Date: Sun, 26 Jan 2003 10:57:53 -0800 (PST) From: Matthew Dillon <dillon@apollo.backplane.com> To: Doug Barton <DougB@FreeBSD.ORG> Cc: cvs-committers@FreeBSD.ORG, cvs-all@FreeBSD.ORG Subject: Re: cvs commit: src/usr.sbin/mergemaster mergemaster.sh Message-ID: <200301261857.h0QIvr65051817@apollo.backplane.com> References: <200301251918.h0PJI6ld069132@freefall.freebsd.org> <20030126040430.Q26257@12-234-22-23.pyvrag.nggov.pbz>
next in thread | previous in thread | raw e-mail | index | archive | help
:I was one of those people, but I don't think that I made my objection
:explicit. I'm not sure if that was a mistake or not, however I honestly
:didn't expect you to commit this change without passing it by me first.
:
:I've been very clear about my desire to review changes to mergemaster
:before they are committed. My preference is listed in MAINTAINERS, and the
:Makefile.
:
:I have the following technical problems with this patch as well:
:
:1. You've redefined the meaning of the 'v' menu option. This is a _huge_
:UI bozo no-no.
I did? No I didn't! What did I do to 'v'? It's the same as it always
has been. It views the whole diff.
In anycase, lets try to turn this into a positive affair rather then
yet another furball. I'm asking people to *TRY* this patch and improve
upon it. I don't mind making style changes or other changes but I would
like people to at least try it out.
I've had this in my tree for so long I honestly didn't think to check
MAINTAINERS before comitting it. I apologize for that.
:The one part of your patch that I can agree with is the concept of using
:'clear' to start display of new messages. I'd be willing to incorporate
:that part if I can prove that this is a safe thing to do in all
:circumstances. The rest of your patch only applies to diffs which are
:shorter than $ROWS; and the meat of your changes are accomplished just by
:using 'clear'. Diffs that are longer are already handled by PAGER.
:
:So, at this point I'd really like you to back out this change. I'll work
:on a version that incorporates the 'clear,' and send it over to see if it
:meets your needs.
:
:Doug
I actually think it's better to generate a sufficient number of newlines
so the new screen is properly aligned, rather then clear, so the xterm
history mechanism works as expected (which would be an improvement over
running it through a pager), but clear was easier to implement.
I would ask you not to request a backout, because there is really no
reason that a backout should be needed. Instead, allow the changes to
be left in the tree and work on your version then simply commit it over
mine (I would like to see it before hand). While the changes are minor
and a backout would not a big deal (compared to, say, the struct file
changes and backout), I do not believe that the current commit breaks
anything so it would be more efficient simply to leave it in and
commit over it. And I really think that once people start using it they
won't be able to live without it.
-Matt
Matthew Dillon
<dillon@backplane.com>
To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe cvs-all" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?200301261857.h0QIvr65051817>