On Wed, Mar 19, 2008 at 12:01:57PM -0700, [EMAIL PROTECTED] wrote:
> http://cr.opensolaris.org/~comay/webrev-0/
Makefile:
- line 146: This script is rather un-makefiley. Perhaps something like
FILELIST_RAMDISK_PKG_$(BUILDID) = SUNWcakr.i
FILELIST_RAMDISK_PKG_83a = SUNWcsr
FILELIST_RAMDISK_PKG_84 = SUNWckr
proto/boot/solaris/filelist.ramdisk: proto
./get_file_from_pkg.py $(WOS_PKGS)/$(FILELIST_RAMDISK_PKG_$(BUILDID)) >
\
boot/solaris/filelist.ramdisk > $@
echo etc/zfs/zpool.cache >> $@
Of course, if it stays in SUNWckr from 84 on, then the default above
will be wrong. Perhaps it should default to SUNWckr, and the known set
of builds where it's in SUNWcakr.i should be explicit.
Why do you have 83a at all now?
If you choose to keep the case statement, you can ditch the parens
around get_file_from_pkg and the semicolon at the end of those
statements.
svc-dlmgmtd:
- line 42: this now doesn't test the result of dlmgmtd, but of the ppriv
command.
SUNWgnome-media-mp3.84:
- is there a bug filed against the inability to remove packages on
upgrade?
I think for my own sanity in future reviews I'm going to have to fix
solaris.py to search for include files from a list of directories; this
.build thing, with lines whizzing around all over the place is getting
craaazy. (Feel free to say you told me so. :)
Danek
_______________________________________________
pkg-discuss mailing list
[email protected]
http://mail.opensolaris.org/mailman/listinfo/pkg-discuss
