Gerard Lally dixit:
>I am slowly learning the shell at the moment, and I was using ast ksh93,
>but it doesn't compile on NetBSD 7 beta. I would be happy to use mksh if
Haha… I’ve not managed to compile it on MirBSD too, the one time
I tried, but that’s been years ago. And I have tried to look at
its source once, and didn’t understand a thing.
>the two books I bought (Bolsky and Korn, 2nd edition, and Rosenblatt and
>Robbins, 2nd edition) were 80% or 90% applicable. Bolsky and Korn is a
Uh. I only know the O’Reilly book. First edition applies 99% (no
DEBUG trap), second edition less.
>very good book: like a tutorial and a reference. I don't need very
>advanced features. What do you think? Is it enough to go through Bolsky
>and Korn's book, together with the mksh man page?
Probably. In general, we don’t have:
• associative, multidimensional arrays (yet, except with assockit)
• floating point
• discipline functions (very vague maybe)
• types (we could just use assoc arrays instead)
• ${.sh.*} variables
• C-style for loops
• <(…)
• printf (all Korn shells use print+typeset instead)
Indexed arrays are “set -A foo -- 1 2 3” (ksh88 style) or
“foo=(1 2 3)” (bash style), not “typeset -a foo=(1 2 3)”.
We have most of the other stuff, although it sometimes differs,
e.g. nameref is very slightly different due to the scoping models
used by the shells differing.
You can also always ask in IRC (channel #!/bin/mksh on Freenode),
but most times it’s enough to just try things out.
Good luck,
//mirabilos
--
"How can you ban language, words? How're words offensive? And why should I
have to tolerate YOUR interpretation? I'm the one using the word. ASK me how
I'm using it, don't TELL me. And if you don't like the way I'm using it, so
what? It's my right. It's my freedom of expression. Without that, we're
nothing but slaves." -- Johnny Rotten
