Re: [Gnumed-devel] (no subject)re: reverting

[Karsten Hilbert]

> "But if for some reason the file was moved and later moved back the one with
> the lower numer is actually newer"
> 
> This isn't true.
Yes it is.

1.1 -> 1.2 -> mv -> 1.1 -> 1.2 -> 1.3 -> mv -> 1.1

Now the last 1.1 is more recent/more up to date than the 1.3.

What cannot happen is this:

1.1 -> 1.2 -> mv -> 1.1 -> 1.3

Here, 1.2 is missing in the new location (it is a different
1.2 than the one that was in the old location !).

CVS has many deficiencies and this is one way to work around
them. But one must be careful.

> Which is probably why other patches supposedly
> got clobbered by syan's commit.  Making a copy of a bunch of files and putting
> them in test-area, and then copying them back to the previous directory later
> and comitting is a BAD idea.
Well, actually the idea was to commit-in the changes made to
the copies of the files but apparently Syan didn't do that
after all but simply copied files over. The other thing was
that at one point he must have gotten merge conflict messages
from the CVS system (and CVS adds some lines to the files in
such cases) and chose not to resolve them but rather simply
comment out the merge conflict markers and hack around them
until it worked.

> Unless your intent from the beginning is to
> replace the originals with the new.  Otherwise best to work in the main
> directories, make small changes, update, commit, update commit, provide good
> commit comments, comment code, etc...
I wonder how many times we have requested that of people. The
test-area was set up because people did not do precisely that
but rather submitted larger patches to the main tree.

Karsten
-- 
GPG key ID E4071346 @ wwwkeys.pgp.net
E167 67FD A291 2BEA 73BD  4537 78B9 A9F9 E407 1346