April Chin wrote:
> Thanks for the migration plan!
No problem...
> Regarding this part of the plan:
>
> > - Edit all existing scripts in the Solaris code base which use
> > /usr/bin/ksh to use /usr/bin/oksh instead until they have been tested
> > with ksh93.
> > A test in 2001 revealed that from 598 ksh scripts in a Solaris 8
> > installation 29 needed changes (unfortunately very prominent stuff like
> > "patchadd" / "patchrm" are on that list... ;-/ ) - that's AFAIK less
> > than 5%... :-)
>
> Could you please send me a list of which scripts you found which needed
> changes? This would be helpful, though I will still have to do
> a current analysis of the latest Solaris scripts.
Yes, I can do that...
> Some comments on this portion:
>
> > - Put ksh93 into /usr/bin/nksh ("new" korn shell. ksh93 MUST NOT be
> > placed to /usr/bin/ksh93 - arguments can be listed here but that is
> > AFAIK outside the scope of this mail))
> > An alternative may be to just stuff ksh93 to /usr/xpg4/bin/sh (instead
> > of creating /usr/bin/nksh (on the other side the creation of
> > /usr/bin/nksh may be better... it indicates that there is a "new" korn
> > shell... :-) ))
> >
> > - /usr/bin/ksh then gets a link (hard-/softlink - whatever you prefer)
> > to the following positions:
> > Solaris 9: /usr/bin/ksh --> /usr/bin/oksh
> > Solaris 10: /usr/bin/ksh --> /usr/bin/oksh
> > OpenSolaris: /usr/bin/ksh --> /usr/bin/nksh
> > Solaris 11: /usr/bin/ksh --> /usr/bin/nksh
>
> I know there is a lot of anxiety to get ksh93 integrated as /usr/bin/ksh
> very soon, but I don't believe we should initially change /usr/bin/ksh to
> ksh93.
> This will generate serious complaints from existing users,
I agree (for the Solaris case). It's hard to miss their wrath... ;-(
> and there
> are features in the existing /usr/bin/ksh which are needed before
> we can replace it, including internationalization with the proper hooks
> for all the Sun-supported locales and multibyte character handling.
Is there a way to get the testcases/testsuite you were using ?
> I would like to advocate a gentler approach, for which we first integrate
> ksh93 as /usr/bin/nksh and keep /usr/bin/ksh linked to oksh (ksh88).
> However, we may be able to initially replace /usr/bin/ksh in OpenSolaris
> with ksh93 but without the features I mentioned.
Yes, that's something which I would fully agree with. Doing the switch
now for OpenSolaris would make much sense - but some things need to be
checked:
- Would Sun like/dislike the diversion between OpenSolaris vs. Solaris
in this point ? AFAIK SchillyX and other distributions did already
expore this part (e.g. ksh93 residing at /usr/bin/ksh) quite
successfully
- How long would such a diversion be tolerated by Sun (Release of
Solaris 11 is more than a year away - there is theoretically time to
sync Solaris with OpenSolaris before FCS, however this would need a very
detailed plan and the blessing of the PSARC) ?
In general I would like to focus now on getting ksh93 integrated first -
which means: Getting it building, running and commited to the
OpenSolaris tree.
Cyril Plisko <cyril.plisko at gmail.com> from genunix.org said we could
have a branch in their SVN (=Subversion) respository - which may be
ideal for getting a prototype running.
This prototype would also help finding any serious problems before we
move the work to the main OpenSolaris repository, hopefully avoiding
something which was called the "quality death spiral" (e.g. things in
the "trunk" branch are broken so people do not install it and more and
more issues may hide in "trunk" until the tree gets unuseable (something
which happened during Netscape4.x deveopment and caused Netscape to
create the "tinderbox" automated test system)) in the original
discussion...
----
Bye,
Roland
--
__ . . __
(o.\ \/ /.o) roland.mainz at nrubsig.org
\__\/\/__/ MPEG specialist, C&&JAVA&&Sun&&Unix programmer
/O /==\ O\ TEL +49 641 7950090
(;O/ \/ \O;)
