Conference smurf::buildhelp

Date Of Receipt: 	10-OCT-1994 13:26:30.09
From: 	FLAMBE::"[email protected]" "10-Oct-1994 1020"
To: 	[email protected]
CC: 	[email protected], [email protected]
Subj: 	Re: HELP! Can't submit AGAIN!

Peter,

DECode II prepends your principal name to your set name. It looks like somewhere
along the way you've also prepended your principal name to a set name, and also
switched setnames (mixing bl6 and atmproj). When you tried the "-set" option,
you compounded the problem by adding your principal name again, and also
".BCSset-".

DECode II is not as consistent as it should be in handling set names (basically
you should never be specifying your principal name as part of a set name, even
though DECode II sometimes reports it to you with the principal name prepended).

To figure out which set you should be doing this from, do the following :

Look in your sandbox base to see which .BCSset* file contains the name of the
file you want to submit. For example, .BCSset-Peter_Schulter_myset, where
"myset" is the set name you specified for the workon during which you checked
out this file, and quite possibly, in your case, contains another ocurrence of
"Peter_Schulter".

Remove the ".BCSset-Peter_Schulter" from the beginning of that filename (the
first ocurrence only). This now gives you the name of the set that contains the
file you want to submit (in my example, it's "myset").

Exit the workon, and re-enter using "workon -sb bl6 myset" (where "myset" is the
setname calculated above). Re-try the submit.

Now to clean up the naming mess ...

Edit $SOURCEBASE/../rc_files/sets. There is a line in this file for each of your
sets. Each line will be of the form :
	set Peter_Schulter_<setname> <path>
where <setname> is a set name, and <path> is a path to the directory you want to
be in when you workon the set.

For each line starting with "set Peter_Schulter_<setname>" which does *not* have
a corresponding $SOURCEBASE/.BCSset-Peter_Schulter_<setname>, you can delete the
line if you wish (I would recommend this for every line that contains multiple
ocurrences of "Peter_Schulter" for your own sanity). However, don't delete the
line whose setname is the same as that listed in the first line of the file,
which starts with "default".

If you have entries in the sets file that you want to delete, but have a
corresponding .BCSset-* file, you'll have to submit the files from that set
before cleaning up (see the top portion of this mail).

Let me know if you have any more questions, and feel free to send me an 'ls -a'
of your $SOURCEBASE and the contents of your sets file if you're still having
problems.

	Mike (DECode II support this week)

Date Of Receipt: 	10-OCT-1994 13:49:37.60
From: 	FLAMBE::"[email protected]" "10-Oct-1994 1343"
To: 	[email protected]
CC: 	[email protected]
Subj: 	Re: HELP! Can't submit AGAIN!

Mike,

  That didn't work.  Rather than continue to fight with ODE I just
blew away the sandbox and created a new one.  The problem seems to
have been that ODE had become confused about who I was.  The set
name it was using,

	.BCSset-atmproj_Peter_Schulter_Peter_Schulter_bl6

was a composition of the two names by which I'm known to ODE.  The
first, atmproj_Peter_Schulter, is my id for administering the shared
sandbox.  The second, is my normal name, and bl6 was the name of
my sandbox.

Making a new sandbox seems to have fixed the problem for now, but 
now I'm stuck with re-creating my work that I needed to submit.  I
figure doing a few extra days of coding is easier than trying
to understand ODE and trying to get it to work ;-}

Thanks,

 --- pete

Date Of Receipt: 	10-OCT-1994 15:03:28.89
From: 	FLAMBE::jmf "Joshua M. Friedman OSF/UNIX SDE"
To: 	[email protected], [email protected]
CC: 	[email protected]
Subj: 	Re: HELP! Can't submit AGAIN!

Pete, if you can supply a list of names, if the setnames have
become discombobulated within the RCS branch labels, we can reset
the label names to what they should be, and you can use the same
sandbox you had.

When you said you "blew away" your sandbox, did you mksb -undo?
It wouldn't have let you do that if you hadn't outdated all your
branches.  If you've "rm -r"'d the sandbox, your leaving your
branches still sitting around in RCS, which you can still reference
either on purpose or by accident (the latter is bad and causes
confusion and can potentially drop others' submits).

So can we help here or have you even lost the list of what files
you had checked out (recorded in the .BCSset- file in the sandbox)?

-josh	

Date Of Receipt: 	10-OCT-1994 15:22:39.41
From: 	FLAMBE::"[email protected]" "10-Oct-1994 1508"
To: 	Joshua M. Friedman OSF/UNIX SDE <[email protected]>
CC: 	[email protected], [email protected], [email protected]
Subj: 	Re: HELP! Can't submit AGAIN!

Josh,

 I just got fed up with ode and did an "rm -r" on my sandbox (I
have never done a "mksb -undo" and didn't even know it existed,
rm ALWAYS works).  It's working fine now and I was able to properly
submit one of the files I had checked out (the one that was critical
to get in) after bco'ing them again and re-doing my changes.

It's working now and I don't want to touch it.  I don't have time
to try to figure out what went wrong or how to fix it as I'm up
against a code freeze next week and have a lot of work to do.

Thanks for your help,

 --- pete