Webrev:
http://jurassic.us.oracle.com/~ebeasant/webrevs/7183731-v2/
components/mlocate/Makefile
Line 42 - Just to verify, this setting of localstatedir will
result in the local data ending up in /var/cache/mlocate and
not just /var/cache?
components/mlocate/mlocate.p5m
Line 27 - I'd suggest a richer pkg.description. When I updated
many of the ones in ON, I ended up looking at the initial
DESCRIPTION field in the man page for inspiration. So for
mlocate(1), that might result in something like
set name=pkg.description \
value="locate(1) reads one or more databases prepared by updatedb(1M)
and writes file names matching at least one of user specified patterns to standard
output, one per line."
(An aside, although the upstream name of the project and the
package name are "mlocate", do any distributions actually
deliver a binary called "mlocate"?)
Lines 38-44 - Is there a reason to put the binaries under
/usr/bin/$(MACH64) rather than just /usr/bin?
Lines 41, 44 - Just verifying that other distributions put
updatedb in /usr/bin? I would have expected it to be under
/usr/sbin or perhaps /usr/libexec on Linux.
components/mlocate/patches/mlocate-0.25-getmnt.patch
Lines 18-19 - Is there a reason you're not changing
PATH_MOUNTED to just MNTTAB (the latter is defined in
<sys/mnttab.h>?)
components/mlocate/updatedb.conf
Where did this particular file come from? It seems to not be
from the distribution but it also includes some things that
don't apply to Solaris such as AFS references and /usr/tmp (the
latter is already covered by /var/tmp being part of
PRUNEPATHS.)
Finally, what's the story with file/slocate? Will the Desktop
consolidation be doing a coordinated delivery of that package since it
seems both packages cannot coexist at the same time.
_______________________________________________
userland-discuss mailing list
[email protected]
http://mail.opensolaris.org/mailman/listinfo/userland-discuss
