home.social

#pdksh — Public Fediverse posts

Live and recent posts from across the Fediverse tagged #pdksh, aggregated by home.social.

fetched live
  1. Think I've got enough #ksh implementations installed? XD

    rld@Intrepid:~$ pkg info |grep ksh
    ksh-1.0.10                     ksh93u+m is the renewed development of ksh93 based on AT&T ksh93u+m (stable)
    mksh-59c_3                     MirBSD Korn Shell
    oksh-7.8,1                     Portable OpenBSD Korn shell
    pdksh-5.2.14p2_7               The Public Domain Korn Shell
    rld@Intrepid:~$ 
    

    #mksh #pdksh #oksh #testing #unix #unixShell #ShellScripting

  2. Think I've got enough #ksh implementations installed? XD

    rld@Intrepid:~$ pkg info |grep ksh
    ksh-1.0.10                     ksh93u+m is the renewed development of ksh93 based on AT&T ksh93u+m (stable)
    mksh-59c_3                     MirBSD Korn Shell
    oksh-7.8,1                     Portable OpenBSD Korn shell
    pdksh-5.2.14p2_7               The Public Domain Korn Shell
    rld@Intrepid:~$ 
    

    #mksh #pdksh #oksh #testing #unix #unixShell #ShellScripting

  3. #Illumos has another #wordexp implementation (originating with Mortice-Kern, no less!) that expects a Korn (93) shell and uses its printf (and set -o), which might be tweakable to function with #OpenBSD's PD Korn shell; but like the GNU C library licence the Sun CDDL would likely be a problem.

    code.illumos.org/plugins/gitil

    So if one is looking for "easily", where one just imports a compatibly-licenced and #KornShell-compatible existing implementation, the answer seems to be "No.".

    #pdksh

  4. #Illumos has another #wordexp implementation (originating with Mortice-Kern, no less!) that expects a Korn (93) shell and uses its printf (and set -o), which might be tweakable to function with #OpenBSD's PD Korn shell; but like the GNU C library licence the Sun CDDL would likely be a problem.

    code.illumos.org/plugins/gitil

    So if one is looking for "easily", where one just imports a compatibly-licenced and #KornShell-compatible existing implementation, the answer seems to be "No.".

    #pdksh

  5. #Poll: Curious about people's attitudes towards shell scripting.

    Two part question:

    1. Are you a DEVeloper (or working in a development-heavy role), OTHER-IT worker (such as a sysadmin, architect, anything in a non-development-heavy role), or NON-IT (accountant, doctor, whatever)
    2. Do you HATE shell scripting, are you INDIFferent towards (or ignorant of) shell scripting, or do you LOVE it?

    #Unix #UnixShell #ShellScript #ShellScripting #POSIX #PosixShell #sh #bash #zsh #csh #tcsh #ksh #pdksh #oksh #mksh

  6. #Poll: Curious about people's attitudes towards shell scripting.

    Two part question:

    1. Are you a DEVeloper (or working in a development-heavy role), OTHER-IT worker (such as a sysadmin, architect, anything in a non-development-heavy role), or NON-IT (accountant, doctor, whatever)
    2. Do you HATE shell scripting, are you INDIFferent towards (or ignorant of) shell scripting, or do you LOVE it?

    #Unix #UnixShell #ShellScript #ShellScripting #POSIX #PosixShell #sh #bash #zsh #csh #tcsh #ksh #pdksh #oksh #mksh

  7. @pixelherodev
    I mean, it really depends. It's mostly an academic exercise.

    I have at&t ksh installed on my linux daily driver, I know it's 2-4x faster than bash, but all of my scripts are still in bash (even the ones that would be trivial to convert).

    20 years ago, the difference mattered a lot more to me than it does now.

    I guess I'm still just wondering why still uses pdksh (and why is still a thing). Probably just history. Already audited, don't mess with it.

  8. Just as I suspected, the raw loop speed on at&t is a lot faster than , even on .

    Not sure if that really matters, but I've always known "true" ksh to be a decent bit faster than bash or pdksh, for reasons unknown to me.