Subject: Re: bin/20200: Let etcupdate skip files with strictly local modifications (patch)
To: None <>
From: Christopher Richards <richards+netbsd@CS.Princeton.EDU>
List: netbsd-bugs
Date: 02/04/2003 14:57:19
On 4 Feb 2003 19:42:04 GMT, Greg A. Woods wrote:

> Yes, absolutely -- I've been harping over this for years.  Though the
> trick with RCS ID strings works well enough for identifying that changes
> have been made to the "master" version of at least those files which
> contain RCS ID strings [...]

You misunderstand what my change does.  It does *not* use RCS ID
strings to detect changes to the master version for the purpose of
merging those changes.

Quite the opposite: it uses them to detect when an installed file
differs from the master file solely by virtue of local modifications,
and then skip the processing of that file.  If the RCS IDs are the
same, the only effect of installing the "new" master file is to revert
all the local changes.  This is infrequently desired.  And if it is
desired, just run etcupdate without "-l".

How does my patch fail to do the right thing in the cases it covers?
How might this new, optional behavior cause harm to one's system?
I don't see how, but I may have overlooked something.