quilt-dev
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Quilt-dev] [RFC][PATCH] quilt commit command


From: Jean Delvare
Subject: Re: [Quilt-dev] [RFC][PATCH] quilt commit command
Date: Thu, 19 May 2005 15:55:54 +0200 (CEST)

Hi all,

> > If Andreas approves the functionality (which is good, IMHO), finalize
> > would be a much better name. But there's a "if" in my sentence... ;)
>
> "drop" came to my mind first ;)

I tried thinking about a name for that function and came up with quite a
few. Not sure any is better than "finalize" but I thought I might just
throw them here and see if any happens to please the majority.

* accept
* merge
* burrow
* include
* incorporate
* baseline (or mainline)
* graft

Each of these names is (IMHO) meaningful if we choose it for the new
function, but most of them could also mean something different (e.g.
merge could mean "create a single patch from two or more different
patches", somehow an extended version of quilt fold, burrow could mean
"move the given patch at the bottom of the patch stack" etc...) I
realize it's quite hard to find a name which can only mean one thing
and people will get it right at first sight. Oh well, users will have to
read the manual for once ;)

The first name which came to my mind was actually "apply", since after
all that's really what the command does, apply a patch - but obviously
this name would be error-prone due to the "applied" and "unapplied"
commands already implemented in quilt, which give a different meaning to
the term "apply".

I haven't read the proposed patch, sorry. I joined the project not so
long ago and still feel a bit uncomfortable giving my opinion on
"important" things (such as adding a brand new command). Also, my
shell skills are probably not good enough at this point to give an
opinion on or find bugs in the code itself. If I spend enough time on
quilt, I hope I will be able to help more in the future.

One question comes to my mind about this patch. Most commands work on the
top-most patch by default. Will this one work on the bottom-most patch
by default? It looks like "finalizing" patches (or whatever we call
that) only makes sense if you consider the stack bottom-up. This seems
to be a somewhat new concept for quilt. Josh, does my remark make sense?
Could you explain to us how you plan to use the new function? More
precisely, what does your stack look like, and what patches from it do
you "finalize"?

(BTW, if we consider things that way, the new command could as well be
named "shift" in reference to the way perl handles lists - but then
again this name is not exactly explicit for the newcomer.)

Thanks,
--
Jean Delvare




reply via email to

[Prev in Thread] Current Thread [Next in Thread]