starred/terminal
1
0
Fork 0
mirror of https://github.com/microsoft/terminal.git synced 2026-02-03 21:25:34 +00:00
Code Issues 19.0k Packages Projects Releases 20 Wiki Activity

Repeated use of OSC4 causes output corruption #15099

New Issue
Closed
opened 2026-01-31 04:28:32 +00:00 by claunia · 8 comments
claunia commented 2026-01-31 04:28:32 +00:00
Owner
Copy Link

Originally created by @vefatica on GitHub (Sep 5, 2021).

Windows Terminal version (or Windows build number)

Microsoft Windows 10 Pro for Workstations 10.0.19043.1165 (2009, 21H1)

Other Software

Batch files in TCC.EXE

Steps to reproduce

I wrote an app to reset the 16-color console palette reproduce any of WT's nine "schemes". It constructs the OSC4 sequence without ESCs, outputs it (error checking), and issues it with the ESCs added. At it's heart is this (full CPP attached).

	for ( INT i=0; i<16; i++ ) // Does conhost have colors 16 and 17?
	{
		// they're backwards
		r = GetBValue(cr[index][i]);
		g = GetGValue(cr[index][i]);
		b = GetRValue(cr[index][i]);
		wsprintfW(seq, L"]4;%d;rgb:%02X/%02X/%02X", i, r, g, b);
		WriteConsoleW(hOut, seq, lstrlen(seq), &dwWritten, nullptr);
		WriteConsoleW(hOut, L"\r\n", lstrlen(L"\r\n"), &dwWritten, nullptr);
		wsprintfW(seq2, L"%c%s%c", L'\x01b', seq, L'\x01b');
		WriteConsoleW(hOut, seq2, lstrlen(seq2), &dwWritten, nullptr);
	}

The program works, but I see corrupted output. Here's the result when run in the general audience CMD/CONHOST (scheme 8 is "Vintage", my default for consoles). Notice that after OSC4 has been issued once, the subsequent test output is corrupt (missing the leading ']'). Aside from that, the app works.

D:\Projects2019\osc4test\x64\Release> osc4test.exe 8
]4;0;rgb:00/00/00
4;1;rgb:80/00/00
4;2;rgb:00/80/00
4;3;rgb:80/80/00
4;4;rgb:00/00/80
4;5;rgb:80/00/80
4;6;rgb:00/80/80
4;7;rgb:C0/C0/C0
4;8;rgb:80/80/80
4;9;rgb:FF/00/00
4;10;rgb:00/FF/00
4;11;rgb:FF/FF/00
4;12;rgb:00/00/FF
4;13;rgb:FF/00/FF
4;14;rgb:00/FF/FF
4;15;rgb:FF/FF/FF

Exactly the same thing happens if I do it in a batch file with this loop at its heart.

:: envvar x is scheme index; i is color index
:: entries in array a look like RRGGBB (hex)
do i=0 to 15
	set r=%@instr[0,2,%a[%x,%i]]
	set g=%@instr[2,2,%a[%x,%i]]
	set b=%@instr[4,2,%a[%x,%i]]
	set seq=]4;%i;rgb:%r/%g/%b
	echo %seq
	echos ^e%seq^e
enddo

Things get worse if I run the EXE in a current OpenConsole.exe (stable or preview) and CMD. All output after the first line is lost. Likewise for the batch file with TCC.EXE running in OpenConsole.exe.

p:\osc4test\x64\Release> osc4test.exe 8
]4;0;rgb:00/00/00

p:\osc4test\x64\Release>

osc4test.src.zip

Expected Behavior

No corruption.

Actual Behavior

As described above.

Originally created by @vefatica on GitHub (Sep 5, 2021). ### Windows Terminal version (or Windows build number) Microsoft Windows 10 Pro for Workstations 10.0.19043.1165 (2009, 21H1) ### Other Software Batch files in TCC.EXE ### Steps to reproduce I wrote an app to reset the 16-color console palette reproduce any of WT's nine "schemes". It constructs the OSC4 sequence without ESCs, outputs it (error checking), and issues it with the ESCs added. At it's heart is this (full CPP attached). ``` for ( INT i=0; i<16; i++ ) // Does conhost have colors 16 and 17? { // they're backwards r = GetBValue(cr[index][i]); g = GetGValue(cr[index][i]); b = GetRValue(cr[index][i]); wsprintfW(seq, L"]4;%d;rgb:%02X/%02X/%02X", i, r, g, b); WriteConsoleW(hOut, seq, lstrlen(seq), &dwWritten, nullptr); WriteConsoleW(hOut, L"\r\n", lstrlen(L"\r\n"), &dwWritten, nullptr); wsprintfW(seq2, L"%c%s%c", L'\x01b', seq, L'\x01b'); WriteConsoleW(hOut, seq2, lstrlen(seq2), &dwWritten, nullptr); } ``` The program works, but I see corrupted output. Here's the result when run in the general audience CMD/CONHOST (scheme 8 is "Vintage", my default for consoles). Notice that after OSC4 has been issued once, the subsequent test output is corrupt (missing the leading ']'). Aside from that, the app works. ``` D:\Projects2019\osc4test\x64\Release> osc4test.exe 8 ]4;0;rgb:00/00/00 4;1;rgb:80/00/00 4;2;rgb:00/80/00 4;3;rgb:80/80/00 4;4;rgb:00/00/80 4;5;rgb:80/00/80 4;6;rgb:00/80/80 4;7;rgb:C0/C0/C0 4;8;rgb:80/80/80 4;9;rgb:FF/00/00 4;10;rgb:00/FF/00 4;11;rgb:FF/FF/00 4;12;rgb:00/00/FF 4;13;rgb:FF/00/FF 4;14;rgb:00/FF/FF 4;15;rgb:FF/FF/FF ``` Exactly the same thing happens if I do it in a batch file with this loop at its heart. ``` :: envvar x is scheme index; i is color index :: entries in array a look like RRGGBB (hex) do i=0 to 15 set r=%@instr[0,2,%a[%x,%i]] set g=%@instr[2,2,%a[%x,%i]] set b=%@instr[4,2,%a[%x,%i]] set seq=]4;%i;rgb:%r/%g/%b echo %seq echos ^e%seq^e enddo ``` Things get worse if I run the EXE in a current OpenConsole.exe (stable or preview) and CMD. All output after the first line is lost. Likewise for the batch file with TCC.EXE running in OpenConsole.exe. ``` p:\osc4test\x64\Release> osc4test.exe 8 ]4;0;rgb:00/00/00 p:\osc4test\x64\Release> ``` [osc4test.src.zip](https://github.com/microsoft/terminal/files/7112030/osc4test.src.zip) ### Expected Behavior No corruption. ### Actual Behavior As described above.
claunia added the Resolution-Fix-AvailableNeeds-TriageIssue-Docs labels 2026-01-31 04:28:32 +00:00
claunia commented 2026-01-31 04:28:33 +00:00
Author
Owner
Copy Link

@j4james commented on GitHub (Sep 6, 2021):

The OSC sequence needs to be terminated with "\x1b\\" (i.e. escape + backslash).

The behaviour you're seeing in OpenConsole is correct. When the escape isn't followed by a backslash, it's interpreted as the start of a new escape sequence. And since you then follow that with the introducer for an OSC, that text is also going to be interpreted as an OSC sequence and consumed up to the next escape. And this process just keeps repeating itself until all of the output is consumed.

The error handling in the older versions of conhost wasn't quite correct, so you're seeing a different result there, but if you terminated your OSC sequence correctly, it should work in both versions.

@j4james commented on GitHub (Sep 6, 2021): The `OSC` sequence needs to be terminated with `"\x1b\\"` (i.e. escape + backslash). The behaviour you're seeing in OpenConsole is correct. When the escape isn't followed by a backslash, it's interpreted as the start of a new escape sequence. And since you then follow that with the introducer for an `OSC`, that text is also going to be interpreted as an `OSC` sequence and consumed up to the next escape. And this process just keeps repeating itself until all of the output is consumed. The error handling in the older versions of conhost wasn't quite correct, so you're seeing a different result there, but if you terminated your `OSC` sequence correctly, it should work in both versions.
claunia commented 2026-01-31 04:28:34 +00:00
Author
Owner
Copy Link

@vefatica commented on GitHub (Sep 6, 2021):

Thanks, @j4james. Maybe that belongs in https://docs.microsoft.com/en-us/windows/console/console-virtual-terminal-sequences (where I get my info). That doc doesn't show a trailing backslash. Or is there something more general that I should have known about?

@vefatica commented on GitHub (Sep 6, 2021): Thanks, @j4james. Maybe that belongs in [https://docs.microsoft.com/en-us/windows/console/console-virtual-terminal-sequences](https://docs.microsoft.com/en-us/windows/console/console-virtual-terminal-sequences) (where I get my info). That doc doesn't show a trailing backslash. Or is there something more general that I should have known about?
claunia commented 2026-01-31 04:28:34 +00:00
Author
Owner
Copy Link

@j4james commented on GitHub (Sep 6, 2021):

OK, I never noticed that before. That's definitely a bug in the docs. If you're feeling enthusiastic, you can file a report on the documentation issue tracker - there's a link to it at the bottom of the page.

And for future reference, you're probably better off using the Xterm control sequence documentation here:
https://invisible-island.net/xterm/ctlseqs/ctlseqs.html

We don't support everything that Xterm does, but you can see a list of the sequences we do support here:
https://terminalnuget.blob.core.windows.net/packages/TerminalSequences.html

I'm not sure that's completely up to date either, but it'll be better than the microsoft.com documentation.

@j4james commented on GitHub (Sep 6, 2021): OK, I never noticed that before. That's definitely a bug in the docs. If you're feeling enthusiastic, you can file a report on the documentation issue tracker - there's a link to it at the bottom of the page. And for future reference, you're probably better off using the Xterm control sequence documentation here: https://invisible-island.net/xterm/ctlseqs/ctlseqs.html We don't support everything that Xterm does, but you can see a list of the sequences we do support here: https://terminalnuget.blob.core.windows.net/packages/TerminalSequences.html I'm not sure that's completely up to date either, but it'll be better than the microsoft.com documentation.
claunia commented 2026-01-31 04:28:35 +00:00
Author
Owner
Copy Link

@vefatica commented on GitHub (Sep 6, 2021):

Thanks again, @j4james. I filed a report.

Curious ... do you know why ESC \ was chosen as the terminator instead of, say, BEL or ST (which seem to be more normal)?

@vefatica commented on GitHub (Sep 6, 2021): Thanks again, @j4james. I filed a report. Curious ... do you know why `ESC \` was chosen as the terminator instead of, say, BEL or ST (which seem to be more normal)?
claunia commented 2026-01-31 04:28:35 +00:00
Author
Owner
Copy Link

@j4james commented on GitHub (Sep 6, 2021):

ESC \ is actually ST! It's just the 7-bit equivalent of the 8-bit C1 control (see here). You can also use the C1 control itself, but in UTF-8 it's not saving you anything, and probably isn't as widely supported (we only added support for it recently I believe). And BEL is a non-standard string terminator, so isn't recommended, but we do support that too (only with OSC controls though).

@j4james commented on GitHub (Sep 6, 2021): `ESC \` is actually `ST`! It's just the 7-bit equivalent of the 8-bit C1 control (see [here](https://invisible-island.net/xterm/ctlseqs/ctlseqs.html#h2-C1-_8-Bit_-Control-Characters)). You can also use the C1 control itself, but in UTF-8 it's not saving you anything, and probably isn't as widely supported (we only added support for it recently I believe). And `BEL` is a non-standard string terminator, so isn't recommended, but we do support that too (only with `OSC` controls though).
claunia commented 2026-01-31 04:28:36 +00:00
Author
Owner
Copy Link

@vefatica commented on GitHub (Sep 6, 2021):

Thanks for that link (7/8-bit C1). I'd seen stuff like that before but never really caught on. Now I get it ... 7-bit equivalents of 8-bit thingies.

@vefatica commented on GitHub (Sep 6, 2021): Thanks for that link (7/8-bit C1). I'd seen stuff like that before but never really caught on. Now I get it ... 7-bit equivalents of 8-bit thingies.
claunia commented 2026-01-31 04:28:36 +00:00
Author
Owner
Copy Link

@DHowett commented on GitHub (Sep 7, 2021):

(I believe that the bug on our documentation site is because the markdown parser consumed the \ and treated it as an escaping character 😁)

@DHowett commented on GitHub (Sep 7, 2021): (I believe that the bug on our documentation site is because the markdown parser consumed the `\` and treated it as an escaping character :grin:)
claunia commented 2026-01-31 04:28:37 +00:00
Author
Owner
Copy Link

@zadjii-msft commented on GitHub (Sep 18, 2021):

Looks like this was fixed in https://github.com/MicrosoftDocs/Console-Docs/pull/221

@zadjii-msft commented on GitHub (Sep 18, 2021): Looks like this was fixed in https://github.com/MicrosoftDocs/Console-Docs/pull/221
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-Docs Needs-Triage Resolution-Fix-Available
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#15099
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?