh, setting the following: SetConsoleMode(ghstdin, dwmode | ENABLE_VIRTUAL_TERMINAL_INPUT | ENABLE_WINDOW_INPUT)) ### Steps to reproduce This is not directly related to running in Windows Terminal, but IIRC you own the conhost as well, so I hope this question/bug belongs here. I have a Windows native version of tcsh that sets ENABLE_VIRTUAL_TERMINAL_INPUT to handle escape codes and such. What I find is that when you use ReadConsoleInput on a console handle with this setting, the Virtual Key code for ESCAPE is not set (i.e., INPUT_RECORD.KeyEvent.wVirtualKeyCode is 0, not 0x1b) The uChar.AsciiChar is set to 0x1b. Now, this is not a problem for this shell as it is not looking for VK_xxx, but just looking at ascii chars. However, this setting seems to be inherited by the child process, so a console application that is too old to know (or does not care) about this behavior will get shafted if it expects VK_xyz to be populated. In my specific case, the child process is a very old hex editor that seems (by looking at disassembly, since I don't have source) to be expecting a non-zero wVirtualKeyCode (or scan code, not quite sure). I can work around this issue by disabling virtual key processing every time I call createprocess and then re-enabling it after the call. The issue/bug is whether this behavior is correct. ### Expected Behavior Child process should be unaware of parent's VT behavior ### Actual Behavior as described above.
claunia added the Issue-BugResolution-Duplicate labels 2026-01-31 07:08:43 +00:00
claunia commented 2026-01-31 07:08:44 +00:00
Author
Owner
Copy Link

@DHowett commented on GitHub (Jul 21, 2023):

Hey, thanks for the report! You're in the right place for conhost as well, since conhost is built out of this repo!

You're totally right when you say that child processes should be unaware of their parents' behavior (and vice-versa)... but unfortunately, that's the way the console API was designed ages upon ages ago.
I have some significant worries about changing that now, but I'm happy to say that we're tracking making the input modes per-process (and solving the overall story) in #4954. We can chat about what it'll mean to change this over there.

In general, shells on Windows have had to sanitize the console modes before they spawn applications. CMD uses VT output but disables it before spawning an application for this exact reason. PowerShell is in the middle, and it changes the modes all the time (which is a big surprise in all of my trace logging!).

If you don't mind, I'm going to mark this as a /duplicate of #4954. Let me know if you disagree!

@DHowett commented on GitHub (Jul 21, 2023): Hey, thanks for the report! You're in the right place for conhost as well, since conhost is built out of this repo! You're totally right when you say that child processes should be unaware of their parents' behavior (and vice-versa)... but unfortunately, that's the way the console API was designed ages upon ages ago. I have some significant worries about changing that now, _but_ I'm happy to say that we're tracking making the input modes per-process (and solving the overall story) in #4954. We can chat about what it'll mean to change this over there. In general, shells on Windows have had to sanitize the console modes before they spawn applications. CMD uses VT _output_ but disables it before spawning an application for this exact reason. PowerShell is in the middle, and it changes the modes _all the time_ (which is a big surprise in all of my trace logging!). If you don't mind, I'm going to mark this as a /duplicate of #4954. Let me know if you disagree!
claunia commented 2026-01-31 07:08:44 +00:00
Author
Owner
Copy Link

@microsoft-github-policy-service[bot] commented on GitHub (Jul 21, 2023):

Hi! We've identified this issue as a duplicate of another one that already exists on this Issue Tracker. This specific instance is being closed in favor of tracking the concern over on the referenced thread. Thanks for your report!

@microsoft-github-policy-service[bot] commented on GitHub (Jul 21, 2023): Hi! We've identified this issue as a duplicate of another one that already exists on this Issue Tracker. This specific instance is being closed in favor of tracking the concern over on the referenced thread. Thanks for your report!
claunia commented 2026-01-31 07:08:45 +00:00
Author
Owner
Copy Link

@amoldeshpande commented on GitHub (Jul 22, 2023):

Can you explain why the KEY_EVENT_RECORD does not have the virtual key code in this mode?
It seems like a VT app could ignore the key code while old apps could continue to use it, but I'm probably missing some nuance.

@amoldeshpande commented on GitHub (Jul 22, 2023): Can you explain why the KEY_EVENT_RECORD does not have the virtual key code in this mode? It seems like a VT app could ignore the key code while old apps could continue to use it, but I'm probably missing some nuance.
claunia commented 2026-01-31 07:08:45 +00:00
Author
Owner
Copy Link

@DHowett commented on GitHub (Jul 22, 2023):

You know, that's an excellent question! I'll have to dig a bit to figure it out.

I think there's some compatibility behavior here. Perhaps that applications that are expecting VKs won't do anything with the "synthetic" escape key from the control sequence if it's zero?

@DHowett commented on GitHub (Jul 22, 2023): You know, that's an excellent question! I'll have to dig a bit to figure it out. I think there's some compatibility behavior here. Perhaps that applications that are expecting VKs won't do anything with the "synthetic" escape key from the control sequence if it's zero?
claunia commented 2026-01-31 07:08:45 +00:00
Author
Owner
Copy Link

@amoldeshpande commented on GitHub (Jul 22, 2023):

hmm, I changed my code so I can't look at the behavior right now. But if I recall correctly, pressing ESC only generated the single key code, not a sequence. So it should ideally have just worked if the VK code was set.

Anyway, I would love to know if you ever discover the reason. thanks!

@amoldeshpande commented on GitHub (Jul 22, 2023): hmm, I changed my code so I can't look at the behavior right now. But if I recall correctly, pressing ESC only generated the single key code, not a sequence. So it should ideally have just worked if the VK code was set. Anyway, I would love to know if you ever discover the reason. thanks!
claunia commented 2026-01-31 07:08:45 +00:00
Author
Owner
Copy Link

@amoldeshpande commented on GitHub (Jul 22, 2023):

The more I think about it, the more it seems like ESC should be passed through with Virtual Keycode as before. If it's part of a multi-byte escape sequence, the application should be able to handle that.

@amoldeshpande commented on GitHub (Jul 22, 2023): The more I think about it, the more it seems like ESC should be passed through with Virtual Keycode as before. If it's part of a multi-byte escape sequence, the application should be able to handle that.
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 Issue-Bug Resolution-Duplicate
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#20273
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?