ate build-to-host.m4 had been checked into the repository, but this
is indeed not the case.
--
Russ Allbery (ea...@eyrie.org) <https://www.eyrie.org/~eagle/>
s available.
It's not a panacea (there are no panaceas), but it's less aggressive and
disruptive than some other ideas that have been proposed, and I think it's
mostly best practice already.
--
Russ Allbery (ea...@eyrie.org) <https://www.eyrie.org/~eagle/>
ks, since there isn't
the same mangification of scale as compromising the distro packages.)
> Well, it'd at least make things more difficult for the attacker, even if
> it wouldn't stop them completely.
This is the whole field of security. Nothing stops attackers completely;
more difficult is
't answer the
question of why that's in the test suite log, though.
--
Russ Allbery (ea...@eyrie.org) <https://www.eyrie.org/~eagle/>
generated code.
Does that get at the problem that you're having?
--
Russ Allbery (ea...@eyrie.org) <https://www.eyrie.org/~eagle/>
out the interaction
between user variables, ‘AM_’ shadow variables, and per-target
variables.
and
https://www.gnu.org/software/automake/manual/automake.html#Flag-Variables-Ordering
--
Russ Allbery (ea...@eyrie.org) <https://www.eyrie.org/~eagle/>
irly sure that, back when I was actively maintaining
and installing software on Solaris, I never bothered with separate
directories for the lettered subsections, and it never caused problems.
--
Russ Allbery (ea...@eyrie.org) <https://www.eyrie.org/~eagle/>
to cause frustration for someone who
has downloaded the package and wants to use it, not develop on it.
--
Russ Allbery (ea...@eyrie.org) <http://www.eyrie.org/~eagle/>
to the native build
system for that language.
--
Russ Allbery (ea...@eyrie.org) <http://www.eyrie.org/~eagle/>
pport nor ignore those
sequences, I'd very much like to know so that I can update my
compatibility chart.
--
Russ Allbery (ea...@eyrie.org) <http://www.eyrie.org/~eagle/>
Those address the much more common case of redirecting test
output to a file, where the ANSI codes cause a lot of problems.)
--
Russ Allbery (ea...@eyrie.org) <http://www.eyrie.org/~eagle/>
ust had Autoconf create the
directory by adding:
AC_CONFIG_COMMANDS([server], [test -d server || mkdir server])
(in my case, the name of the directory containing the scripts was server).
It's kind of a hack, but it works.
--
Russ Allbery (ea...@eyrie.org) <http://www.eyrie.org/~eagle/>
more work over time, since the build system changes a lot.
http://www.eyrie.org/~eagle/journal/2013-01/025.html suggested not to
use ExtUtils::MakeMaker at all.
Yeah, I still need to find time to do that writeup I promised I'd do. :)
--
Russ Allbery (ea...@eyrie.org) http
: @GSSAPI_LDFLAGS@ @GSSAPI_LIBS@
--
Russ Allbery (ea...@eyrie.org) http://www.eyrie.org/~eagle/
Arthur Schwarz aschwarz1...@att.net writes:
automake (GNU automake) 1.14.1
Chapter 15.4.1, page 112 Test::Harness::TAP link is no longer active. I
found http://testanything.org/; which may be a suitable replacement.
TAP::Parser::Grammar is the direct replacement, I think.
--
Russ Allbery
:: man
although I'm not positive that's portable to every make.
--
Russ Allbery (ea...@eyrie.org) http://www.eyrie.org/~eagle/
it also has to integrate with
the native build systems of Perl, Python, PHP, and Ruby.
--
Russ Allbery (ea...@eyrie.org) http://www.eyrie.org/~eagle/
set -e; if [ -f ruby/Makefile ] ; then\
cd ruby $(MAKE) distclean ; \
fi
--
Russ Allbery (ea...@eyrie.org) http://www.eyrie.org/~eagle/
that
would have been in the convenience library being built multiple times with
meaningless compiler flag variations. That can reintroduce quite a lot of
build slowness, as well as duplication of warning messages, etc.
--
Russ Allbery (ea...@eyrie.org) http://www.eyrie.org/~eagle/
packages now use it extensively in their
configure probes. I'm not sure what the uptake outside of the Linux world
looks like. But systemd itself is only available on Linux, so you're
pretty much guaranteed that any system that has systemd also has
pkg-config or can easily install it.
--
Russ
in daemon(7).
--
Russ Allbery (ea...@eyrie.org) http://www.eyrie.org/~eagle/
.
--
Russ Allbery (ea...@eyrie.org) http://www.eyrie.org/~eagle/
information, it's usually best
to just rebuild all the *.java files that make up a JAR if any of them
have changed, to ensure that no errors have been introduced and no
internal APIs have changed.
--
Russ Allbery (r...@stanford.edu) http://www.eyrie.org/~eagle/
information, it's usually best
to just rebuild all the *.java files that make up a JAR if any of them
have changed, to ensure that no errors have been introduced and no
internal APIs have changed.
--
Russ Allbery (r...@stanford.edu) http://www.eyrie.org/~eagle/
anything for
which automatic deps are particularly meaningful or important. Usually
normal make dependencies are sufficient and unexciting.
I can't comment on cross-compilation having never done it myself.
--
Russ Allbery (r...@stanford.edu) http://www.eyrie.org/~eagle/
maintainer and may have more context.
--
Russ Allbery (r...@stanford.edu) http://www.eyrie.org/~eagle/
of supporting tools (dh and dh-autoreconf on
the Debian side, for example). There's an advantage to sticking with a
suboptimal process if it's the process that everyone else is using. It
all depends on how significant the gains are.
--
Russ Allbery (r...@stanford.edu) http://www.eyrie.org
-bit multiarch system library path.
--
Russ Allbery (r...@stanford.edu) http://www.eyrie.org/~eagle/
way of expressing things and just adds complexity to no real purpose.
--
Russ Allbery (r...@stanford.edu) http://www.eyrie.org/~eagle/
the stable branch and
version them (either minor or micro) according to the nature of the
changes that have accumulated.
--
Russ Allbery (r...@stanford.edu) http://www.eyrie.org/~eagle/
Ineiev ine...@gmail.com writes:
On 02/08/2013 08:30 PM, Russ Allbery wrote:
Another place where the default behavior frequently breaks is if one is
applying a patch to both the generated file and the source file,
usually because one explicitly *doesn't* want to re-run Automake (often
because
, that also works. It just seems kind of silly to have to deceive
make rather than just removing the make rules that one doesn't want.
--
Russ Allbery (r...@stanford.edu) http://www.eyrie.org/~eagle/
before the source file, make then tries to re-run
Automake and everything explodes.
--
Russ Allbery (r...@stanford.edu) http://www.eyrie.org/~eagle/
fun with it, and I think that's important!
To be honest, the main reason why I've not already started playing with it
is that it's not packaged for Debian, which is enough of a hurdle that I
haven't found the time to fiddle with it.
--
Russ Allbery (r...@stanford.edu) http
to want it to continue to be easy to run tests serially on an ongoing,
permanent basis, and may even want to consider some way to mark specific
tests as non-parallelizable (so that as many tests as possible can be run
in parallel anyway).
--
Russ Allbery (r...@stanford.edu) http
be disabled by default as well, in case people are
reading the bit about casual user and going well, my package doesn't
have any casual users, so I don't care.)
--
Russ Allbery (r...@stanford.edu) http://www.eyrie.org/~eagle/
this. However, I agree that when you have
dozens of them, there may be something amiss.
--
Russ Allbery (r...@stanford.edu) http://www.eyrie.org/~eagle/
for working compilers and
compiler properties would cause some buggy vendor compilers to core dump.
--
Russ Allbery (r...@stanford.edu) http://www.eyrie.org/~eagle/
). The
easiest way to work around this is probably to change the Automake helper
functions that sit between the code and the Perl open command and have
them switch to calling open with two arguments if the file name is -.
--
Russ Allbery (r...@stanford.edu) http://www.eyrie.org/~eagle/
, and then tests/runtests
-o test will work as expected.
--
Russ Allbery (r...@stanford.edu) http://www.eyrie.org/~eagle/
or some other scripting language
rather than C, while the core test code has to be written in C because of
what it's testing.)
--
Russ Allbery (r...@stanford.edu) http://www.eyrie.org/~eagle/
it and it sounds like the
Automake support for it will be there.
--
Russ Allbery (r...@stanford.edu) http://www.eyrie.org/~eagle/
could chmod +x the configure script before running it, but
I've never had to do that.
--
Russ Allbery (r...@stanford.edu) http://www.eyrie.org/~eagle/
Bob Friesenhahn bfrie...@simple.dallas.tx.us writes:
On Wed, 21 Mar 2012, Russ Allbery wrote:
Miles Bader mi...@gnu.org writes:
[Relying on source-code execute bits always being correctly maintained
is one of those things that ... well... doesn't really feel very robust.
I dunno, maybe it's
to just:
bin_PROGRAMS = myprog
and having the script just be named that and already be executable, which
is what I do with any exectuable shell scripts.
--
Russ Allbery (r...@stanford.edu) http://www.eyrie.org/~eagle/
.
--
Russ Allbery (r...@stanford.edu) http://www.eyrie.org/~eagle/
Automake normally
arranged to run that at the end of make install. Is that not happening
for some reason?
libtool --mode=finish will run ldconfig. It's much of the point of having
a separate finish mode.
--
Russ Allbery (r...@stanford.edu) http://www.eyrie.org/~eagle/
complicated and other parts of the compilation
process dominate the time. C++ is significantly different than C in how
much complexity it puts into its headers.
--
Russ Allbery (r...@stanford.edu) http://www.eyrie.org/~eagle/
complicated and other parts of the compilation
process dominate the time. C++ is significantly different than C in how
much complexity it puts into its headers.
--
Russ Allbery (r...@stanford.edu) http://www.eyrie.org/~eagle/
' then the output format can be certain.
The problem, of course, being that pax probably isn't available, which
makes it hard for a default choice.
--
Russ Allbery (r...@stanford.edu) http://www.eyrie.org/~eagle/
, but isn't part of a typical install.
--
Russ Allbery (r...@stanford.edu) http://www.eyrie.org/~eagle/
; \
FILES=`find . -type f`; \
for file in $$FILES; do \
echo Processing $$file; \
todos -p $$file; \
mv $$file $$file.txt; \
done)
--
Russ Allbery (r...@stanford.edu) http://www.eyrie.org/~eagle/
list for the given target.
In other words, if this is used to build libfoo.a, then add:
EXTRA_libfoo_a_SOURCES = foo.c
to your Makefile.am.
--
Russ Allbery (r...@stanford.edu) http://www.eyrie.org/~eagle/
to be completed within the given time frame.
You may be interested in:
http://www.eyrie.org/~eagle/software/c-tap-harness/
--
Russ Allbery (r...@stanford.edu) http://www.eyrie.org/~eagle/
and then set LUA_DIR as an AC_SUBST variable in your Autoconf
configure.ac, which can then apply platform-specific logic, use a --with
flag to specify the path, etc.
--
Russ Allbery (r...@stanford.edu) http://www.eyrie.org/~eagle/
is set, then the
object file will be named, for instance, `maude-sample.o'. (See
also *note Renamed Objects::.)
--
Russ Allbery (r...@stanford.edu) http://www.eyrie.org/~eagle/
Bob Friesenhahn bfrie...@simple.dallas.tx.us writes:
On Thu, 29 Apr 2010, Russ Allbery wrote:
I think this just varies based on what your developers are like and how
closed your project is, basically. People often say that they find it
fairly easy to make all developers on a project use
,
I'm finding myself more and more just using git clean -x -f -d rather than
worrying about the maintainer-clean target.
--
Russ Allbery (r...@stanford.edu) http://www.eyrie.org/~eagle/
Trevor Harmon tre...@vocaro.com writes:
On Apr 29, 2010, at 12:25 AM, Russ Allbery wrote:
Although I must admit that now that I've switched to Git for
everything, I'm finding myself more and more just using git clean -x -f
-d
What does that do?
Deletes every file and directory that isn't
merge conflicts from auto-generated files probably aren't making your life
much worse the way that they would in Git.
--
Russ Allbery (r...@stanford.edu) http://www.eyrie.org/~eagle/
NightStrike nightstr...@gmail.com writes:
On Thu, Apr 29, 2010 at 7:53 PM, Russ Allbery r...@stanford.edu wrote:
I suspect it depends on what sort of activities you expect people using
a VCS checkout directly to be doing, and also how sophisticated of a
VCS you're using. If you're using CVS
with either TAB or
DOTWHITESPACE? Or would that break too much backwards
compatability?
It breaks the basic assumption that Makefile.am is basically a makefile.
I suppose that Automake could try to transform the whitespace as part of
its processing, but I'm not sure that's a good idea.
--
Russ
Ralf Wildenhues ralf.wildenh...@gmx.de writes:
* Ralf Wildenhues wrote on Mon, Apr 19, 2010 at 06:55:30PM CEST:
* Russ Allbery wrote on Fri, Apr 16, 2010 at 10:27:11PM CEST:
parallel.1: parallel
pod2man --release='$(PACKAGE_VERSION)' --center='$(PACKAGE_NAME)' \
--section=1
Ralf Wildenhues ralf.wildenh...@gmx.de writes:
* Russ Allbery wrote on Mon, Apr 19, 2010 at 08:39:48PM CEST:
Ralf Wildenhues ralf.wildenh...@gmx.de writes:
* Ralf Wildenhues wrote on Mon, Apr 19, 2010 at 06:55:30PM CEST:
* Russ Allbery wrote on Fri, Apr 16, 2010 at 10:27:11PM CEST:
parallel
program.1: program
pod2man --release='$(PACKAGE_VERSION)' --center='$(PACKAGE_NAME)' \
--section=1 program program.1
CLEANFILES = program.1
--
Russ Allbery (r...@stanford.edu) http://www.eyrie.org/~eagle/
the output of the latter command?
Also, if make install knows to install the man page, make should
automatically attempt to build it as well, so I'm confused as to why that
wouldn't happen.
--
Russ Allbery (r...@stanford.edu) http://www.eyrie.org/~eagle/
it into standards are
widespread existing practice before then, and it's too much work to change
them.
I suspect this is part of why, as you point out, the standard also
introduces intsize_t at the same time, but long long is more widely
supported, probably because it's older than the standard.
--
Russ
they look like individual source files. I suspect that
somewhere you're setting the wrong Automake variable to a list of header
files, although I don't see that in the Makefile.am excerpt that you
included.
--
Russ Allbery (r...@stanford.edu) http://www.eyrie.org/~eagle/
them off and somewhat lesser but probably still reasonable consensus that
turning them on as an Automake-wide default is a bad idea, but I don't
think any of the maintainers are going to object to making it silent when
that's explicitly requested.
--
Russ Allbery (r...@stanford.edu
. I just
write explicit rules.
--
Russ Allbery (r...@stanford.edu) http://www.eyrie.org/~eagle/
#! /bin/sh
## $Id: fixconfig.in 8471 2009-05-17 08:25:35Z iulius $
##
## Make variable substitutions in configuration files.
##
## This script does something very similar to what
is cached in a config.h file.
Because of the requirements for expansion of shell variables, usually it's
best to not put paths into config.h. Instead, passing them on the
compiler command line is usually easier.
See info autoconf FAQ 'Defining Directories'
--
Russ Allbery (r...@stanford.edu
between
source files and output *roff files, at which point (if they're named
accordingly) the pattern rule would work.
--
Russ Allbery (r...@stanford.edu) http://www.eyrie.org/~eagle/
Ralf Wildenhues ralf.wildenh...@gmx.de writes:
* Russ Allbery wrote on Wed, Dec 16, 2009 at 05:08:14AM CET:
Do you really want to generate separate copies of the man page for each
function documented by the same source POD file? It's more normal to
generate one copy of the man page
-local: $(BINDINGS)
install-data-local: $(BINDINGS_INSTALL)
--
Russ Allbery (r...@stanford.edu) http://www.eyrie.org/~eagle/
checks
normally go against the philosophy of the Autotools, but in this case
they're files shipped by Automake itself, and it might be the easiest
path.
--
Russ Allbery (r...@stanford.edu) http://www.eyrie.org/~eagle/
.
--
Russ Allbery (r...@stanford.edu) http://www.eyrie.org/~eagle/
...
subdir/my-dirstamp:
@$(MKDIR_P) subdir
@: $@
DISTCLEANFILES += subdir/my-dirstamp
A possibly simpler way, depending on what you're trying to accomplish, is
to do this at configure time:
AC_CONFIG_COMMANDS([tests/data/.placeholder], [touch tests/data/.placeholder])
--
Russ
])
and then use code like:
RRA_LIB_KRB5_SWITCH
AC_CHECK_FUNCS([krb5_cc_copy_creds \
krb5_free_keytab_entry_contents \
krb5_get_init_creds_opt_set_default_flags \
krb5_get_renewed_creds])
AC_CHECK_TYPES([krb5_realm], , , [#include krb5.h])
RRA_LIB_KRB5_RESTORE
for cases like that.
--
Russ
anything
in the FHS, but it's possible that LSB has additional file layout
standards outside the FHS.
--
Russ Allbery (r...@stanford.edu) http://www.eyrie.org/~eagle/
([foo], [1.0], [...@stanford.edu])
Seems to work for me, although I've not used it for a real package. You
then just automatically generate version.m4 as part of your bootstrap
process before running autoconf.
--
Russ Allbery (r...@stanford.edu) http://www.eyrie.org/~eagle/
with:
./configure --prefix=/usr/local
make
make install prefix=/usr/local/stow/package-version
DESTDIR cannot easily be used here because the stow layout should have
simple bin, etc, lib, etc. directories directly under that directory, not
with an extra /usr/local in the way.
--
Russ Allbery
Russ Allbery r...@stanford.edu writes:
Ralf Wildenhues ralf.wildenh...@gmx.de writes:
[1] I'm asking because Automake 1.11 will reliably not install files if
their respective installation directory is empty. This is not yet
functional in Automake 1.10.2. The test for emptiness in 1.11
important.
--
Russ Allbery (r...@stanford.edu) http://www.eyrie.org/~eagle/
autoscan will
suggest using them if you ever use either of those functions, but you may
or may not need the portability there.
--
Russ Allbery (r...@stanford.edu) http://www.eyrie.org/~eagle/
manual has the documentation for both of these.
--
Russ Allbery (r...@stanford.edu) http://www.eyrie.org/~eagle/
that aren't worth bothering checking for unless you're porting
to Windows (which particularly for shared libraries involves quite a bit
of work that Automake can't entirely help with).
--
Russ Allbery (r...@stanford.edu) http://www.eyrie.org/~eagle/
Gerald I. Evenden geraldi.even...@gmail.com writes:
On Thursday 26 March 2009 10:26:13 pm Russ Allbery wrote:
However, more fundamentally, you're reinventing AC_REPLACE_FUNC, which
you probably don't want to do. Check the Autoconf manual for
AC_REPLACE_FUNC, which does exactly what you're
its output files without having make
installed, I believe. The resulting generated files are not useful
without make, but they may not be intended for use on that same system.
--
Russ Allbery (r...@stanford.edu) http://www.eyrie.org/~eagle/
. It doesn't
have a mechanism for eliminating entire lines of output, but adding or not
adding comment characters is fairly straightforward.
--
Russ Allbery ([EMAIL PROTECTED]) http://www.eyrie.org/~eagle/
above.
--
Russ Allbery ([EMAIL PROTECTED]) http://www.eyrie.org/~eagle/
-around.
Thanks for the queued TODO item!
--
Russ Allbery ([EMAIL PROTECTED]) http://www.eyrie.org/~eagle/
) $(KAFS_LDFLAGS)
it has no effect, which after thinking about it for a moment makes some
sense. However, setting LIBOBJS_CPPFLAGS doesn't do anything either.
Could some facility to handle this be added?
--
Russ Allbery ([EMAIL PROTECTED]) http://www.eyrie.org/~eagle/
Russ Allbery [EMAIL PROTECTED] writes:
If I add:
portable_libportable_a_CPPFLAGS = $(KRB5_LDFLAGS) $(KAFS_LDFLAGS)
Er, obviously that should be KRB5_CPPFLAGS and KAFS_CPPFLAGS. Sorry about
the confusion.
it has no effect, which after thinking about it for a moment makes some
sense
some workaround code that
I had doesn't appear to be necesary any more).
--
Russ Allbery ([EMAIL PROTECTED]) http://www.eyrie.org/~eagle/
to make everything a GPL-covered GNU project but who are quite
willing to share code, bug reports, and general software infrastructure.
--
Russ Allbery ([EMAIL PROTECTED]) http://www.eyrie.org/~eagle/
confusion.
People would do things like have two build trees, install from the first
tree, install the second tree while testing something, and then go back
and try to reinstall from the first tree and be very confused when it
didn't do anything.
--
Russ Allbery ([EMAIL PROTECTED]) http
, very different beast that doesn't work
the same way. Debian and the Texinfo developers are currently actively
working on merging the programs so that we can get rid of this workaround
eventually.
--
Russ Allbery ([EMAIL PROTECTED]) http://www.eyrie.org/~eagle/
behind Automake's back and breaking assumptions.
--
Russ Allbery ([EMAIL PROTECTED]) http://www.eyrie.org/~eagle/
module whenever the kernel API changes,
which is about twice a month).
This area is something of a mess and there really isn't much
cross-distribution agreement on how to do it.
--
Russ Allbery ([EMAIL PROTECTED]) http://www.eyrie.org/~eagle/
an application to install a new kernel module. But I guess that
would work around some of the integration issues.
Note that you'll still need to build a new module every time the kernel
API changes (which changes more frequently than the version).
--
Russ Allbery ([EMAIL PROTECTED]) http
1 - 100 of 144 matches
Mail list logo
