Re: makeinfo: missing file argument

[Peter Dyballa]

Am 5.4.2012 um 02:18 schrieb Kuroishi Mitsuo:

> (cd doc/emacs; make info)
> /usr/bin/makeinfo --force --enable-encoding -I /home/kuroishi/download/emacs/doc/emacs -o /home/kuroishi/download/emacs/doc/emacs/../../info/emacs
> makeinfo: missing file argument.

Here is the bug! I have, slightly edited:

(cd doc/emacs; make -k info)
/opt/local/bin/makeinfo --force --enable-encoding -I .../emacs-24.0.95/doc/emacs -o .../emacs-24.0.95/doc/emacs/../../info/emacs .../emacs-24.0.95/doc/emacs/emacs.texi
(cd doc/misc; make -k info)

The argument "<path to>/doc/emacs/emacs.texi" is missing. A possible reason might be that your terminal is limiting the line. Do you have already some variant of GNU Emacs installed? Then you could try this:

launch it, in some terminal or as X client. Then either cd in *shell* buffer to /home/kuroishi/download/emacs or open this directory in dired mode. If you want you can M-x compile RET and remove the whole text in minibuffer and substitute it with any of your configure commands, say

configure --without-x --prefix=/home/kuroishi/local

Press RETURN. Nice output, isn't it? When configure has finished you can type again M-x compile RET and change the text in minibuffer to:

make bootstrap

Press RETURN. If the failure happens again, then try this:

M-x compile RET <a bit of editing in minibuffer> make distclean RET

This should remove all files that do not belong to the distribution. Now perform the configure step again and start 'make bootstrap'. If this fails again, then it's likely that there is some failure in the build system. It could be in configure and some other GNU Emacs related files, or it is in your FreeBSD system. Although I am not using FreeBSD but Mac OS X, which is a bit related to FreeBSD or NetBSD, you can send me your config.log file (pay attention here: it could be very small, 1...2 KB, then I would need the config.log~ file with more than 1 MB size) /privately/ and I can try to examine it and find a possible cause.

Do you have the file ~/download/emacs/doc/emacs/emacs.texi?

--
Greetings

Pete

"By filing this bug report you have challenged the honor of my family. Prepare to die!"
- Answer from Klingon software developer

[Kuroishi Mitsuo]

Message-id: <FD9ADD51-3915-4817...@Web.DE>
From: Peter Dyballa <Peter_...@Web.DE>
Subject: Re: makeinfo: missing file argument
Date: Thu, 5 Apr 2012 11:23:36 +0200

>
> > (cd doc/emacs; make info)
> > /usr/bin/makeinfo --force --enable-encoding -I /home/kuroishi/download/emacs/doc/emacs -o /home/kuroishi/download/emacs/doc/emacs/../../info/emacs
> > makeinfo: missing file argument.
>
> Here is the bug! I have, slightly edited:
>
> (cd doc/emacs; make -k info)
> /opt/local/bin/makeinfo --force --enable-encoding -I .../emacs-24.0.95/doc/emacs -o .../emacs-24.0.95/doc/emacs/../../info/emacs .../emacs-24.0.95/doc/emacs/emacs.texi
> (cd doc/misc; make -k info)
>
> The argument "<path to>/doc/emacs/emacs.texi" is missing. A
> possible reason might be that your terminal is limiting the
> line.

OK. I understood the difference as you mentioned I'm not sure my
terminal isn't good though.

> Do you have already some variant of GNU Emacs installed? Then you could try this:
>
> launch it, in some terminal or as X client. Then either cd
in *shell* buffer to /home/kuroishi/download/emacs or open
this directory in dired mode. If you want you can M-x
compile RET and remove the whole text in minibuffer and
substitute it with any of your configure commands, say
>
> configure --without-x --prefix=/home/kuroishi/local
>
> Press RETURN. Nice output, isn't it? When configure has
> finished you can type again M-x compile RET and change the
> text in minibuffer to:
>
> make bootstrap
>
> Press RETURN. If the failure happens again, then try this:
>
> M-x compile RET <a bit of editing in minibuffer> make distclean RET
>
> This should remove all files that do not belong to the
> distribution. Now perform the configure step again and start
> 'make bootstrap'.

You mean it needs to try to do entire build process in another
Emacs or so, right?

I tried building, make distclean, configure and make bootstrap,
as well but didn't go well again.

> If this fails again, then it's likely that there is some
failure in the build system. It could be in configure and
some other GNU Emacs related files, or it is in your FreeBSD
system. Although I am not using FreeBSD but Mac OS X, which
is a bit related to FreeBSD or NetBSD, you can send me your
config.log file (pay attention here: it could be very small,
1...2 KB, then I would need the config.log~ file with more
than 1 MB size) /privately/ and I can try to examine it and
find a possible cause.

OK. I'll send it to you. Thanks for your help.

> Do you have the file ~/download/emacs/doc/emacs/emacs.texi?

Yes, like below.

% ls -la ~/download/emacs/doc/emacs/emacs.texi
-rw-r--r-- 1 kuroishi staff 71616 4 4 10:17 /home/kuroishi/download/emacs/doc/emacs/emacs.texi

--
Kuroishi Mitsuo

[Peter Dyballa]

Am 5.4.2012 um 15:22 schrieb Kuroishi Mitsuo:

> You mean it needs to try to do entire build process in another
> Emacs or so, right?

No, it was just meant to use a build environment that is not polluted or limited by some terminal emulation programme. This programme can be the culprit. Or a setting. Settings can be in your shell or they can be for the programme (or X client). This is much too hard to debug. So some GNU Emacs is a perfect choice.

> Do you have the file ~/download/emacs/doc/emacs/emacs.texi?
>
> Yes, like below.
>
> % ls -la ~/download/emacs/doc/emacs/emacs.texi
> -rw-r--r-- 1 kuroishi staff 71616 4 4 10:17 /home/kuroishi/download/emacs/doc/emacs/emacs.texi

Has the same size as mine – there must be some bug in the build system.

Since the docs are not really necessary to run GNU Emacs, you could try to compile with 'make -k bootstrap' or just 'make -k' if the bootstrap argument makes make very picky and not forgive errors in this early stage. Then make will not stop when it sees an error. This way the src/emacs binary could be built. And it is, without installation, fully usable!

--
Greetings

Pete

Bake pizza not war!

[Glenn Morris]

Kuroishi Mitsuo wrote:

> (cd doc/emacs; make info)
> /usr/bin/makeinfo --force --enable-encoding -I /home/kuroishi/download/emacs/doc/emacs -o /home/kuroishi/download/emacs/doc/emacs/../../info/emacs
> makeinfo: missing file argument.

The questions I would ask are:

What version of make are you using?
What is /bin/sh for you (eg bash, dash, etc)?
After you configure, what does doc/emacs/Makefile look like?
(best to send the whole thing as an attachment)

Just to see if it works, try passing the --without-makeinfo argument to
configure.

[Peter Dyballa]

Am 5.4.2012 um 18:26 schrieb Glenn Morris:

> Kuroishi Mitsuo wrote:
>
>> (cd doc/emacs; make info)
>> /usr/bin/makeinfo --force --enable-encoding -I /home/kuroishi/download/emacs/doc/emacs -o /home/kuroishi/download/emacs/doc/emacs/../../info/emacs
>> makeinfo: missing file argument.
>
> The questions I would ask are:
>
> What version of make are you using?

Config.log tells:

checking whether we are using GNU Make
result: no

> What is /bin/sh for you (eg bash, dash, etc)?

SHELL='/bin/bash' (from config.log too)


The make programme Mitsuo is using looks suspicious. I also wonder why configure uses lines like:

checking for getifaddrs
gcc -std=gnu99 -o conftest -g -O2 -Wl,-znocombreloc conftest.c >&5 -L/usr/local/lib
$? = 0
result: yes

Shouldn't ">&5" come last? On a few occasions configure reports:

gnome-config: not found
result: no
(checking for lgetfilecon in -lselinux)



The comparison also showed in *my* config.log file mostly the use of gcc, but there are also lines like:

checking sys/un.h usability
gcc -std=gnu99 -I/opt/local/include -L/opt/local/lib -c -g -O2 -I/opt/local/include conftest.c >&5
$? = 0
result: yes
checking sys/un.h presence
cc -E -no-cpp-precomp -I/opt/local/include -L/opt/local/lib conftest.c
$? = 0
result: yes
checking for sys/un.h
result: yes

At its end config.log has:

ac_cv_prog_CPP='cc -E -no-cpp-precomp -I/opt/local/include -L/opt/local/lib'
...
CC='gcc -std=gnu99 -I/opt/local/include -L/opt/local/lib'
CPP='cc -E -no-cpp-precomp -I/opt/local/include -L/opt/local/lib'

This looks also suspicious.


I must say that I don't know FreeBSD well enough to be more helpful in future. Looking into doc/emacs/Makefile (and comparing it with mine) is certainly a good option. Maybe we can see the cause why no file name is passed to makeinfo. Or are these lines obviously suspicious:

EMACS_XTRA= \
${srcdir}/emacs-xtra.texi \
$(srcdir)/arevert-xtra.texi \
$(srcdir)/cal-xtra.texi \
$(srcdir)/dired-xtra.texi \
$(srcdir)/picture-xtra.texi \
$(srcdir)/emerge-xtra.texi \
$(srcdir)/vc-xtra.texi \
$(srcdir)/vc1-xtra.texi \
$(srcdir)/fortran-xtra.texi \
$(srcdir)/msdog-xtra.texi

EMACSSOURCES= \
${srcdir}/emacs.texi \
${srcdir}/emacsver.texi \
...
${srcdir}/kmacro.texi \
$(EMACS_XTRA)

Usually the syntax is:

MACRO <white space> =

--
Greetings

Pete

We have to expect it, otherwise we would be surprised.

[Kuroishi Mitsuo]

Thanks for your replies.


Message-id: <1F227D90-7C7A-48CF...@Web.DE>

From: Peter Dyballa <Peter_...@Web.DE>
Subject: Re: makeinfo: missing file argument

Date: Thu, 5 Apr 2012 15:31:51 +0200

> > You mean it needs to try to do entire build process in another
> > Emacs or so, right?
>
> No, it was just meant to use a build environment that is not
polluted or limited by some terminal emulation
programme. This programme can be the culprit. Or a
setting. Settings can be in your shell or they can be for
the programme (or X client). This is much too hard to
debug. So some GNU Emacs is a perfect choice.

I now know.


Message-id: <2my5qa4...@fencepost.gnu.org>
From: Glenn Morris <r...@gnu.org>

Subject: Re: makeinfo: missing file argument

Date: Thu, 05 Apr 2012 12:26:29 -0400

> After you configure, what does doc/emacs/Makefile look like?
> (best to send the whole thing as an attachment)

copying below.

> Just to see if it works, try passing the --without-makeinfo argument to
> configure.

I'll try to build with -k option or --without-makeinfo.

--
Kuroishi Mitsuo


*
* doc/emacs/Makefile
*

#### Makefile for the Emacs Manual

# Copyright (C) 1994, 1996-2012 Free Software Foundation, Inc.

# This file is part of GNU Emacs.

# GNU Emacs is free software: you can redistribute it and/or modify
# it under the terms of the GNU General Public License as published by
# the Free Software Foundation, either version 3 of the License, or
# (at your option) any later version.

# GNU Emacs is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.

# You should have received a copy of the GNU General Public License
# along with GNU Emacs. If not, see <http://www.gnu.org/licenses/>.

SHELL = /bin/sh

# Where to find the source code. $(srcdir) will be the doc/emacs subdirectory
# of the source tree. This is set by configure's `--srcdir' option.
srcdir=/home/kuroishi/download/emacs/doc/emacs

# Only for make dist.
version=24.0.95

## Where the output files go.
## Note that the setfilename command in the .texi files assumes this.
## This is a bit funny. Because the info files are in the
## distribution tarfiles, they are always made in $scrdir/../../info,
## even for out-of-tree builds.
infodir = $(srcdir)/../../info
# Directory with the (customized) texinfo.tex file.
texinfodir = $(srcdir)/../misc

# The makeinfo program is part of the Texinfo distribution.
# Use --force so that it generates output even if there are errors.
MAKEINFO = /usr/bin/makeinfo
MAKEINFO_OPTS = --force --enable-encoding -I $(srcdir)

TEXI2DVI = texi2dvi
TEXI2PDF = texi2pdf
DVIPS = dvips


ENVADD = TEXINPUTS="$(srcdir):$(texinfodir):$(TEXINPUTS)" \
MAKEINFO="$(MAKEINFO) $(MAKEINFO_OPTS)"

EMACS_XTRA= \
${srcdir}/emacs-xtra.texi \
$(srcdir)/arevert-xtra.texi \
$(srcdir)/cal-xtra.texi \
$(srcdir)/dired-xtra.texi \
$(srcdir)/picture-xtra.texi \
$(srcdir)/emerge-xtra.texi \
$(srcdir)/vc-xtra.texi \
$(srcdir)/vc1-xtra.texi \
$(srcdir)/fortran-xtra.texi \
$(srcdir)/msdog-xtra.texi

EMACSSOURCES= \
${srcdir}/emacs.texi \
${srcdir}/emacsver.texi \

${srcdir}/doclicense.texi \
${srcdir}/gpl.texi \
${srcdir}/screen.texi \
${srcdir}/commands.texi \
${srcdir}/entering.texi \
${srcdir}/basic.texi \
${srcdir}/mini.texi \
${srcdir}/m-x.texi \
${srcdir}/help.texi \
${srcdir}/mark.texi \
${srcdir}/killing.texi \
${srcdir}/regs.texi \
${srcdir}/display.texi \
${srcdir}/search.texi \
${srcdir}/fixit.texi \
${srcdir}/files.texi \
${srcdir}/buffers.texi \
${srcdir}/windows.texi \
${srcdir}/frames.texi \
${srcdir}/mule.texi \
${srcdir}/modes.texi \
${srcdir}/indent.texi \
${srcdir}/text.texi \
${srcdir}/programs.texi \
${srcdir}/building.texi \
${srcdir}/maintaining.texi \
${srcdir}/abbrevs.texi \
${srcdir}/sending.texi \
${srcdir}/rmail.texi \
${srcdir}/dired.texi \
${srcdir}/calendar.texi \
${srcdir}/misc.texi \
${srcdir}/package.texi \
${srcdir}/custom.texi \
${srcdir}/trouble.texi \
${srcdir}/cmdargs.texi \
${srcdir}/xresources.texi \
${srcdir}/anti.texi \
${srcdir}/macos.texi \
${srcdir}/msdog.texi \
${srcdir}/gnu.texi \
${srcdir}/glossary.texi \
${srcdir}/ack.texi \
${srcdir}/kmacro.texi \
$(EMACS_XTRA)

## This seems pointless. The info/ directory exists in both the
## repository and the release tarfiles. We do not use any
## equivalent of mkdir -p/install-sh -d, so this is not a general
## solution anyway. The second test -d is for parallel builds.
mkinfodir = @test -d ${infodir} || mkdir ${infodir} || test -d ${infodir}

.PHONY: info dvi html pdf ps

info: $(infodir)/emacs
dvi: emacs.dvi
html: emacs.html
pdf: emacs.pdf
ps: emacs.ps

# Note that all the Info targets build the Info files in srcdir.
# There is no provision for Info files to exist in the build directory.
# In a distribution of Emacs, the Info files should be up to date.

$(infodir)/emacs: ${EMACSSOURCES}
$(mkinfodir)
$(MAKEINFO) $(MAKEINFO_OPTS) -o $@ $<

emacs.dvi: ${EMACSSOURCES}
$(ENVADD) $(TEXI2DVI) $<

emacs.ps: emacs.dvi
$(DVIPS) -o $@ $<

emacs.pdf: ${EMACSSOURCES}
$(ENVADD) $(TEXI2PDF) $<

emacs.html: ${EMACSSOURCES}
$(MAKEINFO) $(MAKEINFO_OPTS) --html -o $@ $<

emacs-xtra.dvi: $(EMACS_XTRA)
$(ENVADD) $(TEXI2DVI) $<

emacs-xtra.ps: emacs-xtra.dvi
$(DVIPS) -o $@ $<

emacs-xtra.pdf: $(EMACS_XTRA)
$(ENVADD) $(TEXI2PDF) $<

.PHONY: mostlyclean clean distclean maintainer-clean infoclean

## Temp files.
mostlyclean:
rm -f *.aux *.log *.toc *.cp *.cps *.fn *.fns *.ky *.kys \
*.op *.ops *.pg *.pgs *.tp *.tps *.vr *.vrs

## Products not in the release tarfiles.
clean: mostlyclean
rm -f emacs.dvi emacs-xtra.dvi emacs.pdf emacs-xtra.pdf \
emacs.ps emacs-xtra.ps
rm -rf emacs.html/
rm -f emacs-manual-${version}.tar*

distclean: clean

## In the standalone tarfile, the clean rule runs this.
infoclean:
-cd $(infodir) && rm -f emacs emacs-[1-9] emacs-[1-9][0-9]

maintainer-clean: distclean infoclean

.PHONY: dist

## Make a standalone tarfile of the Emacs manual sources.
## The [c] is a dumb way to prevent configure expanding it.
dist:
rm -rf emacs-manual-${version}
mkdir emacs-manual-${version}
cp ${srcdir}/*.texi ${texinfodir}/texinfo.tex \
${srcdir}/ChangeLog* emacs-manual-${version}/
sed -e 's/@sr[c]dir@/./' -e 's/^\(texinfodir *=\).*/\1 ./' \
-e 's/^\(infodir *=\).*/\1 ./' -e 's/^\(clean:.*\)/\1 infoclean/' \
-e "s/@ver[s]ion@/${version}/" \
${srcdir}/Makefile.in > emacs-manual-${version}/Makefile
tar -cf emacs-manual-${version}.tar emacs-manual-${version}
rm -rf emacs-manual-${version}

### Makefile ends here

[Glenn Morris]

Kuroishi Mitsuo wrote:

> > After you configure, what does doc/emacs/Makefile look like?
> > (best to send the whole thing as an attachment)
>
> copying below.

I don't see anything wrong with that.
You did not say what version of Make you are using. If you have GNU Make
available, please try with that.

Also, please try the following test:
Save the end of this mail to a file called `Makefile'.
Then run:

touch file1 file2

Then what is printed if you run the commands:

make one

and

make two

?


Test Makefile follows:
-----------------

SHELL = /bin/sh

var1= \
file1 \
file2

var2= file1 \
file2

one: ${var1}
@echo $@ $<

two: ${var2}
@echo $@ $<

[Kuroishi Mitsuo]

Message-id: <4i7gxt5...@fencepost.gnu.org>

From: Glenn Morris <r...@gnu.org>
Subject: Re: makeinfo: missing file argument

Date: Fri, 06 Apr 2012 03:06:59 -0400

> > > After you configure, what does doc/emacs/Makefile look like?
> > > (best to send the whole thing as an attachment)
> >
> > copying below.
>
> I don't see anything wrong with that.
> You did not say what version of Make you are using.

sorry. It's like this.

% LC_ALL=C gmake --version
GNU Make 3.81
Copyright (C) 2006 Free Software Foundation, Inc.
This is free software; see the source for copying conditions.
There is NO warranty; not even for MERCHANTABILITY or FITNESS FOR A
PARTICULAR PURPOSE.

This program built for i386-portbld-freebsd6.2

> If you have GNU Make
> available, please try with that.

And I tried that. Is it OK?
If I do a something wrong with it, let me know about it.

Thanks,

--
Kuroishi Mitsuo


|
|~/tmp/2012-04-06% ls
|Makefile
|~/tmp/2012-04-06% cat Makefile

|SHELL = /bin/sh
|
|var1= \
| file1 \
| file2
|
|var2= file1 \
| file2
|
|one: ${var1}
| @echo $@ $<
|
|two: ${var2}
| @echo $@ $<

|~/tmp/2012-04-06% touch file1 file2
|~/tmp/2012-04-06% ls
|Makefile file1 file2
|~/tmp/2012-04-06% make one
|one
|~/tmp/2012-04-06% gmake one
|one file1
|~/tmp/2012-04-06% make two
|two
|~/tmp/2012-04-06% gmake two
|two file1
|~/tmp/2012-04-06%
|

[Glenn Morris]

Kuroishi Mitsuo wrote:

> sorry. It's like this.
>
> % LC_ALL=C gmake --version
> GNU Make 3.81

Thanks. Can you build Emacs with gmake? I think that goes something like:

export MAKE=gmake
./configure [...]
gmake

Also, what is the version of the "make" program?

> |~/tmp/2012-04-06% make one
> |one
> |~/tmp/2012-04-06% gmake one
> |one file1
> |~/tmp/2012-04-06% make two
> |two
> |~/tmp/2012-04-06% gmake two
> |two file1

It looks like whatever "make" is, it does not understand the $<
construct, and silently replaces it with nothing.

One final test, to check that the continuation lines are not the issue:

touch file1
make one

Using test Makefile as follows:
-----------------------

SHELL = /bin/sh

var1= file1

[Glenn Morris]

Glenn Morris wrote:

> It looks like whatever "make" is, it does not understand the $<
> construct, and silently replaces it with nothing.

Huh, according to

http://www.gnu.org/savannah-checkouts/gnu/autoconf/manual/autoconf-2.68/html_node/_0024_003c-in-Ordinary-Make-Rules.html

"$<" is not portable in ordinary rules.

Emacs has always used '$<' in its Make rules, but maybe all the other
instances were in non-ordinary rules. So the Emacs Makefiles should
probably be fixed to not use '$<' in ordinary rules, like the ones in
doc/*/Makefile.

Thanks for helping to track this down!

[Kuroishi Mitsuo]

Message-id: <k9fwchm...@fencepost.gnu.org>

From: Glenn Morris <r...@gnu.org>
Subject: Re: makeinfo: missing file argument

Date: Fri, 06 Apr 2012 03:32:08 -0400

> export MAKE=gmake
> ./configure [...]
> gmake

After "make distclean (not gmake)", I did like below,

% export MAKE=gmake
% ./configure --without-x --prefix=/home/kuroishi/local
% gmake bootstrap

And it seems to overcome the problem but stopped building on the
way unfortunately.

attating the log at the end.

> Also, what is the version of the "make" program?
>
> > |~/tmp/2012-04-06% make one
> > |one
> > |~/tmp/2012-04-06% gmake one
> > |one file1
> > |~/tmp/2012-04-06% make two
> > |two
> > |~/tmp/2012-04-06% gmake two
> > |two file1
>

> It looks like whatever "make" is, it does not understand the $<
> construct, and silently replaces it with nothing.

Hmm, I don't konw how to show the version.
I suppose it's the system(Freebsd 6.2) default varsion probably.

> One final test, to check that the continuation lines are not the issue:
>
> touch file1
> make one
>
> Using test Makefile as follows:
> -----------------------
>
> SHELL = /bin/sh
>
> var1= file1
>
> one: ${var1}
> @echo $@ $<

OK. I tried.

|
|~/tmp/2012-04-06.02% cat Makefile

|SHELL = /bin/sh
|
|var1= file1
|
|one: ${var1}
| @echo $@ $<

|~/tmp/2012-04-06.02% ls
|Makefile
|~/tmp/2012-04-06.02% touch file1
|~/tmp/2012-04-06.02% ls
|Makefile file1
|~/tmp/2012-04-06.02% make one
|one
|~/tmp/2012-04-06.02% gmake one
|one file1
|~/tmp/2012-04-06.02%
|

Thanks for your help.

--
Kuroishi Mitsuo

:
:
Generating autoloads for mh-xface.el...
Generating autoloads for mh-xface.el...done
Saving file /home/kuroishi/download/emacs/lisp/net/tramp-loaddefs.el...
Wrote /home/kuroishi/download/emacs/lisp/net/tramp-loaddefs.el
(No changes need to be saved)
cd /home/kuroishi/download/emacs/lisp && chmod +w ps-print.el emulation/tpu-edt.el emacs-lisp/cl-loaddefs.el mail/rmail.el dired.el ibuffer.el htmlfontify.el emacs-lisp/eieio.el
cd /home/kuroishi/download/emacs/lisp; subdirs=`(find . -type d -print)`; for file in $subdirs; do case $file in */.* | */.*/* | */=* | */obsolete | */term ) ;; *) wins="$wins $file" ;; esac; done; \
echo Directories: $wins; \
EMACSLOADPATH=/home/kuroishi/download/emacs/lisp LC_ALL=C /home/kuroishi/download/emacs/src/bootstrap-emacs -batch --no-site-file --no-site-lisp -l autoload --eval '(setq generated-autoload-file "/home/kuroishi/download/emacs/lisp/loaddefs.el")' -f batch-update-autoloads $wins
Directories: . ./calc ./calendar ./cedet ./cedet/ede ./cedet/semantic ./cedet/semantic/analyze ./cedet/semantic/bovine ./cedet/semantic/decorate ./cedet/semantic/symref ./cedet/semantic/wisent ./cedet/srecode ./emacs-lisp ./emulation ./erc ./eshell ./gnus ./international ./language ./mail ./mh-e ./net ./nxml ./org ./play ./progmodes ./textmodes ./url ./vc
Wrote /home/kuroishi/download/emacs/lisp/loaddefs.el
Loading /home/kuroishi/download/emacs/lisp/vc/vc-bzr.el (source)...
Generating autoloads for align.el...
Generating autoloads for align.el...done
Generating autoloads for allout-widgets.el...
Generating autoloads for allout-widgets.el...done
Generating autoloads for allout.el...
Generating autoloads for allout.el...done
Generating autoloads for ansi-color.el...
Generating autoloads for ansi-color.el...done
Generating autoloads for apropos.el...
Generating autoloads for apropos.el...done
Generating autoloads for arc-mode.el...
Generating autoloads for arc-mode.el...done
Generating autoloads for array.el...
Generating autoloads for array.el...done
Generating autoloads for autoarg.el...
Generating autoloads for autoarg.el...done
Generating autoloads for autoinsert.el...
Generating autoloads for autoinsert.el...done
Generating autoloads for autorevert.el...
Generating autoloads for autorevert.el...done
Generating autoloads for avoid.el...
Generating autoloads for avoid.el...done
Generating autoloads for battery.el...
Generating autoloads for battery.el...done
Generating autoloads for bookmark.el...
Generating autoloads for bookmark.el...done
Generating autoloads for bs.el...
Generating autoloads for bs.el...done
Generating autoloads for calculator.el...
Generating autoloads for calculator.el...done
Generating autoloads for cdl.el...
Generating autoloads for cdl.el...done
Generating autoloads for chistory.el...
Generating autoloads for chistory.el...done
Generating autoloads for cmuscheme.el...
Generating autoloads for cmuscheme.el...done
Generating autoloads for color.el...
Generating autoloads for color.el...done
Generating autoloads for comint.el...
Generating autoloads for comint.el...done
Generating autoloads for completion.el...
Generating autoloads for completion.el...done
Generating autoloads for cus-dep.el...
Generating autoloads for cus-dep.el...done
Generating autoloads for cus-edit.el...
Generating autoloads for cus-edit.el...done
Generating autoloads for cus-theme.el...
Generating autoloads for cus-theme.el...done
Generating autoloads for dabbrev.el...
Generating autoloads for dabbrev.el...done
Generating autoloads for delim-col.el...
Generating autoloads for delim-col.el...done
Generating autoloads for delsel.el...
Generating autoloads for delsel.el...done
Generating autoloads for descr-text.el...
Generating autoloads for descr-text.el...done
Generating autoloads for desktop.el...
Generating autoloads for desktop.el...done
Generating autoloads for dframe.el...
Generating autoloads for dframe.el...done
Making generated-autoload-file local to *autoload-file* while let-bound!
Generating autoloads for dired-aux.el...
Error: (error "Running bzr status dired.el...FAILED (status 1)")
Running bzr status ../../../../../a/jc-nhfs/export/nfs/home/kuroishi/download/emacs/lisp/dired.el...FAILED (status 1)
gmake[3]: *** [autoloads] エラー 255
gmake[3]: ディレクトリ `/a/jc-nhfs/export/nfs/home/kuroishi/download/emacs/lisp' から出ます
gmake[2]: *** [/home/kuroishi/download/emacs/src/../lisp/loaddefs.el] エラー 2
gmake[2]: ディレクトリ `/a/jc-nhfs/export/nfs/home/kuroishi/download/emacs/src' から出ます
gmake[1]: *** [src] エラー 2
gmake[1]: ディレクトリ `/a/jc-nhfs/export/nfs/home/kuroishi/download/emacs' から出ます
gmake: *** [bootstrap] エラー 2

[Peter Dyballa]

Am 6.4.2012 um 10:01 schrieb Kuroishi Mitsuo:

> Error: (error "Running bzr status dired.el...FAILED (status 1)")
> Running bzr status ../../../../../a/jc-nhfs/export/nfs/home/kuroishi/download/emacs/lisp/dired.el...FAILED (status 1)
> gmake[3]: *** [autoloads] エラー 255

You need to be on-line, the bzr server needs to be on-line... This is, IMO, not really a build failure – it looks good. The other cure is to use a different method to check-out the sources and update with 'bzr pull'. Then no on-line connection is needed at compile time. I think this method is (or was) also described on the Savannah server.


BTW, Mitsuo, you can use

env LC_ALL=C MAKE=gmake ./configure ...
env LC_ALL=C gmake ...

in order to get output messages in English.

--
Greetings

Pete

"Klingons do not believe in indentation - except perhaps in the skulls of their project managers."

[Kuroishi Mitsuo]

Message-id: <40EE573D-AA6A-4130...@Web.DE>
From: Peter Dyballa <Peter_...@Web.DE>

Subject: Re: makeinfo: missing file argument

Date: Fri, 6 Apr 2012 10:59:32 +0200

> > Error: (error "Running bzr status dired.el...FAILED (status 1)")
> > Running bzr status ../../../../../a/jc-nhfs/export/nfs/home/kuroishi/download/emacs/lisp/dired.el...FAILED (status 1)
> > gmake[3]: *** [autoloads] エラー 255
>
> You need to be on-line, the bzr server needs to be

> > on-line... This is, IMO, not really a build failure �#| it

> > looks good. The other cure is to use a different method to
> > check-out the sources and update with 'bzr pull'. Then no
> > on-line connection is needed at compile time. I think this
> > method is (or was) also described on the Savannah server.

yes, it was my own environment problem.
Now, I'm happy to be able to build the lastest Emacs.

> BTW, Mitsuo, you can use
>
> env LC_ALL=C MAKE=gmake ./configure ...
> env LC_ALL=C gmake ...
>
> in order to get output messages in English.

OK.

Thank you very much.

--
Kuroishi Mitsuo

[Peter Dyballa]

Am 6.4.2012 um 09:37 schrieb Glenn Morris:

> Huh, according to
>
> http://www.gnu.org/savannah-checkouts/gnu/autoconf/manual/autoconf-2.68/html_node/_0024_003c-in-Ordinary-Make-Rules.html
>
> "$<" is not portable in ordinary rules.

I tried to use the bsdmake programme in Mac OS X, which also won't tell its version, and it had no problems to build the docs – but ended with executing this:

(cd doc/lispintro; bsdmake -j 1 -j 1 -j 1 -j 1 info)

(To have a clear sight I invoked 'bsdmake -j 1 bootstrap'.) The compilation of objects and binaries with GCC never started...

--
Greetings

Pete

»¿ʇı̣ əsnqɐ ʇ,uɐɔ noʎ ɟı̣
ɓuı̣ɥʇʎuɐ sı̣ pooɓ ʇɐɥʍ«

[Eli Zaretskii]

> From: Peter Dyballa <Peter_...@Web.DE>

> Date: Fri, 6 Apr 2012 10:59:32 +0200

> Cc: help-gn...@gnu.org

>
> > Error: (error "Running bzr status dired.el...FAILED (status 1)")
> > Running bzr status ../../../../../a/jc-nhfs/export/nfs/home/kuroishi/download/emacs/lisp/dired.el...FAILED (status 1)
> > gmake[3]: *** [autoloads] エラー 255
>
> You need to be on-line, the bzr server needs to be on-line...

??? Only if that's a lightweight checkout, something that is not
recommended. For normal checkouts, "bzr status" is a local operation
and doesn't need to hit the network.

[Kuroishi Mitsuo]

sorry for the fuss.

Message-id: <20120406.170147.211115...@iij.ad.jp>
From: Kuroishi Mitsuo <kuro...@iij.ad.jp>

Subject: Re: makeinfo: missing file argument

Date: Fri, 06 Apr 2012 17:01:47 +0900 (JST)

> After "make distclean (not gmake)", I did like below,
>
> % export MAKE=gmake
> % ./configure --without-x --prefix=/home/kuroishi/local
> % gmake bootstrap
>
> And it seems to overcome the problem but stopped building on the
> way unfortunately.
>
> attating the log at the end.

I'm installing bzr and pythong to ~/local because of various
reasons and it's the cause.

So I did like this

% PYTHONPATH=/home/kuroishi/local/lib/python gmake install

then, it went well and I'm happy to be able to build the latest Emacs now.

[Glenn Morris]

I believe that the emacs-24 branch should now build with a non-GNU make.
It would be great if you could test that it works.

bzr branch bzr://bzr.savannah.gnu.org/emacs/emacs-24

(or however you fetched the trunk)

[Peter Dyballa]

Am 7.4.2012 um 21:26 schrieb Glenn Morris:

> I believe that the emacs-24 branch should now build with a non-GNU make.

I have on intel Mac OS X 10.6.8 (Leopard) a problem with bsdmake: after temacs was compiled, a failure happens (my login shell is tcsh 6.15.00 (Astron) 2007-03-03 (x86_64-apple-darwin) options wide,nls,dl,al,kan,sm,rh,color,filec):

cd ../lisp; bsdmake -j 1 -j 1 -j 1 -j 1 update-subdirs
cd .../emacs-24.0.95/lisp; subdirs=`(find . -type d -print)`; for file in $subdirs; do case $file in */.* | */.*/* | */=* | */cedet* ) ;; *) wins="$wins $file" ;; esac; done; for file in $wins; do .../emacs-24.0.95/update-subdirs $file; done;

(find . -type d -print)

if test "no" = "yes"; then ln -f temacs bootstrap-emacs; else `/bin/pwd`/temacs --batch --load loadup bootstrap || exit 1; mv -f emacs bootstrap-emacs; fi
/bin/pwd
sh: line 2: .../emacs-24.0.95/lisp/temacs: No such file or directory
*** Error code 1
1 error
*** Error code 2
1 error
*** Error code 2
1 error

(Since CWD is lisp the path to temacs cannot be `/bin/pwd`/temacs.) When I then type g in the *compilation* buffer (to repeat the compilation) temacs is deleted and not made again. The compilation finishes with no error, but emacs is not built and the ELC files are not byte-compiled. So it is not really a bootstrap that starts again or is continued.


With GNU Make 3.81 this happens after temacs was built:

cd ../lisp; make update-subdirs
cd .../emacs-24.0.95/lisp; subdirs=`(find . -type d -print)`; for file in $subdirs; do case $file in */.* | */.*/* | */=* | */cedet* ) ;; *) wins="$wins $file" ;; esac; done; \
for file in $wins; do \
.../emacs-24.0.95/update-subdirs $file; \
done;
if test "no" = "yes"; then \
ln -f temacs bootstrap-emacs; \
else \
`/bin/pwd`/temacs --batch --load loadup bootstrap || exit 1; \
mv -f emacs bootstrap-emacs; \
fi
Loading loadup.el (source)...

".../" stands for the absolute path with my home directory.


--
Greetings

Pete

A common mistake that people make when trying to design something completely foolproof is to underestimate the ingenuity of complete fools.
- Douglas Adams, »Mostly Harmless«

[Kuroishi Mitsuo]

Message-id: <cisjgff...@fencepost.gnu.org>
From: Glenn Morris <r...@gnu.org>

Subject: Re: makeinfo: missing file argument

Date: Sat, 07 Apr 2012 15:26:41 -0400

> I believe that the emacs-24 branch should now build with a non-GNU make.

> It would be great if you could test that it works.
>
> bzr branch bzr://bzr.savannah.gnu.org/emacs/emacs-24

I did bzr like above. And configure, make bootstrap, make
install. (not gmake)

It has no problem and works well. Great!

Regards,

--
Kuroishi Mitsuo