On 7/31/13 8:16 AM, David Korn wrote:
cc: ast-users@lists.research.att.com
Subject: Re: Problem in ksh93 93t+ (my reference 46004)
AIX 6.1 (ddaap128/ Version M-12/28/93e):
6100-06-06-1140
hdisk0 root@wlc:/ set | grep -i vers
.sh.version='Version M-12/28/93e'
$ ksh93
$
On 8/2/13 1:24 PM, Roland Mainz wrote:
On Fri, Aug 2, 2013 at 8:12 PM, Terrence J. Doyle
terrence.j.doyle+...@gmail.com wrote:
That explains the format change. While this seems a bit inconsistent
with the output of ${!array[@]}, I'm not going to quibble about format.
However
I my last message I said I had a problem building
ast-open-2014-09-29 with debug=1. So, here it is. When the build reached
the programs dlls and pax, the following error message appeared:
Undefined symbols:
__dll_state, referenced from:
__dll_state$non_lazy_ptr in
While addressing the problem in my last message I took a closer look
at the cc.darwin scripts in src/cmd/INIT. Here's some suggestions for
improvements:
* With ast-open-2014-09-29 you upped MACOSX_DEPLOYMENT_TARGET from
10.3 to 10.7 in cc.darwin and cc.darwin.i386. For cc.darwin in
I was investigating my options for controlling the new built-ins
with ksh-20120929, and I happened to notice some garbled output for the
-n option of the builtin command:
$ builtin --?
Usage: builtin [ options ] [pathname ...]
OPTIONS
-d Deletes each of the specified
I tried using the whence command for some of the new built-ins in
ksh-20140929 and was quite confused by what I saw as the following
illustrates:
$ whence grep
/usr/bin/grep
$ : What? I thought grep was built in.
$ whence -v grep
grep is a shell builtin version of /usr/bin/grep
$ : OK, so I
I recently happened to invoke the restricted shell for ksh-20140929
and an error message immediately appeared. I didn't remember this
happening with prior versions of ksh, so I tried the older versions that
I have and found this:
$ rksh +E
$ rksh[1]: /dev/null: restricted
print
Here's another bit of strange behavior I've known about since
ksh-20120612 but was hoping would be fixed in ksh-20140929. Most of the
time, when I try to execute a file that doesn't exist I get the expected
error message: ... not found [No such file or directory]. However, as
the following
File name completion still has problems when there's only one file
in a directory. If it makes any difference I have FCEDIT=vi. The
misbehavior is hard to explain, but here goes. Let's use /System on Mac
OS 10.5:
$ ls /System
Library
If I go again and hit tab here:
$ ls /System/
No, this is not another bug report.
I've been experimenting with the discipline function PS1.get. It has
the advantage that it can place more dynamic information in the prompt
than the variable. For instance, it's possible to get the current load
average in the prompt with PS1.get.
I've been considering the impact of the new built-in commands in
ksh-2014-09-29 on restricted mode. Many of the new built-ins circumvent
the restrictions imposed by restricted mode. So, I started rksh with
PATH=/rbin, and to my surprise none of the new built-ins were
accessible. So far, so
The ksh(1) man page states:
Commands.
A simple-command is... . The value of a simple-command is its
exit status; 0-255 if it terminates normally; 256+signum
if it terminates abnormally... .
However, when I do the following:
$ read
^C
$ echo $?
2
$
I
I just saw the new -a option to the trap command, and I thought I'd try
it. Trap --? states:
-a append the current trap setting to the
specified action.
However, instead of being appended the added command was prefixed as
shown below:
$ trap
I've been doing some work with compound arrays lately, and I've run
into some confusing issues. The command sequence below illustrates:
$ # Given the following:
$
$ typeset -a array1=(a b c)
$ typeset -a array2=((a b c) (d e f) (g h i))
$ typeset -A array3=([a]=(1 2 3) [b]=(4 5 6) [c]=(7 8
Attempts to send SIGQUIT to a process by using kill -QUIT are now
considered an error and elicit the kill usage message. This is
apparently because of the new -Q option for built-in kill. The
documentation for the built-in kill doesn't describe the legacy -signame
options, but kill -QUIT
There are two points, here.
First, you're using an old version of ksh.
Second, ksh has a different spin from bash on reading fixed-length
data. Read sees variables in two flavors, text and binary. When read
puts characters into a textual variable newlines are always
I've been finding so many bugs in Version ABIJM 93v- 2014-12-24 that I
don't see how it can be deemed release-worthy.
We need more people in the community to download the beta, test it,
make bug reports and even provide patches. Feel free to join in.
I'm trying to get
17 matches
Mail list logo
