Re: Build Emacs from CVS

[Andreas Schwab]

Eli Zaretskii <address@hidden> writes:

|> On Tue, 23 Oct 2001, Kenichi Handa wrote:
|> 
|> > They are re-build by "make bootstrap", thus get different
|> > from what stored in the repository.  If we do "cvs update"
|> > next time, there's a possiblity of conflict (although
|> > usually cvs merges the difference successfully).
|> 
|> Yes, but I think removing them will make more trouble than leaving them.
|> 
|> People who use the CVS sources should know what they are doing.  I don't 
|> think any of the developers had any problems with loaddefs.el during the 
|> entire pretest, right?  So, evidently, this problem is not too hard to 
|> solve.

The problem is that loaddefs.el contains timestamps which are different in
every working directory.  This causes CVS conflicts.  Since I normally
pack the CVS contents in a tar file and feed that to rpm I only get such
conflicts only occasionally and it is not difficult to resolve them.

|> If that's not good enough, we could arrange for automatic updates of 
|> loaddefs.el.  I think the same machinery that mails me every commit to 
|> the CVS tree can also update loaddefs.el.

That would not fix the basic problem with make bootstrap in the CVS
working directory.  cus-load.el, on the other hand, is easier to deal
with.

Andreas.

-- 
Andreas Schwab                                  "And now for something
address@hidden                          completely different."
SuSE Labs, SuSE GmbH, Schanzäckerstr. 10, D-90443 Nürnberg
Key fingerprint = 58CA 54C7 6D53 942B 1756  01D3 44D5 214B 8276 4ED5