23.0.60; compilation fails with: Wrong number of arguments: autoload, 12

23.0.60; compilation fails with: Wrong number of arguments: autoload, 12

[Peter Dyballa]

Hello!

Still no success this sunny morning:

Loading ..../lisp/startup.el (source)...
Lists of integers (garbage collection statistics) are normal output
while building Emacs; they do not indicate a problem.
((106836 . 38433) (6098 . 29) (593 . 86) 441045 1034302 (27 . 1)  
(19 . 0) (8971 . 416))
Loading ..../lisp/loaddefs.el (source)...
Wrong number of arguments: autoload, 12
make[2]: *** [bootstrap-emacs] Error 255
make[2]: Target `bootstrap' not remade because of errors.
make[1]: *** [bootstrap-build] Error 2
make: *** [bootstrap] Error 2

I have to admit that the code (C and ELisp) contains patches for  
Emacs.app (http://emacs-app.sourceforge.net/) that were OK last week ...

Mac OS X 10.4.10, GCC 4.0.1

Here it is, line #29157:

	(autoload 'url-file-nondirectory "url-util" "\
	Return the nondirectory part of FILE, for a URL.
								    <==========
	(autoload 'url-parse-query-string "url-util" "\
	Not documented
	
	\(fn QUERY &optional DOWNCASE ALLOW-NEWLINES)" nil nil)



Can it be due to these changes (50 vs. 60, not all differences cited):

	448d395
	< (autoload 'dirtrack-mode "dirtrack" "\
	470,471c417
	< (autoload 'doc-view-mode "doc-view" "\
	< (autoload 'doc-view-minor-mode "doc-view" "\
	---
	> (autoload 'doc-view "doc-view" "\
	1250d1198
	< (autoload 'mail-abbrevs-mode "mailabbrev" "\
	1289,1290c1237,1238
	< (autoload 'message-bold-region "message" "\
	< (autoload 'message-unbold-region "message" "\
	---
	> (autoload 'bold-region "message" "\
	> (autoload 'unbold-region "message" "\
	1416d1367
	< (autoload 'turn-on-orgstruct++ "org" "\
	1427d1377
	< (autoload 'org-agenda-to-appt "org" "\
	1642,1645d1591
	< (autoload 'remember "remember" "\
	< (autoload 'remember-other-frame "remember" "\
	< (autoload 'remember-clipboard "remember" "\
	< (autoload 'remember-diary-extract-entries "remember-diary" "\
	2044c1994
	< (autoload 'vc-revision-other-window "vc" "\
	---
	> (autoload 'vc-version-other-window "vc" "\

or due to missing nil nil or such here:

(autoload 'double-mode "double" "\
Toggle Double mode.
Setting this variable directly does not take effect;
use either \\[customize] or the function `double-mode'.")




--
Greetings

   Pete

Think of XML as Lisp for COBOL programmers.
                               -- Tony-A (some guy on /.)

Re: 23.0.60; compilation fails with: Wrong number of arguments: autoload, 12

[Dan Nicolaescu]

Peter Dyballa <Peter_Dyballa@Freenet.DE> writes:

  > Hello!
  >
  > Still no success this sunny morning:
  >
  > Loading ..../lisp/startup.el (source)...
  > Lists of integers (garbage collection statistics) are normal output
  > while building Emacs; they do not indicate a problem.
  > ((106836 . 38433) (6098 . 29) (593 . 86) 441045 1034302 (27 . 1) (19
  > . 0) (8971 . 416))
  > Loading ..../lisp/loaddefs.el (source)...
  > Wrong number of arguments: autoload, 12
  > make[2]: *** [bootstrap-emacs] Error 255
  > make[2]: Target `bootstrap' not remade because of errors.
  > make[1]: *** [bootstrap-build] Error 2
  > make: *** [bootstrap] Error 2

Please remove lisp/loaddefs.el and do a cvs update and try again.

Re: 23.0.60; compilation fails with: Wrong number of arguments: autoload, 12

[Peter Dyballa]

Am 12.11.2007 um 14:53 schrieb Dan Nicolaescu:

> Please remove lisp/loaddefs.el and do a cvs update and try again.

Compiles now! (Even runs!)

--
Greetings

   Pete

Increase the size of your bike by at least *five* inches!

Re: 23.0.60; compilation fails with: Wrong number of arguments: autoload, 12

[Richard Wordingham]

Peter Dyballa wrote:
> Am 12.11.2007 um 14:53 schrieb Dan Nicolaescu:

>> Please remove lisp/loaddefs.el and do a cvs update and try again.

> Compiles now! (Even runs!)

It's still not working for me (checked out at 19:43 GMT 12/11/07).  I'm 
using MinGW on Windows XP SP2, with gcc Version 3.4.5.  I'm using GNU Make 
3.81 (tarball mingw32-make-3.81-2.tar.gz).

1. With the command 'mingw32-make SHELL=cmd.exe' the generated makefile 
fails at the first linking.  I have to convert the DOS line-ends to Unix 
line ends.

2. Having made that change, I then get the advice to 'make bootstrap'.  I 
issue the command:

mingw32-make SHELL=cmd.exe bootstrap

and get the final outputs:

Generating autoloads for kermit.el...
Generating autoloads for kermit.el...done
Generating autoloads for kmacro.el...
Generating autoloads for kmacro.el...done
Local variables entry is missing the suffix
mingw32-make[1]: *** [autoloads] Error -1
mingw32-make[1]: Leaving directory `C:/programs/emacs-unicode-2/lisp'
mingw32-make: *** [bootstrap-gmake] Error 2

It seems that the offending entry is in ldefs-boot.el:

;; Local Variables:
;; version-control: never
;; no-byte-compile: t
;; no-update-autoloads: t
;; End:

- but I can't see what is wrong with it.  Perhaps the fault is in the 
function hack-local-variables in files.el.

Richard. 

Re: 23.0.60; compilation fails with: Wrong number of arguments: autoload, 12

[Jason Rumney]

Richard Wordingham wrote:
> It's still not working for me (checked out at 19:43 GMT 12/11/07). 
> I'm using MinGW on Windows XP SP2, with gcc Version 3.4.5.  I'm using
> GNU Make 3.81 (tarball mingw32-make-3.81-2.tar.gz).
>
> 1. With the command 'mingw32-make SHELL=cmd.exe' the generated
> makefile fails at the first linking.  I have to convert the DOS
> line-ends to Unix line ends.
Please read nt/INSTALL carefully, including the "Preliminaries" section.

Tiro still can't usefully link emacs-unicode-2 on Windows

[Richard Wordingham]

Jason Rumney wrote, under subject heading 'Re: 23.0.60; compilation fails 
with: Wrong number of arguments: autoload, 12':
> Richard Wordingham wrote:
>> It's still not working for me (checked out at 19:43 GMT 12/11/07).
>> I'm using MinGW on Windows XP SP2, with gcc Version 3.4.5.  I'm using
>> GNU Make 3.81 (tarball mingw32-make-3.81-2.tar.gz).

>> 1. With the command 'mingw32-make SHELL=cmd.exe' the generated
>> makefile fails at the first linking.  I have to convert the DOS
>> line-ends to Unix line ends.

> Please read nt/INSTALL carefully, including the "Preliminaries" section.

If this is a reference to running 'cvs update -kb' in the nt directory, note 
that I found that I also had to run it in at least the leim\CXTERM-dic and 
lisp\international dictionaries.  Emacs then linked, and the makefile 
appeared to complete, but Emacs wouldn't start up.  I then downloaded the 
whole again using the command 
'cvs -z3 -d:pserver:anonymous@cvs.savannah.gnu.org:/sources/emacs co -kb -r 
emacs-unicode-2 emacs'.  (Note the '-kb'.)  I then made it cleanly (apart 
from numerous warnings during byte-compilation and messages saying that 
files could not be deleted), but again it would not run.  The only clue I 
have as to why is that when I run emacs.exe from the cmd.exe command line I 
get the message, ' Wrong type argument: stringp, nil'.  Could it be failing 
because I build it with none of the graphics options xpm, png, jpeg, tiff 
and gif?  (I didn't need them to rebuild a usable Emacs 22.1 from the source 
distribution.)

Incidentally, while lisp/version.el identifes the current version of 
emacs-unicode-2 as 23.0.60, nt/emacs.rc identifies it as 23.0.50.0, and this 
is the number stamped into the Windows executables I managed to generate.

Richard. 

Re: Tiro still can't usefully link emacs-unicode-2 on Windows

[Jason Rumney]

Richard Wordingham wrote:
> If this is a reference to running 'cvs update -kb' in the nt
> directory, note that I found that I also had to run it in at least the
> leim\CXTERM-dic and lisp\international dictionaries.
These seem like regressions that were previously fixed in Emacs 22.

> I then made it cleanly (apart from numerous warnings during
> byte-compilation and messages saying that files could not be deleted),
> but again it would not run.  The only clue I have as to why is that
> when I run emacs.exe from the cmd.exe command line I get the message,
> ' Wrong type argument: stringp, nil'.
This is triggered by a recent merge of w32-win.el. Some code that was
moved was previously commented out in the emacs-unicode-2 branch due to
not working, and the merge reenabled it. I have fixed the code now (the
problem was caused by the default fontset now being indicated by t
rather than nil).

Re: Tiro still can't usefully link emacs-unicode-2 on Windows

[Richard Wordingham]

Jason Rumney wrote:


> Richard Wordingham wrote:

>> I then made it cleanly (apart from numerous warnings during
>> byte-compilation and messages saying that files could not be deleted),
>> but again it would not run.

> This is triggered by a recent merge of w32-win.el. ... I have fixed the 
> code now (the
> problem was caused by the default fontset now being indicated by t
> rather than nil).

Thanks.  In my copy I've changed nil to t in the four forms calling 
set-fontset-font and now I can build and run Emacs 23.0.60.0.

Richard.