starred/terminal
1
0
Fork 0
mirror of https://github.com/microsoft/terminal.git synced 2026-09-24 16:05:21 +00:00
Code Issues 19.0k Packages Projects Releases 20 Wiki Activity

A question about emulating openpty with the new CreatePseudoConsole API #455

New Issue
Closed
opened 2026-01-30 21:52:45 +00:00 by claunia · 8 comments
claunia commented 2026-01-30 21:52:45 +00:00
Owner
Copy Link

Originally created by @sad75 on GitHub (Nov 25, 2018).

Hello,

I wanted to use the new CreatePseudoConsole() API to implement Pythonos.openpty() in order to ease the use of Visual Studio Python environment.
In Python, the call just returns the master and slave file descriptors.
However I'm not able to find out a way to map the HANDLEs to the master/slave file descriptors.
My understanding is that the new API uses unidirectional pipe HANDLE and that the file descriptors are bidirectional.

Do you have already an example of an emulation of Linux openpty() on Windows using the new API?

Thank You

Originally created by @sad75 on GitHub (Nov 25, 2018). Hello, I wanted to use the new` CreatePseudoConsole()` API to implement Python`os.openpty()` in order to ease the use of Visual Studio Python environment. In Python, the call just returns the master and slave file descriptors. However I'm not able to find out a way to map the HANDLEs to the master/slave file descriptors. My understanding is that the new API uses unidirectional pipe HANDLE and that the file descriptors are bidirectional. Do you have already an example of an emulation of Linux `openpty()` on Windows using the new API? Thank You
claunia added the Issue-QuestionResolution-AnsweredArea-ServerProduct-Conpty labels 2026-01-30 21:52:45 +00:00
claunia commented 2026-01-30 21:52:46 +00:00
Author
Owner
Copy Link

@Ivan171 commented on GitHub (Nov 26, 2018):

Have you tried _open_osfhandle to a get file descriptor from the HANDLE?

@Ivan171 commented on GitHub (Nov 26, 2018): Have you tried [_open_osfhandle](https://docs.microsoft.com/en-us/cpp/c-runtime-library/reference/open-osfhandle?view=vs-2017l) to a get file descriptor from the HANDLE?
claunia commented 2026-01-30 21:52:46 +00:00
Author
Owner
Copy Link

@sad75 commented on GitHub (Nov 26, 2018):

Yes, this is what I planned to do.

However in the example below, the API is associate to 4 PIPE HANDLEs, 2 for the slave part and 2 for the MASTER part.

HRESULT SetUpPseudoConsole(COORD size)
{
    HRESULT hr = S_OK;

    // Create communication channels

    // - Close these after CreateProcess of child application with pseudoconsole object.
    HANDLE inputReadSide, outputWriteSide; 

    // - Hold onto these and use them for communication with the child through the pseudoconsole. 
    HANDLE outputReadSide, inputWriteSide; 

    if (!CreatePipe(&inputReadSide, &inputWriteSide, NULL, 0))
    {
        return HRESULT_FROM_WIN32(GetLastError());
    }

    if (!CreatePipe(&outputReadSide, &outputWriteSide, NULL, 0))
    {
        return HRESULT_FROM_WIN32(GetLastError())
    }

    hr = CreatePseudoConsole(size, inputReadSide, outputWriteSide, 0, &hPC);
    if (FAILED(hr))
    {
        return hr;
    }

    // ... 

}

The openpty API returns only 2 file descriptors, but they are read/write (bidirectional).

@sad75 commented on GitHub (Nov 26, 2018): Yes, this is what I planned to do. However in the example below, the API is associate to 4 PIPE HANDLEs, 2 for the slave part and 2 for the MASTER part. ``` HRESULT SetUpPseudoConsole(COORD size) { HRESULT hr = S_OK; // Create communication channels // - Close these after CreateProcess of child application with pseudoconsole object. HANDLE inputReadSide, outputWriteSide; // - Hold onto these and use them for communication with the child through the pseudoconsole. HANDLE outputReadSide, inputWriteSide; if (!CreatePipe(&inputReadSide, &inputWriteSide, NULL, 0)) { return HRESULT_FROM_WIN32(GetLastError()); } if (!CreatePipe(&outputReadSide, &outputWriteSide, NULL, 0)) { return HRESULT_FROM_WIN32(GetLastError()) } hr = CreatePseudoConsole(size, inputReadSide, outputWriteSide, 0, &hPC); if (FAILED(hr)) { return hr; } // ... } ``` The `openpty `API returns only 2 file descriptors, but they are read/write (bidirectional).
claunia commented 2026-01-30 21:52:47 +00:00
Author
Owner
Copy Link

@zadjii-msft commented on GitHub (Nov 26, 2018):

@sad75 I'd take a look at this sample code:

https://github.com/Microsoft/console/blob/master/samples/ConPTY/EchoCon/EchoCon/EchoCon.cpp#L91-L122

That function is going to:

  1. Create a pair of anonymous pipes for use with the pseudoconsole
  2. Get the size of the console window the app is currently running in. That sample is a console app, so it's already running in the console, but for your use case you can probably omit this, and pass the size you're expecting the pseudoconsole to be into that function.
  3. It then creates the conpty using the pipes from before.
  4. It then closes our handles to the master side of the conpty. The Pseudoconsole has already duped those handles into it's process space, so we need to make sure to close them in our process to correctly refcount the pipes. We won't need them anymore.

When the function returns, the parameters will be filled with the values you need:

  1. the HPCON can be used with other Pseudoconsole functions to manipulate it. You'll need that to:
  • attach child processes (see this sample)
  • resize the pseudoconsole
  • close the pseudoconsole when you're done with it.
  1. phPipeIn and phPipeOut can be used for reading the output from and writing input to the conpty. It's true that the HANDLEs returned by this API are not bi-directional, though managing a pair of unidirectional pipes is trivially harder than a single bidirectional pipe.

I agree that having a sample openconpty (and forkconpty) function might be helpful, especially for people who are solely familiar with linux. Obviously the signatures wouldn't be exactly the same, but if anyone wants to throw a sample together, I'd happily review the PR :)

@zadjii-msft commented on GitHub (Nov 26, 2018): @sad75 I'd take a look at this sample code: https://github.com/Microsoft/console/blob/master/samples/ConPTY/EchoCon/EchoCon/EchoCon.cpp#L91-L122 That function is going to: 1. Create a pair of anonymous pipes for use with the pseudoconsole 2. Get the size of the console window the app is currently running in. That sample is a console app, so it's already running in the console, but for your use case you can probably omit this, and pass the size you're expecting the pseudoconsole to be into that function. 3. It then creates the conpty using the pipes from before. 4. It then closes our handles to the master side of the conpty. The Pseudoconsole has already duped those handles into it's process space, so we need to make sure to close them in our process to correctly refcount the pipes. We won't need them anymore. When the function returns, the parameters will be filled with the values you need: 1. the HPCON can be used with other Pseudoconsole functions to manipulate it. You'll need that to: * attach child processes (see [this sample](https://github.com/Microsoft/console/blob/master/samples/ConPTY/EchoCon/EchoCon/EchoCon.cpp#L126-L165)) * resize the pseudoconsole * close the pseudoconsole when you're done with it. 2. `phPipeIn` and `phPipeOut` can be used for reading the output from and writing input to the conpty. It's true that the HANDLEs returned by this API are not bi-directional, though managing a pair of unidirectional pipes is trivially harder than a single bidirectional pipe. I agree that having a sample `openconpty` (and `forkconpty`) function might be helpful, especially for people who are solely familiar with linux. Obviously the signatures wouldn't be *exactly* the same, but if anyone wants to throw a sample together, I'd happily review the PR :)
claunia commented 2026-01-30 21:52:47 +00:00
Author
Owner
Copy Link

@miniksa commented on GitHub (May 18, 2019):

Looks like question answered to me. Closing.

@miniksa commented on GitHub (May 18, 2019): Looks like question answered to me. Closing.
claunia commented 2026-01-30 21:52:47 +00:00
Author
Owner
Copy Link

@marcomorain commented on GitHub (Aug 26, 2019):

@sad75 did you manage to get this to work in the end? I'm having the same problem trying to make the Windows PTY match the Linux API for a golang project. The existing code that I'm porting to Windows expects a single file handle where it can both read and write to the same file.

What did you see?

@marcomorain commented on GitHub (Aug 26, 2019): @sad75 did you manage to get this to work in the end? I'm having the same problem trying to make the Windows PTY match the Linux API for a `golang` project. The existing code that I'm porting to Windows expects a single file handle where it can both read and write to the same file. ![What did you see?](https://imgs.xkcd.com/comics/wisdom_of_the_ancients.png)
claunia commented 2026-01-30 21:52:47 +00:00
Author
Owner
Copy Link

@sad75 commented on GitHub (Oct 2, 2019):

@marcomorain
Sorry for my late answer. No, I haven't find a solution.

@sad75 commented on GitHub (Oct 2, 2019): @marcomorain Sorry for my late answer. No, I haven't find a solution.
claunia commented 2026-01-30 21:52:47 +00:00
Author
Owner
Copy Link

@Kreijstal commented on GitHub (Jul 26, 2024):

@marcomorain did you find a solution?

@Kreijstal commented on GitHub (Jul 26, 2024): @marcomorain did you find a solution?
claunia commented 2026-01-30 21:52:48 +00:00
Author
Owner
Copy Link

@fdmarc commented on GitHub (Jul 26, 2024):

No I never did, and I've since changed company, so I can no longer check how my previous company have solved this problem.

@fdmarc commented on GitHub (Jul 26, 2024): No I never did, and I've since changed company, so I can no longer check how my previous company have solved this problem.
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-Server Issue-Question Product-Conpty Resolution-Answered
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#455
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?