I. Szczesniak wrote:
> On 9/22/06, James Carlson <james.d.carlson at sun.com> wrote:
>> I. Szczesniak writes:
>> > On 9/22/06, Joseph Kowalski <Joseph.Kowalski at eng.sun.com> wrote:
>> Why wouldn't this scan a ksh93-specific location first, such as a
>> /usr/lib/ksh93/ directory? You could have your own libcmd.so{,.1}
>> buried in there, encourage others to deliver plugin libraries to that
>> separate directory, and there'd be no conflict.
>
> Because it is not compatible to existing ksh93 versions and would not
> be portable within scripts. If you ask me Sun should avoid introducing
> Solaris-specific modifications to the shell.
I don't see how this would break scripts. If it sounds did then you are
saying that there is no flexibility what so ever in where libcmd must be
located on ever single platform - and I'm sure you aren't implying that
because then we couldn't possibly be having the /lib vs /usr/lib
discussion in another thread.
Forgive my ignorance of how this ksh93 feature really works since I'm a
zsh (and have been since before ksh93 existed :-)).
If it is possible for the ksh93 libcmd to be in /lib or /usr/lib and for
it to have non ksh93 content in it (ie the def* functions) then I don't
understand why it can't look for it in /usr/lib/ksh. If I understand
things correctly it wouldn't break scripts because the scripts don't say
what the library search path is and it may not even need a code change
it might be a built time linker option.
--
Darren J Moffat
