Bug#60573: yo dats whack Goldeneye
loading Text ochrisooo hours ago im http://img444.imageshack.us/img444/6735/sqjh8.png keen problems sureIs -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Poll: Anybody using debpool?
On Friday 30 March 2007 21:34, Andreas Fester wrote: > I do. I am currently a littlebit short of time, but I also already > hacked some improvements into debpool, among them > > - Added optional File Alteration Monitor support to make the > incoming queue more responsive (plan is to also add GAMIN support) > > - Started to implement automatic tests to check if packages are > properly installed in the pool (see the tests subdirectory) > > - Added support for multi arch archives (the .package files were > not arch specific and therefore overwritten when an additional > arch-specific package was uploaded) > > - Some minor bug fixes > > The package is at > http://littletux.homelinux.org/debian/pool/main/d/debpool/ Some comments: * You've indented the main loop in bin/debpool. While I think that's how it should be, I also think it's best to undo it until we have merged all contributions. * SGI::FAM is not a module available in Debian, even in experimental. While that's not required of a package that is merely suggested by an experimental package, I think it would be preferable. I implemented the same thing using inotify (which Gamin uses) with Linux::Inotify2, which exists in Debian (liblinux-inotify2-perl) and is also simpler (doesn't need a C library and a running daemon). The drawback is that it requires a Linux 2.6.13 kernel. There is also the older dnotify, which FAM uses if available, but which I haven't found any Perl package of whatsoever. But I believe in freedom of choice and think that any number of alternatives can be made available for the user to choose from. * Did you get the init script to work? start-stop-daemon --stop --exec /usr/bin/debpool shouldn't work, since /proc/(pid)/exe points to /usr/bin/perl. -- Magnus Holmgren[EMAIL PROTECTED] (No Cc of list mail needed, thanks) pgp6n5HAFoXpT.pgp Description: PGP signature
Re: net-tools maintenance status
On 8/2/05, Olaf van der Spek <[EMAIL PROTECTED]> wrote: Hi, What's the maintenance status of the net-tools package? It has 88 bugs: Serious policy violations - outstanding (1 bug) Important bugs - outstanding (8 bugs) Normal bugs - outstanding (38 bugs) Minor bugs - outstanding (9 bugs) Wishlist items - outstanding (32 bugs) A lot older than a year, 17 with patch a lot even without a single response. http://bugs.debian.org/cgi-bin/pkgreport.cgi?pkg=net-tools http://packages.qa.debian.org/n/net-tools.html Hi again, I'm sorry to bump this so soon, as it's only less than two years since I send the original mail. There are now 107 bugs, a lot older than three years, 18 with a patch and a lot without any response. Also, there hasn't been a single upload in 2006 and 2007. Please CC me, I'm not subscribed to the list. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
When to upload a package which fixes FTBFS with GCC 4.3 ?
Le Mon, Apr 02, 2007 at 10:21:22PM +0100, Martin Michlmayr a écrit : > > Your package fails to build with GCC 4.3. Version 4.3 has not been > released yet but I'm building with a snapshot in order to find errors > and give people an advance warning. In GCC 4.3, the C++ header > dependencies have been cleaned up. The advantage of this is that > programs will compile faster. The downside is that you actually > need to directly #include everything you use (but you really should > do this anyway, otherwise your program won't work with any compiler > other than GCC). Some background of this can be found at > http://gcc.gnu.org/PR28080 Dear Martin, thanks a lot for your work ! I am very impressed by your tenacity (I got a similar bug for a package starting with the A letter one month ago). I have a question : I read in the link you provided that binary program will be smaller if compiled with gcc 4.3. Does it mean that I should wait for gcc 4.3 to become the default compiler before uploading a new version which fixes the bug, in order to take advantage of a rebuild, or will there be a mass rebuilding anyway ? (CCing to -devel in the hope that the question is not too naive and has a general interest.) Have a nice day, -- Charles Plessy http://charles.plessy.org Wako, Saitama, Japan -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: When to upload a package which fixes FTBFS with GCC 4.3 ?
Charles Plessy <[EMAIL PROTECTED]> writes: > Le Mon, Apr 02, 2007 at 10:21:22PM +0100, Martin Michlmayr a écrit : >> >> Your package fails to build with GCC 4.3. Version 4.3 has not been >> released yet but I'm building with a snapshot in order to find errors >> and give people an advance warning. In GCC 4.3, the C++ header >> dependencies have been cleaned up. The advantage of this is that >> programs will compile faster. The downside is that you actually >> need to directly #include everything you use (but you really should >> do this anyway, otherwise your program won't work with any compiler >> other than GCC). Some background of this can be found at >> http://gcc.gnu.org/PR28080 > > I have a question : I read in the link you provided that > binary program will be smaller if compiled with gcc 4.3. Does it mean > that I should wait for gcc 4.3 to become the default compiler before > uploading a new version which fixes the bug, in order to take > advantage of a rebuild, or will there be a mass rebuilding anyway ? > > (CCing to -devel in the hope that the question is not too naive and > has a general interest.) Fixing the bugs well in advance of a compiler transition makes the transition far smoother. It should be trivial to schedule binNMUs once the new compiler is the default, and it's probably going to be a while for 4.3 to be the default, given that 4.2 isn't yet released. Given that all these fixes are in fact bugs due to missing headers, I can't see any reason not to fix them at the earliest opportunity. Personally, I'll be uploading fixed versions of my packages as soon as etch is released. -- .''`. Roger Leigh : :' : Debian GNU/Linux http://people.debian.org/~rleigh/ `. `' Printing on GNU/Linux? http://gutenprint.sourceforge.net/ `-GPG Public Key: 0x25BFB848 Please GPG sign your mail. pgpxp8VTYSqkW.pgp Description: PGP signature
Re: [Debconf-announce] DebConf7 - Reconfirmation phase started
2007/4/2, Joerg Jaspert <[EMAIL PROTECTED]>: Reminder: DebConf7 Reconfirmation Phase --- We would like to ask you to reconfirm your attendance by May 3. Please go to https://penta.debconf.org/submission/dc7/person this page isn't working... ** The connection was reset The connection to the server was reset while the page was loading. ** Greeting! -- Dydier J. Rojas Guerrero - (+591) 761 7/(4)6434867 GPG Key: 1024D/739E6295 2006-11-07 Fingerprint: 6D1C 9FB7 98BB 06C0 D7CF 587E 9D29 AE3E 739E 6295 Usuario Linux: 408603 [http://counter.li.org] Sucre, Bolivia
Bug#72140: WebsiteOur
Press Release selected http://img444.imageshack.us/my.php?image=fshc6.jpg products address developer needs -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#224692: log above notebook
Type TDMA No UK http://img444.imageshack.us/my.php?image=cbrhav4.gif Asia GPRS technology characters Difference -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#279983: USD hour
at offshore center India leveraging superior workforce http://img444.imageshack.us/my.php?image=3ocu4.gif Best Magnus dkmisra Lucknow UR Nirmal Surat -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#354417: Many claimants are using
with meanAnswer according aA means descramble http://img444.imageshack.us/img444/8921/ehadfy5.jpg Sender by WayRedmond WA USA -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#98549: programs files
Start includes searching browser additional http://img444.imageshack.us/my.php?image=wkqqwt0.png are not MHz Super VGA -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#166937: leave executing hopefully worst
MetaSolv IPBased Activation http://img444.imageshack.us/my.php?image=2dgc7.gif red eye adjust exposure color -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#78782: Alsop Bosss
videocasts Onnet provided early CompuServe http://img444.imageshack.us/my.php?image=nejhij6.jpg point might bad health instance -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: net-tools maintenance status
Hi, On Tue, Apr 03, 2007 at 02:09:00PM +0200, Olaf van der Spek wrote: > On 8/2/05, Olaf van der Spek <[EMAIL PROTECTED]> wrote: > >What's the maintenance status of the net-tools package? Are you applying for co-maintenance? If so, you should rather talk the the package maintainer, I think. Alternatively, you could do some bug triaging and/or trying to reproduce the older bugs, if you haven't done so already. Michael -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#148993: prayibg manti nothi
seril numger wdedlands freee http://img444.imageshack.us/my.php?image=wgdpci2.gif meaning ofn ames parks -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#199653: designed support
Art Science ROINetwork http://img444.imageshack.us/my.php?image=unqy9.png Sequential August Detecting -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#62878: Inc rights reserved Privacy
TricksCNET TVCompare Prices Life Blogs Extra My http://img444.imageshack.us/img444/3350/jsxm5.gif green office space -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#354417: Gates
Germany prove successful addressing problemIn word useful http://img444.imageshack.us/my.php?image=cq0ws6.png window Iraniansin way help empower -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
help requested [EMAIL PROTECTED]: Re: Bug#366146: libfann: new upstream release includes debian packaging]
According to the bug report, the upstream author of this package wants help. As you can see in the message below, upstream can take care of the actual packaging. However, he needs help on a couple of fine points of the package. Also, it seems as if he is in need of a new sponsor to get this package actually updated. I can't sponsor it myself since I'm not a DD. - Forwarded message from Steffen Nissen <[EMAIL PROTECTED]> - Subject: Re: Bug#366146: libfann: new upstream release includes debian packaging The 2.0.0 package was made and uploaded, but it was rejected due to some problems with the versioning. I was not complete sure how to alter the package so that it could be accepted, and due to various circumstances a new package was not made at that point. Later on I tried to contact the sponsor for the package, in order to get his help, but with no luck. If anyone is interested in sponsoring the package and help creating the new package, please let me know. Best Regards, Steffen On 01/04/07, Allan Lyons <[EMAIL PROTECTED]> wrote: > >Not only has a new upstream release be available for quite a while, the >upstream source tree seems to have all of the files for building a deb >package. Is there a reason why the newer packages were not uploaded >back in 2005 and 2006? (I'm assuming that they weren't uploaded since >packages.qa.debian.org says that unstable is still at 1.2.0-1) > > -- Steffen Nissen - http://MySpace.com/SteffenNissen Project Administrator - Fast Artificial Neural Network Library (fann) http://fann.sf.net - End forwarded message - -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Bug#416490: acpi-support: IBM Ultrabase X4 DVD/CDOM Undetected On Dock
Hi, On Wed, 28 Mar 2007, Sheridan Hutchinson wrote: > I've noticed that if I have my Thinkpad X40 seated in its X4 Ultrabase > and boot it up fresh, that I have a DVDROM (hdc). > > However, if I boot up without it, then go to dock, the DVDROM is not > detected :( > > I noticed that if hotswap is installed, I could run: > hotswap -c 1 rescan-ide > > and this will register /dev/hdc > > Which is pretty nice result.. however, although I have a /dev/hdc > /dev/dvd (which some things depend on) wasn't brought up. > > Debian should be able to handle automatically detecting IDE > CDROM's after docking by now, it is 2007 after all! :-) I would like this too. Unfortunately, I have no clue on how to detect that we're plugged into a dock. And it's not even clear that acpi-support is really the package that should take care of that. After all, acpi-support is only a set of hacks so that supend/resume works with as many laptops as possible. I'm ccing debian-devel to have some more comments on this issue. > Furthermore, ideally, I should be able to boot up my laptop out of the > dock, hibernate it (which works), plug it into the dock, wake it up from > its torpid state, and have a working IDE CDROM on /dev/hdc and > /dev/dvd What happens currently when you wake it up? Do you have no cdrom at all, or do you simply have it as /dev/hdc only? > That said, it's not a huge bug but it would lend a certain amount of > functionality. Feel free to provide a patch. :-) I don't know it it's reasonable to call hotswap systematically after a resume (if it's installed). For everybody: new (co-)maintainers for acpi-support are always welcome as I'm certainly not the best maintainer for that package (I don't care enough about it, I simply wanted it to make my laptop suspend/resume). Also such a package would benefit from having several maintainers with different hardware configurations... Cheers, -- Raphaël Hertzog Premier livre français sur Debian GNU/Linux : http://www.ouaza.com/livre/admin-debian/ -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Modifying /etc/apt/sources.list in postinst ; determining the suite in postinst
Two related problems, both in the same package: emdebian-tools[0] [1]. Neither are etch related, this only affects the next release of the package as it needs to work within a chroot. (No version of emdebian-tools exists in testing.) To install a cross-building toolchain, apt/aptitude needs to be told about the Emdebian [2] toolchain repository [3]. Currently, that is done by the emsetup program in emdebian-tools but I'm wondering if it is better done in postinst. The question then is: how to determine the suite (only unstable, testing, sid, etch and lenny are supported by Emdebian) from postinst (using Perl)? I'm currently parsing a call to 'apt-cache policy' within emsetup but is that OK to call in postinst? There is also a problem with installations created using debootstrap or which end up with a single apt source of ftp.debian.org by some other means - this mirror does not support all architectures. In particular, it does not support the typical cross-building architectures like arm. In these situations, or when any installation ends up with a non-primary mirror, emdebian-tools needs to add a primary mirror to the sources list. How can this be determined cleanly? Which primary mirror? (preferably not another debconf question). Finally, if this is done in postinst, presumably the changes will have to be removed in postrm or can dpkg be persuaded to do this for me? (Could I ship a sources.list file in the package and move the previous one to sources.dpkg-old?) update-manager and netselect-apt already modify /etc/apt/sources.list but not in postinst. [0] http://packages.debian.org/unstable/utils/emdebian-tools [1] http://buildd.emdebian.org/svn/browser/current/host/trunk/emdebian-tools/trunk/ [2] http://www.emdebian.org/ [3] http://www.emdebian.org/debian/pool/main/ -- Neil Williams = http://www.data-freedom.org/ http://www.nosoftwarepatents.com/ http://www.linux.codehelp.co.uk/ pgppzZrer0fJw.pgp Description: PGP signature
Bug#128852: bya withthem xorg org
have own into government strategy difficult appear http://img444.imageshack.us/my.php?image=q7hw0.gif devices lets sync instead syncs Contacts Memos -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#128852: forums extended tech
Give Investors MyPoints MySite Juno http://img444.imageshack.us/my.php?image=cxnlsi0.jpg amore secure uptodate myself everbeing infected virus -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#25837: perpixel alpha channel PNG
Refresh Go keyboard http://img444.imageshack.us/my.php?image=o0mwyy6.jpg checking see if Microsoft -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Bug#416490: acpi-support: IBM Ultrabase X4 DVD/CDOM Undetected On Dock
(Please keep the BTS in copy when replying) On Tue, 03 Apr 2007, Sheridan Hutchinson wrote: > Raphael Hertzog wrote: > >I would like this too. Unfortunately, I have no clue on how to detect > >that we're plugged into a dock. And it's not even clear that acpi-support > >is really the package that should take care of that. After all, > >acpi-support is only a set of hacks so that supend/resume works > >with as many laptops as possible. > > Detection may be possible with values in: > /proc/acpi/ibm/bay > or maybe (could be obsolete) > /proc/acpi/ibm/dock Can you give sample values that those file are exporting ? (I have no IBM laptop and don't feel like searching for documentation on this specific aspect) Feel free to point us to any existing documentation on that topic. > >Feel free to provide a patch. :-) > > My programming skills might not be up to these intricacies, how about a > donation to the Debian project instead? :P We're lacking more man-power than money. But any way to help us is appreciated. ;-) > >I don't know it it's reasonable to call hotswap systematically after a > >resume (if it's installed). > > As an acpi script it's probably not reasonable, it might be easier to > look to see exactly what hotswap does in this context. The more interesting thing is to know if the kernel is aware of the fact that something has been hot-plugged and how can it give us this information. I fear that the IDE bus is not really hotplug friendly and that we simply have no way to discover this automatically (hence the need of a tool to ask for a rescan of the IDE bus by the kernel). But I'm neither an hardware expert nor a kernel specialist, so I may be wrong. I hope someone from debian-devel can enlighten us on this aspect. Cheers, -- Raphaël Hertzog Premier livre français sur Debian GNU/Linux : http://www.ouaza.com/livre/admin-debian/ -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: When to upload a package which fixes FTBFS with GCC 4.3 ?
* Charles Plessy <[EMAIL PROTECTED]> [2007-04-03 21:55]: > I have a question : I read in the link you provided that binary > program will be smaller if compiled with gcc 4.3. Does it mean that > I should wait for gcc 4.3 to become the default compiler before > uploading a new version which fixes the bug, in order to take > advantage of a rebuild, or will there be a mass rebuilding anyway ? Actually, I said compilation time, not size. Anyway, while there is no rush since GCC 4.3 is still far away, it would obviously be good to get these bugs fixed soon since this will give us plenty of time to find more bugs. -- Martin Michlmayr http://www.cyrius.com/ -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Modifying /etc/apt/sources.list in postinst ; determining the suite in postinst
On ti, 2007-04-03 at 16:28 +0100, Neil Williams wrote: > To install a cross-building toolchain, apt/aptitude needs to be told > about the Emdebian [2] toolchain repository [3]. Currently, that is done > by the emsetup program in emdebian-tools but I'm wondering if it is > better done in postinst. The question then is: how to determine the > suite (only unstable, testing, sid, etch and lenny are supported by > Emdebian) from postinst (using Perl)? I'm currently parsing a call to > 'apt-cache policy' within emsetup but is that OK to call in postinst? With regards to Debian Policy, you need to read section 10.7. Basically, since sources.list is a configuration file belonging to apt, your package may only edit it in postinst if apt provides commands for doing so. I don't think apt does that, but I could be wrong. With regards to the life of a sysadmin, don't do it. Changing sources.list will surprise a lot of people, and can cause bad things to happen, for example higher network bills because the wrong mirror is used. Provide a README.Debian that tells what needs to be done, and possibly provide a command to do it easily. -- It's 1978! Things should be round by now -- Michael Kelso (That 70's show) -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Modifying /etc/apt/sources.list in postinst ; determining the suite in postinst
* Neil Williams: > Finally, if this is done in postinst, presumably the changes will have > to be removed in postrm or can dpkg be persuaded to do this for me? > (Could I ship a sources.list file in the package and move the previous > one to sources.dpkg-old?) There is /etc/apt/sources.list.d. However, I don't think official Debian packages should enable non-official package sources. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Modifying /etc/apt/sources.list in postinst ; determining the suite in postinst
On Tue, 03 Apr 2007 19:27:17 + Lars Wirzenius <[EMAIL PROTECTED]> wrote: > On ti, 2007-04-03 at 16:28 +0100, Neil Williams wrote: > With regards to Debian Policy, you need to read section 10.7. That's why I'm changing the existing method (which does simply append data to /etc/apt/sources.list). The previous hack isn't good. :-) However, determining the suite from within postinst is my main issue at the moment. Is it safe to use `apt-cache policy` ? > Basically, since sources.list is a configuration file belonging to > apt, your package may only edit it in postinst if apt provides > commands for doing so. I don't think apt does that, but I could be > wrong. It does have a method: /etc/apt/sources.list.d/ I'm going to put a sources file in there which can be purged when the package is purged. > With regards to the life of a sysadmin, don't do it. I can't see sysadmins having to manage many cross-building buildd machines but I will provide a README.Debian. The Emdebian repository only stores the cross-building toolchains. > Provide a README.Debian that tells what needs to be done, and > possibly provide a command to do it easily. The package is unusable without the added sources. Providing a userspace command doesn't ensure that the repository is removed from the sources list when the package is purged. The modified package is currently in the Emdebian repository, v0.1.4. -- Neil Williams = http://www.data-freedom.org/ http://www.nosoftwarepatents.com/ http://www.linux.codehelp.co.uk/ pgpkAAcmZ7NWp.pgp Description: PGP signature
Re: Modifying /etc/apt/sources.list in postinst ; determining the suite in postinst
On Tue, 03 Apr 2007 21:43:43 +0200 Florian Weimer <[EMAIL PROTECTED]> wrote: > * Neil Williams: > > > Finally, if this is done in postinst, presumably the changes will > > have to be removed in postrm or can dpkg be persuaded to do this > > for me? (Could I ship a sources.list file in the package and move > > the previous one to sources.dpkg-old?) > > There is /etc/apt/sources.list.d. Yes, I found that after posting the original query. Thanks. It is easy to implement that in postinst and postrm. > However, I don't think official > Debian packages should enable non-official package sources. Emdebian is an official subproject of Debian - does that make a difference? None of the scripts in the package are usable without this repository and the toolchain packages that are available. These aren't arbitrary packages, they are cross-built versions of official Debian packages. What about determining the suite? -- Neil Williams = http://www.data-freedom.org/ http://www.nosoftwarepatents.com/ http://www.linux.codehelp.co.uk/ pgpwDlNEVZKoJ.pgp Description: PGP signature
Override CFLAGS when using cdbs
Hi, I try to package a project that needs to add some include pathes to the existing default CFLAGS and thus I tried to patch a Makefile in one subdir to CFLAGS=$(CFLAGS) -I$(INCLUDEDIR) but I had to learn that the CFLAGS variable that is used in the call cd subtree; $(MAKE) just overrides any CFLAGS setting and the line above is just ignored. Is there any chance to force setting of CFLAGS in the makefile? (Sorry, I guess this is a faq but Google was not friendly to my search atempts.) Kind regards Andreas. -- http://fam-tille.de -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Modifying /etc/apt/sources.list in postinst ; determining the suite in postinst
On Tue, 2007-04-03 at 21:14 +0100, Neil Williams wrote:
> What about determining the suite?
You should probably use debconf to ask, though. I wouldn't be surprised
to find that some Emdebian users want a stable host and an unstable
target.
If you still want to find the default suite, you should be able to do it
by parsing apt-cache policy output, though I'm not sure how stable that
format is. Here's some Perl code that worked for me:
my ($cur_pri, $max_pri, $max_pri_suite);
open POLICY, "apt-cache policy |" or die "open: $!";
for () {
if (/^\s*(\d+)/) {
$cur_pri = $1;
} elsif (defined($cur_pri)
&& (!defined($max_pri) || $cur_pri > $max_pri)
&& /^\s+release o=Debian,a=(\w+),l=Debian,c=main\s*$/) {
$max_pri = $cur_pri;
$max_pri_suite = $1;
}
}
close POLICY;
if (defined($max_pri_suite)) {
print "Default suite is $max_pri_suite\n";
} else {
print "No sources for Debian main\n";
}
### END ###
Ben.
--
Ben Hutchings
I'm not a reverse psychological virus. Please don't copy me into your sig.
signature.asc
Description: This is a digitally signed message part
