Re: adding to CFLAGS for all objects in subdir
> "Jeremy" == Jeremy Slade <[EMAIL PROTECTED]> writes: Jeremy> In a project, I want all objects (both C and C++) in a certain Jeremy> subdirectory to include additional flags (defined by Jeremy> PERL_CFLAGS) in the compile line, but I don't want those flags Jeremy> used to build object in other directories. Use AM_CFLAGS. Jeremy> CFLAGS = $(PERL_CFLAGS) CFLAGS is a user variable. Don't set it yourself. Jeremy> CPPFLAGS=$(PERL_CFLAGS) CPPFLAGS is also a user variable. Jeremy> # Must create the $perllibarchdir/auto directory to enable auto-loading Jeremy> # Also have to chmod Module.sl to be executable after install -- Jeremy> # not initially made executable since it is a _DATA object. Jeremy> install-data-local: Jeremy> $(mkinstalldirs) $(DESTDIR)$(perllibarchdir)/auto Jeremy> chmod +x $(DESTDIR)$(perllibarchdir)/Module.sl That's a bit ugly :-(. I wouldn't object to a cleaner way to build Perl modules. Jeremy> Please tell me there's an easier way to do this... I thought everybody who wrote Perl modules used Perl's MakeMaker mechanism (isn't that what it is called? I haven't looked at it since 1996...). Tom
Re: "unknown extension in source" perl error (with patch)
> "edward" == edward <[EMAIL PROTECTED]> writes:
edward> + if (!defined($extension_map{$extension}))
edward> + {
edward> + &am_error ("unknown extension for `$full' in $var");
edward> + next;
edward> + }
We can't do this. We intentionally don't warn in this case. See the
comment later in that same function:
# No error message here. Used to have one, but it was
# very unpopular.
This error message was the number one bug report generator for a
while.
I agree that something should be done though.
Tom
adding to CFLAGS for all objects in subdir
In a project, I want all objects (both C and C++) in a certain subdirectory to include additional flags (defined by PERL_CFLAGS) in the compile line, but I don't want those flags used to build object in other directories. I'm not sure how to accomplish this. If I do this in the Makefile.am: CFLAGS = $(PERL_CFLAGS) Doesn't that clobber the standard CFLAGS, such as '-g -O2'? What I'm ultimately trying to do is build a shared library that can be dynamically loaded into a perl interpreter -- perhaps this is a problem someone has already solved? If anyone has pointers, I'd appreciate it. I currently have a Makefile.am that looks something like this: -- perllib_DATA = Module.pm perllibarch_DATA = Module.sl # Do noinst libtool library to handle building the objects noinst_LTLIBRARIES = lib_Module.la lib_Module_la_SOURCES = Module.C INCLUDES = -I$(PERL_ARCHLIBCORE) CPPFLAGS=$(PERL_CFLAGS) Module.sl: Module.lo ld -b -E -o $@ $^ # Must create the $perllibarchdir/auto directory to enable auto-loading # Also have to chmod Module.sl to be executable after install -- # not initially made executable since it is a _DATA object. install-data-local: $(mkinstalldirs) $(DESTDIR)$(perllibarchdir)/auto chmod +x $(DESTDIR)$(perllibarchdir)/Module.sl -- I have other macros defined that set perllib, perllibarch, PERL_CFLAGS, etc. Ideally, those same macros would change the CFLAGS, INCLUDES, etc for this specific directory to include all of the perl-related stuff. My platform is HP-UX 10.20, gcc-2.95.2, perl 5.6.0, autoconf-2.49c, automake-1.4b, libtool-1.3.5 Please tell me there's an easier way to do this... Thanks, Jeremy -- Jeremy Slade Hewlett-Packard Company VLSI Technology Center - http://cpus.hp.com/ [EMAIL PROTECTED](970) 898-7881
--Werror and --add-missing don't work toghether
Hello! CVS Automake uses am_line_error() to inform the user that it's installing files. As a result, --Werror causes Automake to exit after it installs the first file: $ automake --Werror --add-missing; echo $? automake: Makefile.am: installing `./INSTALL' 2 $ automake --Werror --add-missing; echo $? automake: Makefile.am: required file `./NEWS' not found 2 Sorry, no patch. Regards, Pavel Roskin
Re: plea for case-insensitive, but case preserving (was: Re: 45-fix-yacc2-test.patch)
> "Lars" == Lars Hecking <[EMAIL PROTECTED]> writes: Lars> edward writes: >> hi akim, >> >> this is a plea to avoid using constructs like the "cp Makefile.am >> Makefile.AM" below. this kills us cygwin/windows people, and i for >> one, very much would like to keep automake, etc. working on window >> based platforms! Lars> I fully support this. There are more platforms with Lars> case-preserving case-insensitive filesystems out there. Yes, I'm sorry, this was really stupid. Had I paid more attention, I would have not written this. Sorry.
Re: plea for case-insensitive, but case preserving (was: Re: 45-fix-yacc2-test.patch)
edward writes: > hi akim, > > this is a plea to avoid using constructs like the "cp Makefile.am > Makefile.AM" below. this kills us cygwin/windows people, and i for one, very > much would like to keep automake, etc. working on window based platforms! I fully support this. There are more platforms with case-preserving case-insensitive filesystems out there.
Re: pr19.test
Right, I reset my automake dir to CVS no patches,
test log in testSubDir (running on cygwin 1.1.8-2)===
$ perl -pi.old -e 's/^\t\@/\t/g' Makefile
$ make distdir
chmod -R a+w ./am_lex_bug-0.1.1 >/dev/null 2>&1; rm -rf
./am_lex_bug-0.1.1
mkdir ./am_lex_bug-0.1.1
for file in Makefile.am Makefile.in aclocal.m4 configure configure.in
depcomp foo.c instal
l-sh missing mkinstalldirs foo.l ; do \
d=.; \
if test -d $d/$file; then \
cp -pR $d/$file ./am_lex_bug-0.1.1 \
|| exit 1; \
else \
test -f ./am_lex_bug-0.1.1/$file \
|| cp -p $d/$file ./am_lex_bug-0.1.1/$file \
|| exit 1; \
fi; \
done
find ./am_lex_bug-0.1.1 -type d ! -perm -777 -exec chmod a+rwx {} \; -o
\
! -type d ! -perm -444 -links 1 -exec chmod a+r {} \; -o \
! -type d ! -perm -400 -exec chmod a+r {} \; -o \
! -type d ! -perm -444 -exec /bin/sh ./install-sh -c -m a+r {} {} \; \
|| chmod -R a+r ./am_lex_bug-0.1.1
=== end test log ===
For the sake of completeness,
== test log in testSubDir (running under ksh on openBSD 2.8)
$ perl -pi -e 's/^\t\@/\t/g' Makefile
$ make distdir
chmod -R a+w ./am_lex_bug-0.1.1 >/dev/null 2>&1; rm -rf
./am_lex_bug-0.1.1
mkdir ./am_lex_bug-0.1.1
for file in Makefile.am Makefile.in aclocal.m4 configure configure.in
depcomp f\
oo.c install-sh missing mkinstalldirs foo.l ; do d=.; if test -d
$d/$file; th\
en cp -pR $d/$file ./am_lex_bug-0.1.1 || exit 1; else test -f
./am_lex_bug-0\
.1.1/$file || cp -p $d/$file ./am_lex_bug-0.1.1/$file || exit 1; fi;
done
find ./am_lex_bug-0.1.1 -type d ! -perm -777 -exec chmod a+rwx {} \; -o
! -type\
d ! -perm -444 -links 1 -exec chmod a+r {} \; -o ! -type d
! -perm -400 -exec\
chmod a+r {} \; -o ! -type d ! -perm -444 -exec /bin/sh
./install-sh -c -m a+r\
{} {} \; || chmod -R a+r ./am_lex_bug-0.1.1
=== end test log
Rob
Bonus content:
make check VERBOSE=e TESTS=pr19.test log ((from cygwin) don't
know if it's useful I haven't included the openBSD one, because we're
AFAIK we're not chasing a platform bug.===
=== Running test ./pr19.test
configure.in:8: warning: AC_PROG_LEX invoked multiple times
checking for a BSD compatible install... /bin/install -c
checking whether build environment is sane... yes
checking for mawk... no
checking for gawk... gawk
checking whether make sets ${MAKE}... yes
checking for gcc... gcc
checking for C compiler default output... a.exe
checking whether the C compiler works... yes
checking whether we are cross compiling... no
checking for executable suffix... .exe
checking for object suffix... o
checking whether we are using the GNU C compiler... yes
checking whether gcc accepts -g... yes
checking how to run the C preprocessor... gcc -E
checking for style of include used by make... #
checking dependency style of gcc... gcc
checking for flex... flex
checking for flex... (cached) flex
checking for yywrap in -lfl... yes
checking lex output file root... lex.yy
checking whether yytext is a pointer... yes
checking for bison... bison -y
configure: creating ./config.status
config.status: creating Makefile
make[3]: Entering directory `/usr/src/automake/tests/testSubDir'
flex foo.l && mv lex.yy.c foo.c
source='foo.c' object='foo.o' libtool=no \
depfile='.deps/foo.Po' tmpdepfile='.deps/foo.TPo' \
depmode=gcc /bin/sh ./depcomp \
gcc -DPACKAGE=\"am_lex_bug\" -DVERSION=\"0.1.1\" -DYYTEXT_POINTER=1 -I.
-I. -g -O2 -c
-o foo.o `test -f foo.c || echo './'`foo.c
gcc -g -O2 -o foo foo.o -lfl
make[3]: Leaving directory `/usr/src/automake/tests/testSubDir'
make[3]: Entering directory `/usr/src/automake/tests/testSubDir'
chmod -R a+w ./am_lex_bug-0.1.1 >/dev/null 2>&1; rm -rf
./am_lex_bug-0.1.1
mkdir ./am_lex_bug-0.1.1
find ./am_lex_bug-0.1.1 -type d ! -perm -777 -exec chmod a+rwx {} \; -o
\
! -type d ! -perm -444 -links 1 -exec chmod a+r {} \; -o \
! -type d ! -perm -400 -exec chmod a+r {} \; -o \
! -type d ! -perm -444 -exec /bin/sh ./install-sh -c -m a+r {} {} \; \
|| chmod -R a+r ./am_lex_bug-0.1.1
/bin/sh /usr/src/automake/tests/testSubDir/missing --run tar chof -
./am_lex_bug-0.1.1 | G
ZIP=--best gzip -c >./am_lex_bug-0.1.1.tar.gz
chmod -R a+w ./am_lex_bug-0.1.1 >/dev/null 2>&1; rm -rf
./am_lex_bug-0.1.1
chmod -R a+w ./am_lex_bug-0.1.1 > /dev/null 2>&1; rm -rf
./am_lex_bug-0.1.1
GZIP=--best gunzip -c ./am_lex_bug-0.1.1.tar.gz | /bin/sh
/usr/src/automake/tests/testSubD
ir/missing --run tar xf -
chmod -R a-w ./am_lex_bug-0.1.1; chmod a+w ./am_lex_bug-0.1.1
mkdir ./am_lex_bug-0.1.1/=build
mkdir ./am_lex_bug-0.1.1/=inst
chmod a-w ./am_lex_bug-0.1.1
dc_install_base=`CDPATH=: && cd ./am_lex_bug-0.1.1/=inst && pwd` \
&& cd ./am_lex_bug-0.1.1/=build \
&& ../configure --srcdir=.. --prefix=$dc_install_base \
&& make \
&& make dvi \
&& make check \
&& make install \
&& make installcheck \
&& make uninstall \
&& test `find $dc_install_base -type f -print | wc -l` -le 1 \
&& make dist \
&& make distclean \
&& rm -f ./am_lex_bug-0.1.1.tar.gz \
&& (test `find . -type f
plea for case-insensitive, but case preserving (was: Re: 45-fix-yacc2-test.patch)
hi akim, this is a plea to avoid using constructs like the "cp Makefile.am Makefile.AM" below. this kills us cygwin/windows people, and i for one, very much would like to keep automake, etc. working on window based platforms! cheers, edward > +cp Makefile.am Makefile.AM - Original Message - From: "Akim Demaille" <[EMAIL PROTECTED]> To: "Automake" <[EMAIL PROTECTED]> Sent: Sunday, March 11, 2001 1:09 PM Subject: 45-fix-yacc2-test.patch > > Index: ChangeLog > from Akim Demaille <[EMAIL PROTECTED]> > * tests/yacc2.test: Don't define several times a variable, as > automake complains. > > Index: tests/yacc2.test > --- tests/yacc2.test Sat, 13 Jan 2001 18:11:09 +0100 akim (am/6_yacc2.test 1.1 775) > +++ tests/yacc2.test Sun, 11 Mar 2001 16:55:21 +0100 akim (am/6_yacc2.test 1.1 775) > @@ -15,11 +15,20 @@ > zardoz_SOURCES = zardoz.y > END > > +# Don't redefine several times the same variable. > +cp Makefile.am Makefile.AM > + > +
"unknown extension in source" perl error (with patch)
hi peeps:
given the following Makefile.am:
noinst_PROGRAMS = foo
foo_SOURCES = foo.i
automake will die with:
+ perl ../../automake --amdir=/usr/beta/src/automake/tests/.. --foreign
-a
Use of uninitialized value in string eq at ../../automake line 5417.
unknown extensions are not tested before used.
ChangeLog:
2001-03-13 Edward M. Lee <[EMAIL PROTECTED]>
* automake.in (handle_single_transform_list): test (and warn the
user) for unknown extensions before doing language features.
Suggested Patch:
--- automake.in.origTue Mar 13 04:49:55 2001
+++ automake.in Tue Mar 13 06:28:11 2001
@@ -1633,6 +1633,11 @@
my $renamed = 0;
my ($linker, $object);
+ if (!defined($extension_map{$extension}))
+ {
+ &am_error ("unknown extension for `$full' in $var");
+ next;
+ }
$extension = &derive_suffix ($extension);
my $lang = $extension_map{$extension};
if ($lang)
__
Do You Yahoo!?
Yahoo! Auctions - Buy the things you want at great prices.
http://auctions.yahoo.com/
Re: pr19.test
- Original Message - From: "Akim Demaille" <[EMAIL PROTECTED]> To: "edward" <[EMAIL PROTECTED]> Cc: <[EMAIL PROTECTED]> Sent: Tuesday, March 13, 2001 5:39 AM Subject: Re: pr19.test > Thanks for the details. As I'm currently working on DIST_COMMON > etc. (I have another bunch of patches waiting at home), I'm not > inclined to fix this now (collisions...). > > But still, I fail to understand the relationship between your > explanation and the failure you observe. In what way is this related > to Cygwin? _That_ is the real question. The scenario you depict > seems to imply that Automake should have pr19 fail everywhere, which > is not the case. So why do _you_ observe a failure? hi Akim, On systems with make programs which delete intermediate targets (including cygwin), I expect pr19 to fail with foo.c not found. I haven't verified this since I don't have access to non-gnu make, but it seems consistent with what others have reported here. On cygwin itself, there is an additional problem due to the "make clean" test failing. foo.exe isn't removed properly. That can be fixed by putting AC_EXEEXT into configure.in cheers, edward ps. looking again at the Makefile.am of pr19.test, foo is listed as noinst. i suppose lang_*_rewrite should at least check for noinst/check targets before pushing into dist common, similar to what happens in am_install_var. that's a separate issue though =) __ Do You Yahoo!? Yahoo! Auctions - Buy the things you want at great prices. http://auctions.yahoo.com/
Re: pr19.test
| From | http://sources.redhat.com/ml/automake/2001-03/msg00183.html | | 2) lines 4946-4947 of automake.in version 1.977 | | # FIXME: nodist. | &push_dist_common ($pfx . $base . '.' . $ext); | | commenting out the last line removes foo.c (which is a temporary file) | from the make distdir target, which fails because according to automake, | foo.c is supposed to be redistributed. however, (gnu) make removes temporary | files like foo.c (generated from foo.l). in my opinion, this is correct and | should not be redistributed. the Makefile.am writer can always explicitly | include foo.c if they really want to. from the comment in the automake.in | file, it seems the automake peeps are aware of this. there are similar | comments in automake.in with other generated files as well. | | Here is the context: | | foo.o: foo.l | | make has enough information to convert foo.l -> foo.c (via lex/flex) | then compile foo.c to foo.o. However, since foo.c is an intermediate file | it is deleted. | | automake adds foo.c to the list of dist files though, via push_dist_common. | from the comments in the automake source, it appears this is already known, | but i don't know if anyone plans on doing anything about it. by the way, | same | issue for other auto-generated files. just search automake.in for "FIXME: | nodist" Thanks for the details. As I'm currently working on DIST_COMMON etc. (I have another bunch of patches waiting at home), I'm not inclined to fix this now (collisions...). But still, I fail to understand the relationship between your explanation and the failure you observe. In what way is this related to Cygwin? _That_ is the real question. The scenario you depict seems to imply that Automake should have pr19 fail everywhere, which is not the case. So why do _you_ observe a failure?
Re: pr19.test
From http://sources.redhat.com/ml/automake/2001-03/msg00183.html 2) lines 4946-4947 of automake.in version 1.977 # FIXME: nodist. &push_dist_common ($pfx . $base . '.' . $ext); commenting out the last line removes foo.c (which is a temporary file) from the make distdir target, which fails because according to automake, foo.c is supposed to be redistributed. however, (gnu) make removes temporary files like foo.c (generated from foo.l). in my opinion, this is correct and should not be redistributed. the Makefile.am writer can always explicitly include foo.c if they really want to. from the comment in the automake.in file, it seems the automake peeps are aware of this. there are similar comments in automake.in with other generated files as well. Here is the context: foo.o: foo.l make has enough information to convert foo.l -> foo.c (via lex/flex) then compile foo.c to foo.o. However, since foo.c is an intermediate file it is deleted. automake adds foo.c to the list of dist files though, via push_dist_common. from the comments in the automake source, it appears this is already known, but i don't know if anyone plans on doing anything about it. by the way, same issue for other auto-generated files. just search automake.in for "FIXME: nodist" cheers
Re: pr19.test
> make[2]: Leaving directory `/usr/src/automake/tests/testSubDir'
> make[2]: Entering directory `/usr/src/automake/tests/testSubDir'
> chmod -R a+w ./am_lex_bug-0.1.1 >/dev/null 2>&1; rm -rf
> ./am_lex_bug-0.1.1
> mkdir ./am_lex_bug-0.1.1
> cp: ./foo.c: No such file or directory
I have this.
make[3]: Leaving directory `/mnt/nostromo/home/lrde/prof/akim/src/am/tests/testSubDir'
make[3]: Entering directory `/mnt/nostromo/home/lrde/prof/akim/src/am/tests/testSubDir'
chmod -R a+w ./am_lex_bug-0.1.1 >/dev/null 2>&1; rm -rf ./am_lex_bug-0.1.1
mkdir ./am_lex_bug-0.1.1
find ./am_lex_bug-0.1.1 -type d ! -perm -777 -exec chmod a+rwx {} \; -o \
! -type d ! -perm -444 -links 1 -exec chmod a+r {} \; -o \
! -type d ! -perm -400 -exec chmod a+r {} \; -o \
! -type d ! -perm -444 -exec /bin/sh ./install-sh -c -m a+r {} {} \; \
|| chmod -R a+r ./am_lex_bug-0.1.1
Could you try to find the guilty lines in tests/testSubDir/Makefile?
I'd suggest that you
make check VERBOSE=e TESTS=pr19.test
perl -pi 's/^\t\@/\t/g' testSubDir/Makefile
cd testSubDir
make distdir
And send the logs.
