Here's a new webrev, at the same place:
http://cr.grommit.com/~stevel/156_ws/
With Stephen's & Rich's comments addressed.
My initial question re: protodefs still stands, since neither Darren,
Rich, nor I were able to address it:
Hi Bill (cc'ing tools-discuss),
I'm looking at 'ws', and wondering about the use of the
Codemgr_ws/sunos/protodefs file... Rich said you were able to convince him not
to throw it away; do you mind convincing me too? :-)
I'm debating about what to do in the Hg case, whether we should keep a symmetrical .hg/sunos/protodefs file, or what..., but I could use some more background.
Anyone know, or have an opinion?
cheers,
steve
Stephen Hahn wrote:
* Stephen Lau <[EMAIL PROTECTED]> [2007-02-15 15:24]:
I've taken Darren's ws, merged in the stuff from my attempt at porting
ws and have a webrev for a unified version here:
http://cr.grommit.com/~stevel/156_ws/
At the moment it only supports Mercurial & Teamware, I can add
Subversion support if people need it.
Anyone want to review it?
On a first pass, I didn't like
tmpwsname=`foo`
if [ -z "$tmpwsname ]; then
tmpwsname=`bar`
if [ -z "$tmpwsname ]; then
...
else
...
fi
else
...
fi
very much. I think
if [ -n "${tmpwsname:=$(cd $wsname >/dev/null && hg root 2>/dev/null)}" ];
then
...
elif [ -n "${tmpwsname:=$(cd $wsname >/dev/null && workspace name)}" ]; then
...
else
...
fi
is clearer for multiple SCMs. Oh, and the "cd $wsname && cmd" should
be well-named shell functions.
- Stephen
--
stephen lau // [EMAIL PROTECTED] | 650.786.0845 | http://whacked.net
opensolaris // solaris kernel development
_______________________________________________
tools-discuss mailing list
tools-discuss@opensolaris.org
