Minor legal questions about usr/src/lib/libshell/common/data/builtins.c ... / was: Re: [ksh93-integration-discuss] Re: [osol-code] Round two:((pre-)pre-review) ksh93-integrationwebrev 2007-02-02

[Roland Mainz]

Bonnie Corwin wrote:
> David.Comay at sun.com wrote:
> > Alan,
> >> I've made some fairly significant changes to perl in the process of
> >> integrating perl into Solaris.  I've *never* added Sun copyrights as a
> >> result.  I have legal approval to contribute to perl, and most of the
> >> contributions are really porting tweaks, so there's no IP that's worth
> >> protecting in any case. In every case I've got the fixes back into the
> >> upstream source before completing the Solaris integration, and that's
> >> much easier to do if I'm not cluttering the source with a load of
> >> questionable copyright assignments.
> >>
> >> Plus Roland doesn't work for Sun, so I don't understand why he should
> >> be putting *our* copyright on *his* work.  Adding the CDDL to files
> >> he's created from scratch is one thing (and even that's debatable if
> >> he wants to get them accepted upstream), but I really don't think
> >> porting changes should need us to assert copyright - it seems kinda
> >> ... discourteous?
> >
> >
> > You make two excellent points here - I still think getting
> > clarification from Bonnie is the right thing to do since I don't
> > understand the issues very well.
> 
> Hi,
> 
> I've been trying to follow this thread, and at this point I'm getting
> confused.
> 
> If someone can send me a pointer to a source hierarchy

The current ksh93-integration webres can be found at
http://www.nrubsig.org/people/gisburn/work/solaris/ksh93_integration/ksh93_integration_prototype004_webrev_20070210/allfiles/webrev/

> and specific
> questions that need answers, I'm happy to help figure this out.

AFAIK the only remaining question is whether the following change to the
AT&T source file "builtins.c"
(http://www.nrubsig.org/people/gisburn/work/solaris/ksh93_integration/ksh93_integration_prototype004_webrev_20070210/allfiles/webrev/usr/src/lib/libshell/common/data/builtins.c.html)
requires that we have to put a CDDL license at the top of that file:
-- snip --
Index: src/lib/libshell/common/data/builtins.c
===================================================================
--- src/lib/libshell/common/data/builtins.c	(revision 515)
+++ src/lib/libshell/common/data/builtins.c	(working copy)
@@ -120,7 +120,9 @@
 	"wait",		NV_BLTIN|BLT_ENV|BLT_EXIT,	bltin(wait),
 	"type",		NV_BLTIN|BLT_ENV,		bltin(whence),
 	"whence",	NV_BLTIN|BLT_ENV,		bltin(whence),
-#if SHOPT_CMDLIB_BLTIN
+#if SHOPT_CMDLIB_SOLARIS_BLTINS
+#include "solaris_cmdlist.h"
+#elif SHOPT_CMDLIB_BLTIN
 #define Q(f)		#f	/* libpp cpp workaround -- fixed 2005-04-11 */
 #define CMDLIST(f)	BDIR Q(f), NV_BLTIN|NV_NOFREE, bltin(f),
 #include <cmdlist.h>
-- snip --
This change is used to alter the list of ksh93 builtin commands (e.g.
those enabled by default) to those specified by PSARC 2006/550.

IMO the change doesn't justify the addition of a CDDL license template
since it only adds an #ifdef/#include/etc.-sequence to include the list
of builtins from another source file ("solaris_cmdlist.h") which is
already licensed as CDDL.

If that justification isn't enougth I'd like to follow
http://mail.opensolaris.org/pipermail/ksh93-integration-discuss/2007-January/002001.html
which AFAIK allows us to ship any of our code as CPL1.0 on demand and
would therefore avoid the need to add a CDDL license template into that
code (e.g. I am not happy with the idea to put a giant CDDL thing into
David&&Glenn's code for that tiny #ifdef/#include/etc.-hack...).

AFAIK that's all... :-)

----

Bye,
Roland

-- 
  __ .  . __
 (o.\ \/ /.o) roland.mainz at nrubsig.org
  \__\/\/__/  MPEG specialist, C&&JAVA&&Sun&&Unix programmer
  /O /==\ O\  TEL +49 641 7950090
 (;O/ \/ \O;)