Re: mksh import

[Alex Goncharov <alex-goncharov%comcast.net@localhost>]

First, I hope your surgery went well and the recovery is going well,
too!

,--- You/Dad - Kent (Fri, 7 Jan 2011 18:30:14 -0600) ----*
| Here is a link to a presentation given at NLUUG several
| years back:
| 
| https://docs.google.com/present/view?id=dqwmc5b_0fpv6h5

Thank you -- this is all extremely interesting.  I am tied up with
other duties till about next weekend, but I am hoping to read this all
carefully and experiment in not so remote a future.
 
| IMO, this is pretty much a no-brainer. With ksh93 performance,
| it's functionality, continued development and improvement, and 
| the fact it is becoming the defacto standard, I think ksh93 should 
| be the default / base shell in NetBSD.

The problem with ksh93 (and even more, with ksh93t+) is the
portability.  As you have seen from my experiments, e.g:

,--- I/Alex (Wed, 05 Jan 2011 12:35:08 -0500) ----*
| FreeBSD 8.2-PRERELEASE
| == /usr/local/bin/ksh93 ==>
| real  0m0.274s
| <== /usr/local/bin/ksh93 ===
|
   [ no ksh93 on Linux and SunOS ]
   
| ============================================================
| AIX 3 5
| == /bin/ksh93 ==>
| ./mandelbrot.sh[35]: let: {-20..20}: arithmetic syntax error
| <== /bin/ksh93 ===
`-------------------------------------------------*

So:

 a. It is less widely installed than bash and ksh (88).
 
 b. When installed, ksh93 behaves differently on various systems.

If the good (and current) ksh93(++) can be easily built on all the
major platforms, then it can be installed into a private location and
this is OK -- not much worse than the Bash's case; the activity of
ksh93 development and fixes need to be looked into, too...

But from what I learned from you, ksh93 is worth a further effort's
for me, thank you!

-- Alex -- alex-goncharov%comcast.net@localhost --

/*
 * I used to think I was indecisive, but now I'm not so sure.
 */