n added to the packages-architecture-specific list
for the autobuilders? If not, it's my understanding that it should be;
and then the package that's already been built on s390 should be
removed.
The other architectures in the xfree86 control file can be ignored --
they don't have autobuilders that are uploading to the archive.
--
Steve Langasek
postmodern programmer
signature.asc
Description: Digital signature
from seeing the
> "Architecture: all" Cernlib binary packages?
You can't, though for better or for worse, arch: all packages are only
evaluated on i386 where propagation into testing is concerned.
--
Steve Langasek
postmodern programmer
signature.asc
Description: Digital signature
On Sun, Jul 11, 2004 at 05:56:05AM -0700, William Ballard wrote:
> Where does this script belong? /usr/bin, /var/lib/,
> /usr/share/ ?
> This is probably in the policy but I missed it.
See the FHS, which is included in policy by reference.
--
Steve Langasek
postmodern programmer
me B is upgraded.
By coordinating with the maintainer of that other package, rather than
trying to undermine his packaging decisions.
--
Steve Langasek
postmodern programmer
The phpapi-* virtual packages are not intended for use by
applications written in PHP, they are there for packages which provide
PHP extensions.
--
Steve Langasek
postmodern programmer
signature.asc
Description: Digital signature
een rejected bu Katie:
> Rejected: Unknown distribution `woody-proposed-updates'.
The two names for this suite are "proposed-updates" and "stable".
--
Steve Langasek
postmodern programmer
signature.asc
Description: Digital signature
packages which follow you.
--
Steve Langasek
postmodern programmer
signature.asc
Description: Digital signature
On Sat, Jul 31, 2004 at 05:58:53PM -0700, Steve Langasek wrote:
> On Sat, Jul 31, 2004 at 08:39:52PM -0400, Justin Pryzby wrote:
> > I have a package (x11iraf) which conflicts with gcc-3.3. For the whole
> > story see below..
> > gcc includes /usr/bin/gcc which is prett
r-us state for a single arch which seems to be your case.
> You need to talk to the mips buildd admin to revert that.
But not before reviewing the build logs on buildd.debian.org, to confirm
whether there's a good reason *why* it's been added to P-a-s.
--
Steve Langasek
postm
privileges
> temporarily, dropping the +s bit, or something else.
Do we really want to be adding to the number of svgalib-based programs
in the archive? Surely this isn't the only security problem lurking...
--
Steve Langasek
postmodern programmer
signature.asc
Description: Digital signature
llable -- don't upload uninstallable packages to the archive!
--
Steve Langasek
postmodern programmer
signature.asc
Description: Digital signature
you had in mind, the
description could be changed to not give this impression and frighten
off sponsors.
I don't know if these comments will actually help you find a sponsor,
but I hope they at least give you some insight into why you haven't
found one so far.
Cheers,
--
Steve Langasek
postmodern programmer
signature.asc
Description: Digital signature
for the
> description of the package. Possibly there is a happy medium. I am
> glad you brought this particular detail to my attention though since it
> had never crossed my mind.
Well, as I said, that was my reaction to the phrase. I don't know if
others would have the same reaction
be the fastest and most efficient way
to get the package requeued on all architectures. Well, you could set a
build-conflicts: with the buggy version of package A, I suppose, so the
upload has *some* changes in it... ;)
--
Steve Langasek
postmodern programmer
signature.asc
Description: Digital signature
the
right solution; if the package has not yet been built because it
couldn't be, a sourceful upload is spurious but also happens to be the
most expedient approach.
--
Steve Langasek
postmodern programmer
signature.asc
Description: Digital signature
the "a", it adds no information to the description.
It makes it a grammatically correct noun phrase in English.
But at least drop the capitalization...
--
Steve Langasek
postmodern programmer
signature.asc
Description: Digital signature
latively uncommon for headers from one library
to require headers from other dependency libraries, which is why such -dev
interdependencies are not as common as they might be.
--
Steve Langasek
postmodern programmer
signature.asc
Description: Digital signature
; expects binary packages not to mix upstream versions.
So what package is this, and why are those functions in the shared library
at all?
--
Steve Langasek
postmodern programmer
signature.asc
Description: Digital signature
; How should I do so? I think that I should just upload a new version
> of freefem3d without any other changes. Is that right?
Assuming the build-dependency is libopts-dev or some such, then no change is
needed besides a changelog entry.
Cheers,
--
Steve Langasek
postmodern programmer
si
t on the blowhard rhetoric, and more effort
reading to the bottom of the Policy section in question, mmmkay?
--
Steve Langasek
postmodern programmer
signature.asc
Description: Digital signature
On Wed, Jul 06, 2005 at 09:39:06AM -0700, John Hendrickson wrote:
> Why in the world did DMs compile KDE to say: You must use LessTif and
> uninstall Motif???
Why you so blur one?
--
Steve Langasek
postmodern programmer
signature.asc
Description: Digital signature
you need to change
> the soname for the g++ transition. The current packages must not be
> uploaded anymore.
I assume you mean package name rather than soname.
In any case, packages that are not yet in the Debian archive are not bound
by the C++ ABI transition.
--
Steve Langasek
postmodern pr
cc really use
different default values on arm and PPC than on other architectures? If so,
why?
Steve Langasek
postmodern programmer
it does not do this.
A workaround (to suppress the warning message) is to do something like this
in your makefile:
LD_LIBRARY_PATH=./debian/tmp/usr/lib/ dh_shlibdeps -a
Not very pretty, of course. :)
Steve Langasek
postmodern programmer
the most recent) can use a
'-l ' option to specify additional shared library directories.
You may want to Build-Depends: on a specific version of dpkg-dev if you use
this.
HTH,
Steve Langasek
postmodern programmer
in the DEBIAN/libcurl.shlibs file; but if you
want the build to be clean, this is how to correct the warning.
Steve Langasek
postmodern programmer
ed to allow rebuilding with
different compilers, although having Qt Build-Depend on a non-free package is
not a very good solution. :/
Is the g++ bug being looked into by the upstream?
Steve Langasek
postmodern programmer
libbar-dev. That way if
tomorrow the maintainer of libbar decides to patch her package so that it
builds "foo-free libbar", the autobuilders will still know what to do with
/my/ package.
If, OTOH, you only need B because A is linked to it, then only A belongs in
your Build-Depends li
es which are not suitable for upload into Debian that we want to
distribute from our own sites?
Steve Langasek
postmodern programmer
t is used for cross-compiling, e.g. from i386 to
powerpc. Unless you have an x86 potato environment set up for cross-building
(complete with a second copy of gcc and binutils :), this option won't be of
much use.
Probably not too important in the present context, but I'm compelled to share
tr
place of braces. I believe
${MPI_HOME} references a non-existent /environment/ variable. (Elsewhere the
braces may indeed be correct, since you mention the use of multiple makefiles,
but if you need to reference a Makefile variable set in the same file, you
should normally use parentheses.)
HTH,
Steve Langasek
postmodern programmer
the correct value of MPI_LIB does appear towards the end of the
library string, well past the point where -L{MPI_HOME} shows up.
So I'm just on crack... pay no attention to anything I say. :)
Steve Langasek
postmodern programmer
xt release is
libfoo1.so.1? The least-bad option that seems to hold forth is to name the
package 'libfoo10' (perhaps with a delimiter between the 1 and the 0?)... and
then petition the upstream to un-break their library names...
HTH,
Steve Langasek
postmodern programmer
ry /version/. Which
makes one much less cumbersome than the other.
Steve Langasek
postmodern programmer
endent files
(e.g., PHP class files). You might want to look at the directory structure
for some of the other PHP modules (e.g., php4-mysql, php4-pear) to see what
directories they use.
Regards,
Steve Langasek
postmodern programmer
ary'. And I'm sure I can find plenty
of examples of libraries that underwent complete rewrites and substantial API
changes but didn't find it necessary to add a version to the library name.
Steve Langasek
postmodern programmer
mer, concerns with versioned directories are
already taken care of for you; if the latter, and if the class files aren't
explicitly bound to php4 and up (like PEAR is), you may be in uncharted
waters.
Steve Langasek
postmodern programmer
mba source that's already available in the Debian archives, or
including the Samba tree in my package (as part of the diff !?)
Any advice, pointers, etc. are welcome.
Regards,
Steve Langasek
postmodern programmer
nd Makefile.in/configure.in DO change frequently.
> IMHO one should also always allow for a local copy of the 'donwloadable'
> data you would want to fetch from the net. It's probably best to store it in
> $TMPDIR
Hmm, I think this should be doable, yes; I do something similar in the
upstream. Thanks for the hint!
Steve Langasek
postmodern programmer
Matt,
> On Tue, Mar 27, 2001 at 09:46:30AM -0600, Steve Langasek wrote:
> Which parts of the Samba source does it use? Does it just include header
> files, or does it compile and link with code there?
> There is a module, libpam-smb, which sounds similar to your program, which
e /package/ would change at the author's discretion. Still, even though
it doesn't need to match the release version 2.6.4, 0.0.0 does seem a little
low. If you search through the makefiles for 'soname', do you find anywhere
that this argument is being passed to libtool (or supposed to be)?
Steve Langasek
postmodern programmer
author releases version 0.5 of the package, he can call he library
libgpgme.so.0.5.0 if he wants to -- IF the library interface is the same as
previous versions. If a program compiled against version 0.4 doesn't work
with version 0.5, he must call the library libgpgme.so.1.x.x ... usually
libgpgme.so.1.0.0.
Steve Langasek
postmodern programmer
c, and a cronjob to take
care of the anonymous rsync on the user's behalf? This way, it's a little
more useful to the user than a documentation package would be, and it helps
the keyring server by lowering the bandwidth requirements. (Users don't have
to go to the central keyserve
tions are not part of the actual license should be enough to
get it into main.
If the regulation being referenced actually brings the rights of US Gov't.
users into line with the rights of everyone else, I think it's DFSG-compliant,
but IANAL and that question should probably be asked on deb
gt; country affecting a global distribution.
>
Efforts are being made to resolve this to everyone's satisfaction. It's not
(yet) out of the question that woody will include crypto in main by the time
of release.
Steve Langasek
postmodern programmer
ettext, but no-one has bother translating the program into any
> other languages yet).
We have plenty of software in Debian which is English-only, with no support
for gettext at all. I see no reason why English should be given preferential
treatment.
Steve Langasek
postmodern programmer
on
for wanting licenses in English is so that we can protect ourselves from legal
troubles.
Steve Langasek
postmodern programmer
ne in
debian/control? Or do I need to list out all the architectures I /do/ want to
build it on?
Thanks,
Steve Langasek
postmodern programmer
On Mon, 21 May 2001, Gordon Sadler wrote:
> On Mon, May 21, 2001 at 04:57:56PM -0500, Steve Langasek wrote:
> > Hi all,
> > I may just be looking in the wrong places in the developers' documentation,
> > in
> > which case I apologize in advance. I'm adopti
seeing ./configure, can
reasonably expect that if they run './configure --sysconfdir=/etc/foo/bar'
this setting will be honored. If the program's build scripts are /not/
honoring such settings, I would suggest that this is a genuine bug that ought
to be fixed. This doesn't require any Debian-specific changes to the code.
HTH,
Steve Langasek
postmodern programmer
right'
> Packages that are not Debian-native MUST contain a copy of
> `debian/changelog' file from the Debian source tree in
> `/usr/share/doc/' as `changelog.Debian.gz'.
> (Caps mine.)
/usr/share/doc/, not /usr/share/. Two different beasts.
Steve Langasek
postmodern programmer
le,
> so the package names should include also the micro number.
If they want to call the upstream package 'gtk 1.3.x' to indicate where in the
development cycle they are, that's fine; but the library soname should not be
governed by marketing, and upstream /should/ be LARTed for this.
Steve Langasek
postmodern programmer
ts own
major so number. Anything else is a hazard for our users, regardless of what
developers think of the situation.
And RedHat is allowed to do what it wants to with its own packages -- Debian
didn't get its reputation as a high-quality distribution by following RedHat's
lead.
Steve Langasek
postmodern programmer
seem possible, there are other ways that have been used in the past to
handle the identification requirement. In the meantime, sponsored uploads are
a great way to get started with Debian.
Regards,
Steve Langasek
postmodern programmer
> For now I want a sponsor to get garchiver in to Debian.
&g
on't already know a lot about PKI before they become DD's,
never learn more than the mechanics of signing a key -- and that's ok, until
we start encouraging them to go out and sign other people's keys. :)
Steve Langasek
postmodern programmer
KI that uses CAs (such as Thawte) are
not entirely compatible with those of a peer-to-peer system (such as PGP).
Steve Langasek
postmodern programmer
[1] And is Thawte really so impervious to corruption that there's not even a
*remote* possibility of falsification? Remember that they're now owned by
Network Solutions. Anything is possible...
s fall into the same category as clothing; if you're a Debian
developer and it's not for profit, I think it's safe to use it. However, if
you're going to use these cards in professional dealings, I wouldn't use the
logo without getting permission first.
HTH,
Steve Langasek
postmodern programmer
of the uids of
someone I found out had asked me to sign a uid that wasn't theirs. I just
can't understand why we have to worry about anyone misrepresenting a key as
their own when it has their name on it, since that only hurts their digital
identity.
Steve Langasek
postmodern programmer
On Fri, 29 Jun 2001, Manoj Srivastava wrote:
> >>"Steve" == Steve Langasek <[EMAIL PROTECTED]> writes:
> >> Are you implying that ensuring the person whose identity you
> >> verified actually controls the email address and the secret pass
>
ty of changes noted are specific
to the shared debian directory, of which there is precisely one for any set of
binary packages that are built from a single source package.
Steve Langasek
postmodern programmer
g changes to the Debian packaging itself. I suspect you'll find
that other uses of this file would not be well-supported by the makers of the
related Debian packaging tools.
Regards,
Steve Langasek
postmodern programmer
icy wrt my earlier
decision to mark the config file as a conffile?
Thanks,
Steve Langasek
postmodern programmer
s that don't supply one?
I would suggest that it's a maintainer's responsibility, as liaison to the
upstream developers, to petition them to adopt SONAMEs in any shared libraries
they provide that will be linked by external applications.
Steve Langasek
postmodern programmer
On Wed, 25 Jul 2001, Steve M. Robbins wrote:
> On Wed, Jul 25, 2001 at 03:50:47PM -0500, Steve Langasek wrote:
> > On Wed, 25 Jul 2001, Steve M. Robbins wrote:
> > > So I guess I'm still searching for the answer to my original questions:
> > > 1. Does Debian r
much attention as most. In
fact, last I heard, there was no i386 autobuilder for non-US. You'll probably
want to grab the latest sources, build the package yourself, and upload it.
(As the maintainer, you'll want to grab the NMU diffs anyways...)
Cheers,
Steve Langasek
postmodern programmer
On Sun, Apr 04, 2004 at 12:33:35PM -0600, Benjamin Cutler wrote:
> Steve Langasek wrote:
> >IIRC, the --distribution argument only has an effect the first time you
> >run 'pbuilder create'.
> What should I have done instead to ensure the program was being built
y for the packages
> > uploaded to the archive.
> But WHY ???
Because to do otherwise makes diffs of debian/changelog a PITA.
Including a debian/ dir in CVS is reasonable. Including it in an
upstream source release is not.
--
Steve Langasek
postmodern programmer
signature.asc
Description: Digital signature
It would still be a bug to turn on -O3 optimization when you don't know
what you're doing (which is to say, in the vast majority of cases),
whether or not it actually trips a compiler bug in the process.
--
Steve Langasek
postmodern programmer
signature.asc
Description: Digital signature
to libgds.so? What does 'file' say about the format of this file, if
dpkg-shlibdeps doesn't recognize it?
--
Steve Langasek
postmodern programmer
signature.asc
Description: Digital signature
; /usr/lib/libgds.so.0: ELF 32-bit LSB shared object, Intel 80386, version 1 (SYSV),
> stripped
> It is in one of these packages libfirebird-c32, libfirebird-s64,
> libfirebird-s32, libfirebird-c64 which are orphaned. See
> http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=237532
> Is i
ges
that block its removal, as per the update_output.txt report.
--
Steve Langasek
postmodern programmer
signature.asc
Description: Digital signature
ated.
FWIW, I'm not sure this kind of question is appropriate for
debian-mentors; surely this would be better on debian-devel?
--
Steve Langasek
postmodern programmer
signature.asc
Description: Digital signature
n added to the packages-architecture-specific list
for the autobuilders? If not, it's my understanding that it should be;
and then the package that's already been built on s390 should be
removed.
The other architectures in the xfree86 control file can be ignored --
they don't have autobuilders that are uploading to the archive.
--
Steve Langasek
postmodern programmer
signature.asc
Description: Digital signature
from seeing the
> "Architecture: all" Cernlib binary packages?
You can't, though for better or for worse, arch: all packages are only
evaluated on i386 where propagation into testing is concerned.
--
Steve Langasek
postmodern programmer
signature.asc
Description: Digital signature
On Sun, Jul 11, 2004 at 05:56:05AM -0700, William Ballard wrote:
> Where does this script belong? /usr/bin, /var/lib/,
> /usr/share/ ?
> This is probably in the policy but I missed it.
See the FHS, which is included in policy by reference.
--
Steve Langasek
postmodern programmer
me B is upgraded.
By coordinating with the maintainer of that other package, rather than
trying to undermine his packaging decisions.
--
Steve Langasek
postmodern programmer
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
The phpapi-* virtual packages are not intended for use by
applications written in PHP, they are there for packages which provide
PHP extensions.
--
Steve Langasek
postmodern programmer
signature.asc
Description: Digital signature
een rejected bu Katie:
> Rejected: Unknown distribution `woody-proposed-updates'.
The two names for this suite are "proposed-updates" and "stable".
--
Steve Langasek
postmodern programmer
signature.asc
Description: Digital signature
packages which follow you.
--
Steve Langasek
postmodern programmer
signature.asc
Description: Digital signature
On Sat, Jul 31, 2004 at 05:58:53PM -0700, Steve Langasek wrote:
> On Sat, Jul 31, 2004 at 08:39:52PM -0400, Justin Pryzby wrote:
> > I have a package (x11iraf) which conflicts with gcc-3.3. For the whole
> > story see below..
> > gcc includes /usr/bin/gcc which is prett
r-us state for a single arch which seems to be your case.
> You need to talk to the mips buildd admin to revert that.
But not before reviewing the build logs on buildd.debian.org, to confirm
whether there's a good reason *why* it's been added to P-a-s.
--
Steve Langasek
postm
privileges
> temporarily, dropping the +s bit, or something else.
Do we really want to be adding to the number of svgalib-based programs
in the archive? Surely this isn't the only security problem lurking...
--
Steve Langasek
postmodern programmer
signature.asc
Description: Digital signature
llable -- don't upload uninstallable packages to the archive!
--
Steve Langasek
postmodern programmer
signature.asc
Description: Digital signature
you had in mind, the
description could be changed to not give this impression and frighten
off sponsors.
I don't know if these comments will actually help you find a sponsor,
but I hope they at least give you some insight into why you haven't
found one so far.
Cheers,
--
Steve Langasek
postmodern programmer
signature.asc
Description: Digital signature
for the
> description of the package. Possibly there is a happy medium. I am
> glad you brought this particular detail to my attention though since it
> had never crossed my mind.
Well, as I said, that was my reaction to the phrase. I don't know if
others would have the same reaction
be the fastest and most efficient way
to get the package requeued on all architectures. Well, you could set a
build-conflicts: with the buggy version of package A, I suppose, so the
upload has *some* changes in it... ;)
--
Steve Langasek
postmodern programmer
signature.asc
Description: Digital signature
the
right solution; if the package has not yet been built because it
couldn't be, a sourceful upload is spurious but also happens to be the
most expedient approach.
--
Steve Langasek
postmodern programmer
signature.asc
Description: Digital signature
the "a", it adds no information to the description.
It makes it a grammatically correct noun phrase in English.
But at least drop the capitalization...
--
Steve Langasek
postmodern programmer
signature.asc
Description: Digital signature
latively uncommon for headers from one library
to require headers from other dependency libraries, which is why such -dev
interdependencies are not as common as they might be.
--
Steve Langasek
postmodern programmer
signature.asc
Description: Digital signature
any patches to the BTS and ask someone to do an NMU
for the bug, rather than asking someone to upload based on packages that
you've prepared separately.
To ask for an NMU, debian-devel may be a better list.
Cheers,
--
Steve Langasek
postmodern programmer
signature.asc
Description: Digital signature
denly deemed critical for the architecture after
we were already 3 months into a toolchain freeze is another question.
FWIW, these questions seem more appropriate to debian-devel than
debian-mentors; these are not what I would call novice packaging
questions. :)
--
Steve Langasek
postmodern programmer
signature.asc
Description: Digital signature
AIL PROTECTED])~$
> Then I would do `dpkg -S libSM.so.6` and similar for every other library.
> Then just take -dev version of every package.
Please always use objdump -p /usr/bin/xwrits | grep NEEDED instead. If you
use ldd, it will report indirect lib dependencies as well, resulting in the
same problem of excessive build deps.
--
Steve Langasek
postmodern programmer
signature.asc
Description: Digital signature
t requests *for* the new upstream release; if there
are changes *in* the new upstream release that fix reported bugs, please
enumerate those changes before closing the bugs in the changelog.
--
Steve Langasek
postmodern programmer
signature.asc
Description: Digital signature
On Wed, Jan 12, 2005 at 09:29:40AM +0100, Osamu Aoki wrote:
> On Sat, Dec 18, 2004 at 05:32:48PM -0800, Steve Langasek wrote:
> > On Sun, Dec 19, 2004 at 02:48:02AM +0200, Antti-Juhani Kaijanaho wrote:
> > > Like with many other things in Debian, how you do it doesn't matt
n.
If for some reason this isn't possible, then and only then should you use
NEWS.Debian to document the issue. But not a debconf note -- debconf notes
are not and never were intended for this.
--
Steve Langasek
postmodern programmer
signature.asc
Description: Digital signature
> do then ?
wxwidgets2.5 does not depend on gtk1.
--
Steve Langasek
postmodern programmer
signature.asc
Description: Digital signature
g "j2re1.5 or better") that you
could use as an alternative. Barring that, your only choices are to list
all the existing alternatives one by one, or to remove the alternative
completely and just depend on j2re1.4 (which is much uglier for users).
--
Steve Langasek
postmodern programmer
signature.asc
Description: Digital signature
-z,defs", but if you're not building that way it doesn't matter. If
you are going to keep the shared code in a separate .so, it must stay in
/usr/lib, and you must set an SONAME for it.
--
Steve Langasek
postmodern programmer
signature.asc
Description: Digital signature
libmaserver into maserver, you get a minimal
*reduction* in the total install size of the package, you get a minimal
reduction in the startup time for the application, and you get policy
compliance. Sounds like a winning strategy to me.
--
Steve Langasek
postmodern programmer
signature.asc
Description: Digital signature
il someone creates these packages?
> No, sorry. This would violate DFSG 2 ("The program must include source
> code"). It would also prevent the security team from easily patching
> those libraries.
Which is to say, that it's allowed, but such a package cannot go to main.
--
Steve Langasek
postmodern programmer
signature.asc
Description: Digital signature
201 - 300 of 604 matches
Mail list logo
