Martijn Dekker

*edited my comment because I misread the original question.* I would say no, definitely not. ksh93 (as I just discovered thanks to your question) is the only shell where `test...

No, your understanding is incorrect. The `(...)` notation always creates a subshell, but ksh93 by default implements subshells without forking a new process. Your example `echo $$; ( echo $$...

@stephane-chazelas gave an essential hint in a comment on #480: running the `ulimit` builtin in a subshell forces it to fork off into a separate process right then and there....

Hi @cassioneri , thank you for that report. This AT&T repo is no longer active, but the (to the best of my knowledge) only actively and openly developed fork of...

Now that I'm here, I'd like to give belated thanks to @kernigh for the [parser fix](#issuecomment-384840196) and to @stephane-chazelas for the [reproducers](#issuecomment-384178095), both of which I've incorporated in what is,...

For the record, this bug is now fixed in [ksh 93u+m](/ksh93/ksh). See https://github.com/ksh93/ksh/commit/acf84e9633f03b10da7d674773ca831dc4b9d88e.

What makes you think it's not handling any errors? Not handling any errors is precisely what it did before this commit. An infinite retry loop was the result. > Are...

A fix for this issue is now in my ksh 93u+m bugfix branch. For a full explanation, expand the commit referenced immediately above this comment (ffe5df3).

Hi, I'm not sure why you bothered to post a PR here as this repo is inactive and AT&T will ignore it, but I'm glad you did, because the libast...

For anyone else reading: This is fixed on the [ksh 93u+m](https://github.com/ksh93/ksh) fork.