Re: ITA wmakerconf, wmakerconf-data (will need a sponsor)
Kevin B. McCarty wrote: I am adopting wmakerconf and wmakerconf-data packages -- I've already debianized the newest upstream versions (2.9 and 0.80.0 respectively) and fixed a few outstanding bugs. Being a non-maintainer, I'll need a sponsor -- would anyone like to volunteer? i have agreed to sponsor these packages. -john -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: ITA wmakerconf, wmakerconf-data (will need a sponsor)
Kevin B. McCarty wrote: I am adopting wmakerconf and wmakerconf-data packages -- I've already debianized the newest upstream versions (2.9 and 0.80.0 respectively) and fixed a few outstanding bugs. Being a non-maintainer, I'll need a sponsor -- would anyone like to volunteer? i have agreed to sponsor these packages. -john
Re: Weird broken dependencies according to qa.debian.org
Karl E. Jorgensen wrote: According to: http://qa.debian.org/debcheck.php?dist=unstablepackage=battery-stats My package declares a build time dependency on libapm-dev which cannot be satisfied on sh. SuperH processor. i am not sure what systems use a SuperH CPU, though i think it is on the order of hand-held devices. -john -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Weird broken dependencies according to qa.debian.org
Karl E. Jorgensen wrote: According to: http://qa.debian.org/debcheck.php?dist=unstablepackage=battery-stats My package declares a build time dependency on libapm-dev which cannot be satisfied on sh. SuperH processor. i am not sure what systems use a SuperH CPU, though i think it is on the order of hand-held devices. -john
Re: How to detect if Sendmail/Postfix/Qmail is installed
Brian M. Carlson wrote: On Fri, Feb 21, 2003 at 09:41:18AM +0100, Xavier Roche wrote: In this case, should the package just document what the use should do (it will be different for each MTA) or at least detect which MTA is installed and copy an example config file Ensure that the depends: field in control has something like Depends: (sendmail | qmail | postfix) It would be even better if you could make it work with a generic mail-transport-agent. to throw even more monkeywrenches in the gearbox, some people may have compiled an MTA outside of the packaging system, and be using qmail/ sendmail/postfix (the uspported MTA's) but not have it registered with apt/dpkg at all. it is at your discretion to not support such a setup, of course. -john -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: How to detect if Sendmail/Postfix/Qmail is installed
Brian M. Carlson wrote: On Fri, Feb 21, 2003 at 09:41:18AM +0100, Xavier Roche wrote: In this case, should the package just document what the use should do (it will be different for each MTA) or at least detect which MTA is installed and copy an example config file Ensure that the depends: field in control has something like Depends: (sendmail | qmail | postfix) It would be even better if you could make it work with a generic mail-transport-agent. to throw even more monkeywrenches in the gearbox, some people may have compiled an MTA outside of the packaging system, and be using qmail/ sendmail/postfix (the uspported MTA's) but not have it registered with apt/dpkg at all. it is at your discretion to not support such a setup, of course. -john
Re: signing a GPG key with multiple uids
Osamu Aoki wrote: On Wed, Dec 04, 2002 at 03:05:57AM +0100, Rene Engelhard wrote: which have that address in it. I sign a uid when these uid's address is not bouncing and the person who claims to belong to this key answers a message encrypted to him sent to the specific uid. If the person answers to all the mails sent to him, I can sign all uid's. This sounds like good practice but burden of proof for the activeness of e-mail account is on signer side. A bit unfiar, IMHO. this is as it should be. a signer needs to take Due Diligence when saying ``Yes. I know that this key matches this Name and EMail address.'' failure to do that renders that signature, and potentially all other signatures made by that signer. the whole Web-of-Trust thing. some people do take more care than others when signing, and that is okay. but the onus is always on the signer to verify that the facts as she understands them are true. -john -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: signing a GPG key with multiple uids
Osamu Aoki wrote: On Wed, Dec 04, 2002 at 03:05:57AM +0100, Rene Engelhard wrote: which have that address in it. I sign a uid when these uid's address is not bouncing and the person who claims to belong to this key answers a message encrypted to him sent to the specific uid. If the person answers to all the mails sent to him, I can sign all uid's. This sounds like good practice but burden of proof for the activeness of e-mail account is on signer side. A bit unfiar, IMHO. this is as it should be. a signer needs to take Due Diligence when saying ``Yes. I know that this key matches this Name and EMail address.'' failure to do that renders that signature, and potentially all other signatures made by that signer. the whole Web-of-Trust thing. some people do take more care than others when signing, and that is okay. but the onus is always on the signer to verify that the facts as she understands them are true. -john
Re: user for getting files
David Roundy wrote: On Fri, Nov 22, 2002 at 11:04:27AM -0800, Blars Blarson wrote: but is it realy appropriate for a package to create a new user for a weekly download? depending upon what other processes will be using the data, and those that will be starting them, the answer varies from ``maybe'' to ``yes!'' I'm far from an expert, but I would have thought that 'nobody' would be appropriate for this. _no_. you never want a file owned by nobody. services that do not need any elevated privedges should run as nobody, so if they are compromised, then can do nothing. if you download a file as nobody, then a compromised nobudy-running daemon can then trojan that file. bad. As long as you don't trust the content of those files, this seems safe to me. there are other reasons to not trust the file, other than the ownership (dns cache poisoning, dns takeover, trojans on the server), so we can accept this as a truism. -john
Re: game with small library
Matt Zimmerman wrote: If the shared library is internal to the package, it should not be installed in the system library path (e.g., /usr/lib). Instead, install it under /usr/lib/package. You will need to use a wrapper script to add /usr/lib/package to LD_LIBRARY_PATH wehn running the program. what is so wrong with setting the rpath in the binary itself? -john -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: game with small library
Matt Zimmerman wrote: If the shared library is internal to the package, it should not be installed in the system library path (e.g., /usr/lib). Instead, install it under /usr/lib/package. You will need to use a wrapper script to add /usr/lib/package to LD_LIBRARY_PATH wehn running the program. what is so wrong with setting the rpath in the binary itself? -john
Re: debhelper I have no package to build
On Fri, May 17, 2002 at 10:01:14PM -0400, Steve M. Robbins wrote: The binary-common rule has lots of dh_* stuff there, including dh_compress. I decided that compressing the HTML files would be a bad idea, so I modified it to skip that one package using the dh_compress man page says: By default, dh_compress compresses files that debian pol icy mandates should be compressed, namely all files in usr/share/info, usr/share/man, usr/X11R6/man, files in usr/share/doc that are larger than 4k in size, (except the copyright file, .html files, and files that appear to be already compressed based on their extensions), and all changelog files. dh_compress contains the following snippet: (spacing changed) find usr/doc usr/share/doc -type f \\( -size +4k -or -name changelog* \\) \\ \\( -name changelog.html -or ! -iname *.htm* \ so i would venture to guess that the .html files are already skipped, and you need not do anything additional. -john -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Wcal packaging question
On Wed, Jan 23, 2002 at 07:04:25PM +0100, Wouter Eerdekens wrote: Another thing is the apache config: it must be edited before one can use wcal: I added the following: there are other webservers besides apache. like boa. is there a generic handler to add the cgi information to the webserver backends? i am not aware of any. i doubt that wcal is so closely welded to the apache webserver that it cannot work with any of the other webservers available. i have no solutions for you, but i do want to bring up the awareness issue: there is more than one httpd. -john i count a dozen in sid: % grep-available --count -F Provides httpd 12
Re: Help on policy
On Mon, Nov 26, 2001 at 09:28:47AM +0100, Falco Cesare wrote: What would be best? 1. A non-free debian package including both sources and needed roms 2. Two debian packages: a contrib one including the source and a non-free one with the roms only i'd go the latter route. if there are any DFSG free ROM's, then the DFSG free ROMS and the emulator can be main. now, to speak of your copyright issue, for example, linus owns the copyright to the linux kernel. the free software foundation owns the copyright to the GNU utilities. if the copyright ownership is the only thing that prevents the ROM's from being DFSG free, then consider that a moot point. not having read the license of the ROM's in question, i cannot tell you if that is the case or not. if you have questions, ask debian-legal, if you have not already. -john -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: your mail
On Mon, Nov 12, 2001 at 12:38:46PM +0100, Gundolf Kiefer wrote: I have written a piece of software that I would like to offer for inclusion into a future release of Debian I am not a Debian developer, How can I proceed further? one of two ways: you can apply to become a debian developer http://www.debian.org/devel/join/newmaint or you can file a Request for Package be sending an email to the Bug Tracking System. http://www.debian.org/devel/wnpp/ describes the proper format. whichever way you decide to go, i wish you luck! -john -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: your mail
On Mon, Nov 12, 2001 at 12:38:46PM +0100, Gundolf Kiefer wrote: I have written a piece of software that I would like to offer for inclusion into a future release of Debian I am not a Debian developer, How can I proceed further? one of two ways: you can apply to become a debian developer http://www.debian.org/devel/join/newmaint or you can file a Request for Package be sending an email to the Bug Tracking System. http://www.debian.org/devel/wnpp/ describes the proper format. whichever way you decide to go, i wish you luck! -john
Re: Playing with dpkg's mind
On Wed, Nov 07, 2001 at 11:56:34AM +0100, Robert Bihlmeyer wrote: In following I call the old version of the package OP, the transition package TP, and the game file I want to keep GF. why not have TP depend upon GF1 GF2 GF3 GF4 GF5 ? TP is supposed to be a meta-package, right? upon re-reading the proposal, it looks like you want to have an installer be able to go to the FTP site, and grab the games in question. i see this as being done in basically one of three ways: 1) the task meta pacakges of yore 2) qmail-src, that gets a tarball, and then makes a .deb, which is then installed 3) mscorefonts, that gets some .exe's, extracts the fonts, and places them in the filesystem where they belong. mscorefonts gets all the fonts at the time of installation, and has an updater along with it. i am not certain that it allows the selective removal and addition of various games. if i understand correctly: you want a package that allows the insallation and removal of games on an individual basis, but will get a whole bunch at one time. the first half can be done via apt/dpkg. the second half can be done via a meta-package like the -task packages of yore. my 2¢. -john
Re: unofficial apt source setup
On Sun, Oct 07, 2001 at 09:43:38PM -0500, Scott Dier wrote: * [EMAIL PROTECTED] [EMAIL PROTECTED] [011007 19:44]: Read about the 'apt-ftparchive' program, specifically the 'packages' and 'sources' options to it. Or, if your like me and a glutton for punishment, try using dpkg-scanpackages and dpkg-scansources by hand. been there, done that, and nothing will create the Release file that can be found on the debian archives. it seems as if it is not required, but i dislike the warning messages that apt-get update gives me. does anyone know how to create a proper Release file? -john -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: unofficial apt source setup
On Sun, Oct 07, 2001 at 09:43:38PM -0500, Scott Dier wrote: * [EMAIL PROTECTED] [EMAIL PROTECTED] [011007 19:44]: Read about the 'apt-ftparchive' program, specifically the 'packages' and 'sources' options to it. Or, if your like me and a glutton for punishment, try using dpkg-scanpackages and dpkg-scansources by hand. been there, done that, and nothing will create the Release file that can be found on the debian archives. it seems as if it is not required, but i dislike the warning messages that apt-get update gives me. does anyone know how to create a proper Release file? -john
Re: [mentors] Keysign request
On Tue, Jul 03, 2001 at 04:59:03PM -0500, Taral wrote: I'd prefer to get my new key signed. Anyone? you should be able to sign the new key with the old key to get it into the keyring. -john
GPG Key Signing (Was: Advocate/Sponsor)
On Thu, Jun 28, 2001 at 12:13:37PM -0500, Steve Langasek wrote: we should also require them to demonstrate a clear understanding of PKI as part of the NM process. manoj came up with a pretty good protocol to sign a key. i have it available in HTML at http://people.debian.org/~jaqque/keysign.html it does have some weaknesses, but it is a lot stronger than the ``oh, i've met you, i have checked your ID, and off we go'' comments welcome. -john
Re: License question for truncate.sty (got from CTAN)
On Sun, Jun 10, 2001 at 08:11:43PM +0530, Viral wrote: Does this mean that I can distribute this program in accordance with the DFSG. There is no implicit license, just this statement which says that the file is released to the public domain. Would the file be still copyrighted by the author ? no. the author has recinded all rights to that product. it belongs to everyone. you may use it any way you see fit. you can turn around, change it, and claim it as your own work. this is even more free than the BSD license. -john -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: License question for truncate.sty (got from CTAN)
On Sun, Jun 10, 2001 at 08:11:43PM +0530, Viral wrote: Does this mean that I can distribute this program in accordance with the DFSG. There is no implicit license, just this statement which says that the file is released to the public domain. Would the file be still copyrighted by the author ? no. the author has recinded all rights to that product. it belongs to everyone. you may use it any way you see fit. you can turn around, change it, and claim it as your own work. this is even more free than the BSD license. -john
Re: sponsor for QEmacs maybe
On Fri, Apr 27, 2001 at 10:24:50PM +0100, Edmund GRIMLEY EVANS wrote: QEmacs is a tiny full-screen editor with UTF-8 support, a hex mode and the ability to edit huge files efficiently: http://www-stud.enst.fr/~bellard/qemacs/ have you contacted the author? has he given his blessing to QEmacs being pacakged? i ask because i came across this: Download Current snapshot (please do not redistribute since it is not finished): qemacs-0.2.tar.gz -john
Re: Way to check installed daemon
On Thu, Apr 19, 2001 at 09:41:58PM -0700, Osamu Aoki wrote: dpkg -s lprng| grep -e ^Status: install ok installed dpkg -s lpr| grep -e ^Status: install ok installed It will be very slow, though. Any good way? how about replacing dpkg -s lprng with grep -A1 '^Package: lpr$' /var/lib/dpkg/status and grep -e ^Status: install ok installed with grep -q -e ^Status: install ok installed (helps in suppressing unnecessary output) comparison times: dpkg -s lpr 5.71s user 0.37s system 35% cpu 17.182 total grep -A1 '^Package: lpr$' /var/lib/dpkg/status 0.00s user 0.04s system 155% cpu 0.026 total -john
Re: Undocumented binary
On Tue, Apr 03, 2001 at 04:28:37PM +0200, Dennis Schoen wrote: On Tue, Apr 03, 2001 at 09:49:41AM -0400, Steve M. Robbins wrote: What is the point? policy :) i hate that catch 22. can't file a bug, since the package is not in debian[1] can't use undocumented, unless you have a bug against it, can't leave a manpage out entirely, since it would fail lintian, can't get into the archive, unless it is lintian clean -john [1] okay, you do file a bug against it with the ITP, but that bug is against wnpp. somehow, filing a bug against a package that does not yet exist seems wrong to me. i end up writing a small man page. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Undocumented binary
On Tue, Apr 03, 2001 at 04:28:37PM +0200, Dennis Schoen wrote: On Tue, Apr 03, 2001 at 09:49:41AM -0400, Steve M. Robbins wrote: What is the point? policy :) i hate that catch 22. can't file a bug, since the package is not in debian[1] can't use undocumented, unless you have a bug against it, can't leave a manpage out entirely, since it would fail lintian, can't get into the archive, unless it is lintian clean -john [1] okay, you do file a bug against it with the ITP, but that bug is against wnpp. somehow, filing a bug against a package that does not yet exist seems wrong to me. i end up writing a small man page.
Re: Build-Depends
On Mon, Apr 02, 2001 at 04:46:55PM -0300, Carlos Laviola wrote: What do you do to ensure that your packages aren't missing build-depends? build it in a clean chroot, after getting all the files in build-essential and those listed in build-depends. if your compile fails, add the missing package to the chroot, and the build-depends line. continue until it compiles -john -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Build-Depends
On Mon, Apr 02, 2001 at 04:46:55PM -0300, Carlos Laviola wrote: What do you do to ensure that your packages aren't missing build-depends? build it in a clean chroot, after getting all the files in build-essential and those listed in build-depends. if your compile fails, add the missing package to the chroot, and the build-depends line. continue until it compiles -john
Re: NMU??
On Mon, Mar 19, 2001 at 10:00:46AM +0100, Peter van Rossum wrote: On Sun, Mar 18, 2001 at 10:24:27PM -0800, Osamu Aoki wrote: Please help me clarify one thing. if name in changelog != name in control file: upload is NMU Which one should be the name of real Maintainer? changelog name in .dsc file. The one in the control file (which ends up in .dsc); the one in the changelog is for who actually changed the package. So someone doing an NMU would not change the control file, but would create a new changelog entry with his or her own name. and a porter (w/o changing the source) would use the -m flag to debuild/dpkg-buildpackage ? -john
Re: kernel-module version mismatch
On Tue, Mar 13, 2001 at 05:48:13PM -0500, Andrew D Dixon wrote: Hi All, I'm writing my first module and when I try to run it I get this error: #insmod hello.o hello.o: kernel-module version mismatch hello.o was compiled for kernel version 2.2.15 while this kernel is version 2.2.18pre21. I believe that I need to upgrade linux/module.h and I was wondering what the preferred method of doing this was. /usr/include/linux/module.h correctly refers to the kernel that the libraries were built against. if you know where your current kernel headers are, you can include the path to them with the -I flag gcc -c hello.c -I/usr/src/linux/include -john -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: kernel-module version mismatch
On Tue, Mar 13, 2001 at 05:48:13PM -0500, Andrew D Dixon wrote: Hi All, I'm writing my first module and when I try to run it I get this error: #insmod hello.o hello.o: kernel-module version mismatch hello.o was compiled for kernel version 2.2.15 while this kernel is version 2.2.18pre21. I believe that I need to upgrade linux/module.h and I was wondering what the preferred method of doing this was. /usr/include/linux/module.h correctly refers to the kernel that the libraries were built against. if you know where your current kernel headers are, you can include the path to them with the -I flag gcc -c hello.c -I/usr/src/linux/include -john