>>>>> "Nathan" == Nathan Bush <nathan.bush at sun.com> writes:
Nathan> 1) Why move the mktpl step at all? It seems that it was already
Nathan> happening before findunref. The new location introduces two
Nathan> consecutive if-statements (lines 2130, 2147) that have the same
Nathan> conditional expression. They might as well be run together into
Nathan> one if-statement.
Yes and no. The problem that I see with putting them all in 1 big if
block is that it blurs the logical separation between the two activities
(generating the license files and generating the open-only binaries).
I can put them into 1 "if" block if folks really think that's the right
thing to do, but I'll insist on a comment like
# This step is logically separate from everything else in this
# if block (generation of open-only binaries, etc).
Nathan> 2) This is minor, but w.r.t. the lines:
Nathan> 2639 echo "\n==== Generating OpenSolaris tarballs ====\n" | \
Nathan> 2640 tee -a $LOGFILE >> $mail_msg_file
Nathan> Most of the rest of the script has the "tee" command written as
Nathan> "tee -a $mail_msg_file >> $LOGFILE". Unless it matters here it
Nathan> might as well be made to match.
It appears both ways at various places in the script. I think ">>
$LOGFILE" is more common, but I didn't do a count. I'll update my
changes to use ">> $LOGFILE".
Nathan> 3) In lines 2683-2708, it looks like the order of sdrop and
Nathan> bfudrop were reversed relative to each other vs. the original
Nathan> version. If I followed the bug comments correctly, these two
Nathan> steps do not interact. Is there another reason for this
Nathan> switch?
Mostly historical. I originally just moved sdrop, before I discovered
that I had to move bindrop. Also, come to think of it, I need to see
which steps depend on README.opensolaris and add a comment about that
dependency.
mike
