Build failure: feature not provided

Build failure: feature not provided

[Harald Hanche-Olsen]

Compiling /local/src/emacs/emacs-cvs/lisp/./org/org-export-latex.el

In toplevel form:
org/org-export-latex.el:52:1:Error: Required feature `cal-loaddefs' was not provided

The following patch seems to cure the problem, though it is probably
not the right thing to do. Better to fix the automatic generation of
the -loaddefs files to include an appropriate provide form?

- Harald


Index: lisp/calendar/calendar.el
===================================================================
RCS file: /cvsroot/emacs/emacs/lisp/calendar/calendar.el,v
retrieving revision 1.259
diff -u -r1.259 calendar.el
--- lisp/calendar/calendar.el	12 Apr 2008 03:11:03 -0000	1.259
+++ lisp/calendar/calendar.el	20 Apr 2008 14:33:19 -0000
@@ -116,7 +116,7 @@
 
 ;;; Code:
 
-(require 'cal-loaddefs)
+(load "cal-loaddefs" nil t)
 
 ;; Avoid recursive load of calendar when loading cal-menu.  Yuck.
 (provide 'calendar)
Index: lisp/calendar/diary-lib.el
===================================================================
RCS file: /cvsroot/emacs/emacs/lisp/calendar/diary-lib.el,v
retrieving revision 1.170
diff -u -r1.170 diary-lib.el
--- lisp/calendar/diary-lib.el	16 Apr 2008 03:23:30 -0000	1.170
+++ lisp/calendar/diary-lib.el	20 Apr 2008 14:33:20 -0000
@@ -31,7 +31,7 @@
 ;;; Code:
 
 (require 'calendar)
-(require 'diary-loaddefs)
+(load "diary-loaddefs" nil t)
 
 (defgroup diary nil
   "Emacs diary."
Index: lisp/calendar/holidays.el
===================================================================
RCS file: /cvsroot/emacs/emacs/lisp/calendar/holidays.el,v
retrieving revision 1.77
diff -u -r1.77 holidays.el
--- lisp/calendar/holidays.el	11 Apr 2008 03:48:37 -0000	1.77
+++ lisp/calendar/holidays.el	20 Apr 2008 14:33:21 -0000
@@ -31,7 +31,7 @@
 ;;; Code:
 
 (require 'calendar)
-(require 'hol-loaddefs)
+(load "hol-loaddefs" nil t)
 
 (defgroup holidays nil
   "Holidays support in calendar."

Re: Build failure: feature not provided

[Stefan Monnier]

> Compiling /local/src/emacs/emacs-cvs/lisp/./org/org-export-latex.el
> In toplevel form:
> org/org-export-latex.el:52:1:Error: Required feature `cal-loaddefs' was not provided

This should be a side effect of previous builds (i.e. fixed by a clean
checkout).  "maintainer-clean" should also fix it, but maybe it doesn't.

> The following patch seems to cure the problem, though it is probably
> not the right thing to do. Better to fix the automatic generation of
> the -loaddefs files to include an appropriate provide form?

Actually, I think those patches are good, independently from whether or not
maintainer-clean does the right thing above.


        Stefan

Re: Build failure: feature not provided

[Glenn Morris]

Stefan Monnier wrote:

> "maintainer-clean" should also fix it, but maybe it doesn't.

It assuredly does (though there's nothing really to "fix"), if people
would bother to search the list archives before reporting a problem as
fundamental as a failure to build, over and over again.

>> Better to fix the automatic generation of the -loaddefs files to
>> include an appropriate provide form?

It already does, provided you start from a clean slate.

> Actually, I think those patches are good, independently from whether or not
> maintainer-clean does the right thing above.

Obviously I disagree, since I made the opposite changes in the first place.

Re: Build failure: feature not provided

[Harald Hanche-Olsen]

+ Stefan Monnier <monnier@iro.umontreal.ca>:

> "maintainer-clean" should also fix it, but maybe it doesn't.

Ah, I had missed that part of INSTALL.CVS.
I'll keep it mind for the future. Thanks.

- Harald

Re: Build failure: feature not provided

[Paul R]

Glenn Morris <rgm@gnu.org> writes:

> Stefan Monnier wrote:
>
>> "maintainer-clean" should also fix it, but maybe it doesn't.
>
> It assuredly does (though there's nothing really to "fix"), if people
> would bother to search the list archives before reporting a problem as
> fundamental as a failure to build, over and over again.

Sorry for this OT message, but I really find this tone unadapted when
speaking about someone making the effort to report a bug. If you don't
want to read a post, just skip it, or use your killfile, but please
mind to keep the atmosphere as good as you can. Thank you.

Re: Build failure: feature not provided

[Stefan Monnier]

> Obviously I disagree, since I made the opposite changes in the first place.

I know.  But I have no idea why you bothered to make such a change.
After all, the file that does the `require' should be loaded only once
anyway and this `require' is the only place where we may load this
loaddefs file, so `load' would work just as well.

And if someone ever decides to reload calendar.el, I wonder why you
think it wrong for cal-loaddefs. to also be reloaded.

It's not like it matters much, either way, really.


        Stefan

Re: Build failure: feature not provided

[Harald Hanche-Olsen]

+ Paul R <paul.r.ml@gmail.com>:

> Glenn Morris <rgm@gnu.org> writes:
> 
> > Stefan Monnier wrote:
> >
> >> "maintainer-clean" should also fix it, but maybe it doesn't.
> >
> > It assuredly does (though there's nothing really to "fix"), if people
> > would bother to search the list archives before reporting a problem as
> > fundamental as a failure to build, over and over again.
> 
> Sorry for this OT message, but I really find this tone unadapted when
> speaking about someone making the effort to report a bug.

That's okay. As a long time net user, I have developed thick skin and
am not easily offended. However, I might point out that archives for
this list were not available when I looked, back in February I think
it was. I see they are now, so I could indeed have found the answer in
the archives had I known they existed. Oh, wait a minute ... this is
emacs-pretest-bug, but I see these messages appearing in the
emacs-devel archives? Are the two addresses really the same list?

- Harald