You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
It would to be nice add something like interactiveShell, which would spawn a long running full shell process with working stdin / stdout / stderr.
For a start this would allow us to turn one into successive calls of commands, which we feed the shell via the input stream, instead of relying on concatenating them via &&.
On the other hand, while never the original intent of this module, having a super simple interactiveShell() call to launch a full system shell sounds like a fun idea.
Short prototyping via osproc.startProcess wasn't very successful though. Calling any command will just cause the process to stop once the command ran (which makes sense). On the other hand, explicitly launching a shell as a process, e.g. startProcess("zsh", options = {...}) seems to make it impossible to access the input and output streams. The streams we have access to are probably the ones from the session that launches zsh, instead of the streams of the launched shell. That also makes sense.
However, how to work around this and achieve what I want, I don't know. I will have to do some research on that at some point. :)
The text was updated successfully, but these errors were encountered:
When I tried this for the first time, I simply didn't use the accessed input and output streams correctly. I have a working proof of concept for this lying around. Just need to fix it up and find a way to properly include it into the project.
It would to be nice add something like
interactiveShell
, which would spawn a long running full shell process with working stdin / stdout / stderr.For a start this would allow us to turn
one
into successive calls of commands, which we feed the shell via the input stream, instead of relying on concatenating them via&&
.On the other hand, while never the original intent of this module, having a super simple
interactiveShell()
call to launch a full system shell sounds like a fun idea.Short prototyping via
osproc.startProcess
wasn't very successful though. Calling any command will just cause the process to stop once the command ran (which makes sense). On the other hand, explicitly launching a shell as a process, e.g.startProcess("zsh", options = {...})
seems to make it impossible to access the input and output streams. The streams we have access to are probably the ones from the session that launcheszsh
, instead of the streams of the launched shell. That also makes sense.However, how to work around this and achieve what I want, I don't know. I will have to do some research on that at some point. :)
The text was updated successfully, but these errors were encountered: