Re: PATCH: $Header$ handling

On Wednesday 21 July 2004 09:52, John Peacock wrote:
> I've been closely watching the keyword topic since I joined the list
> (and have a patch in process to change the keyword handling completely),
> and I have to say that $Header$ has not been that highly demanded (no
> offense intended). As I suggested in your original thread, the use of
> $Header$ to create reports is trying to map CVS methodology to Subversion.

I disagree. While I am not sure what "CVS methodology" you have in mind,
$Header$ has very clear advantages over $Id$.

1. A lot of projects use both interchangeably in CVS and RCS. Supporting
$Header$ would prevent an effort of turining $Header$ into $Id$

2. $Header$ allows the programmer to see exactly where a certain file comes
from. This is invaluable when analyzing the executable. You won't believe how
many times in my practice I saw people scratching their head when seeing two
or more different Id strings for the same filename. When you have a
production problem (which is usually the moment that calls for such
measures), having an additional clue is more than welcome.

3. I actually like $Header$ :-) It crams more useful information into the
string compared to $Id$.

> That being said, please read HACKING and include a log message in the
> project style so that this patch can be considered for adoption.

I am most definitely doing so now (who needs to RTFM *before* doing
something?!) ;-)

-- 
Alexander L. Belikoff                      GPG f/pr: 0D58 A804 1AB1 4CD8 8DA9
Bloomberg L.P.                                       424B A86E CD0D 8424 2701
abel *at* vallinor4 *dot* com             (http://pgp5.ai.mit.edu for the key)
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org