Conference smurf::buildhelp

1438.0. "fyi: fixing non-binary files with BIN comment leaders in the OS pools" by AOSG::FILTER (Automatic Posting Software - mail to flume::puck) Wed Apr 26 1995 19:38

Date Of Receipt: 	26-APR-1995 17:46:47.26
From: 	SMURF::FLUME::jmf "Joshua M. Friedman OSF/UNIX SDE  26-Apr-1995 1745"
To: 	plong@DEC:.zko.flume, odehelp@DEC:.zko.flume
CC: 	
Subj: 	fyi: fixing non-binary files with BIN comment leaders in the OS pools

------- Forwarded Message

Return-Path: jmf 
Delivery-Date: Wed, 26 Apr 95 17:42:55 -0400
Return-Path: jmf
Received: from localhost by flume.zk3.dec.com; (5.65v3.0/1.1.8.2/16Jan95-0946AM)
	id AA21854; Wed, 26 Apr 1995 17:42:40 -0400
Message-Id: <[email protected]>
To: mwarren, brett
Cc: tresvik, releng, [email protected]
Subject: fixing non-binary files with BIN comment leaders in the OS pools
Date: Wed, 26 Apr 95 17:42:40 -0400
From: "Joshua M. Friedman, OSF/UNIX SDE 381-1548" <jmf>
X-Mts: smtp

Mark & Brett,

I thought I'd check with you on this for a sanity check; can you please
review these files to see if you have any concern about my resetting
their comment-leader with no submits to any pools?  I've looked at them
all and no submits are needed for any of them to fix the files to work
properly and not be considered "binary" by ODE; the comment leader can
just be changed and that's all.  I'd like to do this soon (this week).

(Actually, there are three files which really should have a submit to
clean up comment lines that start with "BIN" or "NONE" instead of "# ",
but they haven't been bothering anyone so far; I've listed them at the
end of the mail.)

Note this is just for the 'os' pool; I haven't yet investigated the
'x11' or 'cde' pools.

I've done some research and have found 102 files under src (33 under
src/kernel) which are set to type BIN (or NONE, archaic for BIN), but
do have ODE headers with $Log: which are not getting expanded.  These
are the result of people doing a "bcreate" for a file with unknown or
no extension (which prints a message about setting the leader to BIN);
they've included an ODE header manually, and assumed ODE would do the
"right thing".  It doesn't know any better, and the header becomes part
of the Binary file.  In some cases people have put their comments in
the HISTORY part of the header, clearly expecting them to get used.

Files which are considered binary but are not are a hassle because they
can't be diff'd or merged by ODE.  There may be many other files which
are listed as BIN (there were 2300 in the full os pool), which don't
have ode headers, but possibly could.  This should also be investigated,
but it's a larger scale effort.

For all of these 102 files, I was planning to reset the comment leader
to " * " or "# " or "$ " as appropriate.  Subsequent check-outs would
then get the new correct behavior of expanding comments, though these
files should be reviewed to make sure nothing will break as a result.
I've reviewed them all visually in goldminos, ptliteos, and ptos and
believe them all to be ok.

The full list of files to be changed, along with their new comment
leader, is as follows:

Thank you very much...			-josh



'# ' ./etc/.mrg..inittab
' * ' ./etc/eca/builtin_types.proto
' * ' ./kernel/arch/alpha/spl.h
' * ' ./kernel/atm/uni3.0/common/cm1.x
' * ' ./kernel/atm/uni3.0/common/cm4.x
' * ' ./kernel/atm/uni3.0/common/cm6.x
' * ' ./kernel/atm/uni3.0/common/dat.x
' * ' ./kernel/atm/uni3.0/common/gen.x
' * ' ./kernel/atm/uni3.0/common/lm_pt.x
' * ' ./kernel/atm/uni3.0/common/sm.x
' * ' ./kernel/atm/uni3.0/common/ss_pt.x
' * ' ./kernel/atm/uni3.0/q93b/am.x
' * ' ./kernel/atm/uni3.0/q93b/amt.x
' * ' ./kernel/atm/uni3.0/q93b/mf.x
' * ' ./kernel/atm/uni3.0/qsaal/as.x
' * ' ./kernel/atm/uni3.0/qsaal/mac.x
' * ' ./kernel/io/atm/cmm/cmm.c
' * ' ./kernel/io/atm/uni3.0/common/cm1.x
' * ' ./kernel/io/atm/uni3.0/common/cm4.x
' * ' ./kernel/io/atm/uni3.0/common/cm6.x
' * ' ./kernel/io/atm/uni3.0/common/dat.x
' * ' ./kernel/io/atm/uni3.0/common/gen.x
' * ' ./kernel/io/atm/uni3.0/common/lm_pt.x
' * ' ./kernel/io/atm/uni3.0/common/sm.x
' * ' ./kernel/io/atm/uni3.0/common/ss_pt.x
' * ' ./kernel/io/atm/uni3.0/q93b/am.x
' * ' ./kernel/io/atm/uni3.0/q93b/amt.x
' * ' ./kernel/io/atm/uni3.0/q93b/mf.x
' * ' ./kernel/io/atm/uni3.0/qsaal/as.x
' * ' ./kernel/io/atm/uni3.0/qsaal/mac.x
'# ' ./kernel/io/common/.mrg..conf.c
' * ' ./kernel/mach/mach_host.defs
' * ' ./kernel/mach/memory_object.defs
' * ' ./kernel/mach/monitor.defs
'# ' ./kernel/streams/.mrg..str_config.c
'# ' ./root/.mrg..DXsession
'# ' ./sbin/.mrg..bcheckrc
'# ' ./sbin/cfgmgr/.mrg..sysconfigtab
'# ' ./sbin/init.d/autosysconfig
'# ' ./sbin/it.d/bin/twouser.sh
'# ' ./usr/bin/at/at.deny
' * ' ./usr/bin/at/att1.y
'$ ' ./usr/bin/basename/basename.msg
'$ ' ./usr/bin/bdiff/bdiff.msg
'$ ' ./usr/bin/df/df.msg
'$ ' ./usr/bin/expr/expr.msg
'$ ' ./usr/bin/getopt/getopt.msg
'$ ' ./usr/bin/test/test.msg
'# ' ./usr/bin/uucp/Devices
'# ' ./usr/bin/uucp/Dialcodes
'# ' ./usr/bin/uucp/Maxuuscheds
'# ' ./usr/bin/uucp/Maxuuxqts
'# ' ./usr/bin/uucp/Permissions
'# ' ./usr/bin/uucp/Poll
'# ' ./usr/bin/uucp/Systems
'$ ' ./usr/bin/uucp/uuencode.msg
'# ' ./usr/bin/uucp/uugettydefs
'$ ' ./usr/ccs/lib/libchf/chf.msg
'# ' ./usr/ccs/lib/libchf/libchf.a.uu
' * ' ./usr/ccs/lib/libcurses/_redraw.c
' * ' ./usr/ccs/lib/libcurses/addwch.c
' * ' ./usr/ccs/lib/libcurses/addwchstr.c
' * ' ./usr/ccs/lib/libcurses/addwstr.c
' * ' ./usr/ccs/lib/libcurses/bsd/curses.ext
' * ' ./usr/ccs/lib/libcurses/capnames.c
'# ' ./usr/ccs/lib/libcurses/caps
' * ' ./usr/ccs/lib/libcurses/curs_wch.h
' * ' ./usr/ccs/lib/libcurses/curses.h
' * ' ./usr/ccs/lib/libcurses/curshdr.h
' * ' ./usr/ccs/lib/libcurses/cursor_off.c
' * ' ./usr/ccs/lib/libcurses/cursor_on.c
' * ' ./usr/ccs/lib/libcurses/getwch.c
' * ' ./usr/ccs/lib/libcurses/getwstr.c
' * ' ./usr/ccs/lib/libcurses/inswch.c
' * ' ./usr/ccs/lib/libcurses/inswstr.c
'# ' ./usr/ccs/lib/libcurses/libcurses-objs.mk
'# ' ./usr/ccs/lib/libcurses/maketerm.ex
' * ' ./usr/ccs/lib/libcurses/termcap.c
' * ' ./usr/ccs/lib/libcurses/ungetwch.c
' * ' ./usr/ccs/lib/libcurses/vwprintw.c
' * ' ./usr/ccs/lib/libcurses/vwscanw.c
' * ' ./usr/ccs/lib/libcurses/winwch.c
' * ' ./usr/ccs/lib/libcurses/winwstr.c
'# ' ./usr/ccs/lib/libids_nox/Makefile
'# ' ./usr/ccs/lib/libimg/Makefile
'# ' ./usr/ccs/lib/libips/Makefile
'# ' ./usr/ccs/lib/libips/libips.a.uu
'# ' ./usr/examples/devdriver/cb100/Makefile
' * ' ./usr/examples/devdriver/cb100/cbreg.h
'# ' ./usr/examples/devdriver/none100/Makefile
'# ' ./usr/sbin/cron/.mrg...proto
'# ' ./usr/sbin/cron/.mrg..adm
'# ' ./usr/sbin/cron/cron.deny
' * ' ./usr/sbin/latcp/lattelnet.c
'# ' ./usr/sbin/tcpdump/atime.awk
'# ' ./usr/sbin/tcpdump/ostype.awk
'# ' ./usr/sbin/tcpdump/packetdat.awk
'# ' ./usr/sbin/tcpdump/send-ack.awk
'# ' ./usr/sbin/tcpdump/stime.awk
'# ' ./usr/share/mk/posix.mk
'# ' ./usr/shlib/libcurses/Makefile
'# ' ./usr/share/lib/timezone/dst.db

Three files from this list  which are a bit bogus already - look at them
in ptos/src; they have BIN or NONE in the left column:

./kernel/mach/monitor.defs       <--- syntax ok cause in /* */ block
./kernel/io/common/.mrg..conf.c  <--- clearly broken in pt; should probably
				      be reverted to ptlite's version since
				      no update install in pt yet
./usr/share/lib/timezone/dst.db  <--- looks like broken syntax, but may be
				      forgiving if still used; has been
				      there for a long time



------- End of Forwarded Message