info-cvs
[Top][All Lists]
Advanced

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

Re: cvs commitinfo - remote access issues - enforced code beautification


From: Greg A. Woods
Subject: Re: cvs commitinfo - remote access issues - enforced code beautification
Date: Thu, 21 Mar 2002 15:24:22 -0500 (EST)

[ On Thursday, March 21, 2002 at 09:46:06 (-0800), Noel Yap wrote: ]
> Subject: Re: cvs commitinfo - remote access issues - enforced code 
> beautification
>
> If you're going this route, I would suggest your
> commitinfo script also check for the existence of RCS
> keywords within the files.

That's certainly a good suggestion that anyone making use of this hack
should be very careful to follow.  Also check to make sure '-kb' and
'-ko' are not set (either 'sticky' in the working directory [[hmmm... is
that possible?]] or in the RCS file itself, I think).

> I think such enforcement would require too much
> overhead.

Well, this overhead would be a lot lower than the current overhead of
sending the whole file back instead of just the diffs.

Indeed the easiest and most certain way of getting the correct diffs
requires keeping a copy of the original as it was received from the
client anyway.

> Also, even if such a feature were supported, it
> shouldn't be used since resolving conflicts due to
> mass changes from beautification is an extremely
> tedious task.

I think this is the tipping factor.

While anyone can argue that keyword expansion is an example of the
change control tool making changes itself, this argument doesn't hold
for making other cosmetic changes to the checked in revision.  Keyword
expansions are done under the explicit instructions of the user doing
the commit, and in explicit locations identified specifically for this
purpose by the user.

Other more arbitrary cosmetic fixes are done at the surprise, and
sometimes to the dismay, of the user.

Finally I should note that any form of modification done by CVS to the
working file greatly increases the risk that the user will overwrite the
modified file with a stale copy from an editor buffer or similar.

-- 
                                                                Greg A. Woods

+1 416 218-0098;  <address@hidden>;  <address@hidden>;  <address@hidden>
Planix, Inc. <address@hidden>; VE3TCP; Secrets of the Weird <address@hidden>



reply via email to

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