The changes in usr/src/tools/findunref/Makefile process multiple exception_list* files (appropriately plural) into one exception_list, which is then passed to the C version of findunref.
It works, I tested it. There was some debate about why this belonged in usr/src/tools/findunref, instead of a workspace-specific location, and there was talk about modifying findunref.c to take multiple args. I went with this approach, explicitly, and thought that you (and others) were OK with it. That wasn't a "nobody seemed to object," it was more "I think we actually discussed this, and ended up at this point." That discussion even involved explicit advice to "wx mv" the existing exception_list to the new name (open, closed, teamware, mercurial, etc.) with the most commonality. ? --Mark On Wed, 9 Jul 2008, James Carlson wrote: > Date: Wed, 09 Jul 2008 13:51:46 -0400 > From: James Carlson <James.D.Carlson at Sun.COM> > To: Mark J. Nelson <Mark.J.Nelson at Sun.COM> > Cc: scm-migration-dev at opensolaris.org > Subject: Re: [scm-migration-dev] scm-migration tools code review followup > > Mark J. Nelson writes: >>>> Mark asked me send this along on his behalf, since he's out. >>> [...] >>>>> http://cr.opensolaris.org/~richlowe/toolsreview >>> >>> One thing that jumps out at me is that usr/src/tools/README.tools >>> describes using "exception_list*" with findunref, but that multiple >>> list change is gone with the switch back from Python to C. >>> >>> Other than that, it looks good. >> >> Actually, those changes are still present, and were reviewed. >> >> They may be subject to further modification (especially to being moved out >> of usr/src/tools/findunref, leaving the tools dir for tools and putting >> the workspace-specific stuff somewhere else), but they're part of >> supporting a split gate. Or more generically, nested repositories. > > I don't understand that answer, so I'm guessing that I didn't get my > review comment across accurately enough. Let me try again. > > The text for findunref in README.tools (line 192) says this: > > $ findunref $INTELSRC $INTELSRC/tools/findunref/exception_list* | \ > > Note the asterisk. That's going to expand to multiple files. Now > look at line 109 of findunref.c: > > if (argc != 2) { > > That's ENOTGONNAWORK. The C-based version of findunref does *not* > allow you to specify multiple exception files on the command line. It > requires precisely *one* file. > > The change I'm suggesting in the README.tools file is quite simple. > It's just: > > < $ findunref $INTELSRC $INTELSRC/tools/findunref/exception_list* | \ >> $ findunref $INTELSRC $INTELSRC/tools/findunref/exception_list | \ > ---- >> $ findunref $SPARCSRC $SPARCSRC/tools/findunref/exception_list* | \ > < $ findunref $SPARCSRC $SPARCSRC/tools/findunref/exception_list | \ > > Would it be possible to do that? > > -- > James Carlson, Solaris Networking <james.d.carlson at sun.com> > Sun Microsystems / 35 Network Drive 71.232W Vox +1 781 442 2084 > MS UBUR02-212 / Burlington MA 01803-2757 42.496N Fax +1 781 442 1677 >
