Bug#732128: Manpages missing cross-references to socket stuff
Hi Ian, Fixed upstream as commit r1586: http://bazaar.launchpad.net/~upstart-devel/upstart/trunk/revision/1586 upstart-events(7) changes applied to ubuntu branch as commit r1533: http://bazaar.launchpad.net/~ubuntu-branches/ubuntu/trusty/upstart/trusty/revision/1533 Note that prior to this bug being raised, init(5) and init(8) made reference to upstart-events(7). This could have been construed as a error but I think we can handily argue that all distro installations of Upstart _should_ provide upstart-events(7) with appropriate content which ties that issue up rather neatly :-) Kind regards, James. -- James Hunt #upstart on freenode http://upstart.ubuntu.com/cookbook https://lists.ubuntu.com/mailman/listinfo/upstart-devel -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Bug#732131: Please document apparmor directives
Hi Ian, Apparmor is covered in init(5) section "AppArmor support". Did you maybe search for "apparmour" by mistake? Kind regards, James. -- James Hunt #upstart on freenode http://upstart.ubuntu.com/cookbook https://lists.ubuntu.com/mailman/listinfo/upstart-devel -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Bug#727708: tech-ctte: Decide which init system to default to in Debian.
On Mon, Dec 30, 2013 at 02:03:22 -0800, Steve Langasek wrote:
> The upstart "session init" runs as the user, not as root.
Note that a session init can run as root ("sudo init --user") but yes,
conventionally they are run as non-priv users.
> I'm not sure if
> upstart as a user session has any dependencies on upstart being PID 1.
> Cc:ing James, who would know better - James, do you know if upstart session
> init works on non-upstart systems?
I've got a Session Init running fine on Wheezy with SysVinit as PID 1.
To do this:
1) sudo apt-get install libnih1
2) sudo apt-get -y build-dep upstart
3) Download an upstart release (or branch the code from lp:upstart).
4) Unpack.
5) autoreconf -fi && ./configure && make
6) Manually install the following:
util/initctl from the build tree as /sbin/initctl
init/init from the build tree as /sbin/session-init.
7) Setup a few basics and create a single job to start a terminal (no WM :)
mkdir -p ~/.config/upstart
mkdir -p ~/.cache/upstart
cat >>~/.xsession<>~/.config/upstart/terminal.conf
Bug#732123: usage of "stanza" not compatible with ordinary English
Hi Ian, = Definition = I agree that the word 'stanza' is somewhat of an uncommon term in the realm of computing, however there is precedent in the domain of XML [1]. The usage in that spec admittedly implies a sense of plurality and yet an XML element could of course span multiple lines but actually represent an empty value: According to wiktionary and Websters [2], stanza can also mean a division or simply a poetic unit without prescribing a minimum length (as the OED does). I suspect (?) Scott might have been thinking along those lines when he adopted the term. = Multi-line singular stanzas in Upstart = Note that even those singular Upstart stanzas such as 'author' and 'console' can span multiple lines using the usual back-slashing technique if so required: console \ \ \ log author \ \ "james" That facility again lends weight to the use of the term for any type of Upstart stanza; each stanza is a logical unit the encapsulates certain meta-data or requirements for the job environment. = Occurrences of the term in the code = Interestingly, aside from the man pages, there are only 2 occurrences of the word stanza in the code that users will ever see: 1) init --help|grep -iB1 stanza (reformatted below) --default-console=VALUE default value for console stanza Clearly, this would be trivial to change if required. 2) The error "Unknown stanza" resulting from failing to parse an invalid job configuration file. This error comes not from Upstart but from the dependent libnih which actually performs the safe parsing of all the Upstart job configuration files. I'm not convinced we should change libnih. In fact, it is Scott we'd really need to convince since the stanza code in libnih has been around since ~2007 fwics. = Proposed Plan = In summary, would you be happy if we defined 'stanza' more clearly in the manual pages, making reference to terms such as 'command' and 'keyword' and giving examples of single-line and multi-line stanzas so there is no misunderstanding as to what is meant by the term? Kind regards, James. [1] - http://xmpp.org/rfcs/rfc3920.html#stanzas [2] - http://en.wiktionary.org/wiki/stanza -- James Hunt #upstart on freenode http://upstart.ubuntu.com/cookbook https://lists.ubuntu.com/mailman/listinfo/upstart-devel -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Bug#732127: Does setuid also set the group(s) ? It should.
Hi Ian, I've updated the man pages on this point (fixed in r1580): http://bazaar.launchpad.net/~upstart-devel/upstart/trunk/revision/1580 Kind regards, James. -- James Hunt #upstart on freenode http://upstart.ubuntu.com/cookbook https://lists.ubuntu.com/mailman/listinfo/upstart-devel -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Bug#732125: upstart-events(7) title is misleading
Hi Ian, As suggested, I've moved the appropriate sections to init(8) upstream as commit r1581: http://bazaar.launchpad.net/~upstart-devel/upstart/trunk/revision/1581 Note that upstart-events(7) is an Ubuntu+Debian-specific man page so debian/manpages/upstart-events.7 will need to be tweaked in Debian to: 1) Remove the Job States and Job Lifecycle sections from upstart-events.7. 2) Remove reference to Table 6 in the Event Summary section. I've already done this in Ubuntu as commit r1532: http://bazaar.launchpad.net/~ubuntu-branches/ubuntu/trusty/upstart/trusty/revision/1532 Kind regards, James. -- James Hunt #upstart on freenode http://upstart.ubuntu.com/cookbook https://lists.ubuntu.com/mailman/listinfo/upstart-devel -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Bug#732122: semantics of boolean event specification not defined in init(5)
Hi Ian, I've expanded the section on 'start on' in upstream commit r1584: http://bazaar.launchpad.net/~upstart-devel/upstart/trunk/revision/1584 Kind regards, James. -- James Hunt #upstart on freenode http://upstart.ubuntu.com/cookbook https://lists.ubuntu.com/mailman/listinfo/upstart-devel -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Bug#732126: state machine semantics of "initctl restart" are unclear
Hi Ian, Fixed in upstream commit r1585: http://bazaar.launchpad.net/~upstart-devel/upstart/trunk/revision/1585 Kind regards, James. -- James Hunt #upstart on freenode http://upstart.ubuntu.com/cookbook https://lists.ubuntu.com/mailman/listinfo/upstart-devel -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Bug#722242: debian-maintainers: Please add James Hunt as a Debian Maintainer
Thanks Jonathan, However, it appears the updated package hasn't hit the archive yet? Kind regards, James. signature.asc Description: OpenPGP digital signature
Bug#735266: lintian: ppc64el architecture not recognised
Package: lintian Version: 2.5.21 Severity: normal Dear Maintainer, I have just uploaded procenv_0.29-2 which has a Build-Depends on libnuma-dev. That library is very architecture-dependent so I have had to specify all the Linux architectures in procenv's debian/control that are relevant. However, lintian rejects 'linux-ppc64el' as a valid architecture: $ lintian --profile debian -I -E --pedantic procenv_0.29-2_source.changes N: Using profile debian/main. N: Setting up lab in /tmp/temp-lintian-lab-6dBvDYkBJw ... N: Unpacking packages in group procenv/0.29-2 N: N: Processing changes file procenv (version 0.29-2, arch source) ... N: N: Processing source package procenv (version 0.29-2, arch source) ... E: procenv source: invalid-arch-string-in-source-relation linux-ppc64el [build-depends: libnuma-dev [linux-i386 linux-amd64 linux-ia64 linux-mips linux-mipsel linux-powerpc linux-ppc64 linux-ppc64el linux-x32]] P: procenv source: debian-watch-may-check-gpg-signature $ lintian --ftp-master-rejects -I -E --pedantic procenv_0.29-2_source.changes N: Using profile debian/ftp-master-auto-reject. N: Setting up lab in /tmp/temp-lintian-lab-T_cZDOel0A ... N: Unpacking packages in group procenv/0.29-2 N: N: Processing changes file procenv (version 0.29-2, arch source) ... N: N: Processing source package procenv (version 0.29-2, arch source) ... -- System Information: Debian Release: jessie/sid APT prefers trusty-updates APT policy: (500, 'trusty-updates'), (500, 'trusty-security'), (500, 'trusty') Architecture: i386 (i686) Kernel: Linux 3.13.0-2-generic (SMP w/4 CPU cores) Locale: LANG=fr_FR.UTF-8, LC_CTYPE=fr_FR.UTF-8 (charmap=UTF-8) Shell: /bin/sh linked to /bin/dash Versions of packages lintian depends on: ii binutils 2.24-2ubuntu2 ii bzip2 1.0.6-5 ii diffstat 1.58-1 ii file 1:5.14-2ubuntu1 ii gettext0.18.3.1-1ubuntu2 ii hardening-includes 2.5ubuntu1 ii intltool-debian0.35.0+20060710.1 ii libapt-pkg-perl0.1.29build1 ii libarchive-zip-perl1.30-7 ii libclass-accessor-perl 0.34-1 ii libclone-perl 0.36-1 ii libdpkg-perl 1.17.1ubuntu1 ii libemail-valid-perl1.192-1 ii libfile-basedir-perl 0.03-1fakesync1 ii libipc-run-perl0.92-1 ii liblist-moreutils-perl 0.33-1build3 ii libparse-debianchangelog-perl 1.2.0-1ubuntu1 ii libtext-levenshtein-perl 0.06~01-2 ii libtimedate-perl 2.3000-1 ii liburi-perl1.60-1 ii man-db 2.6.5-3 ii patchutils 0.3.2-3 ii perl [libdigest-sha-perl] 5.18.1-5 ii t1utils1.37-2ubuntu1 Versions of packages lintian recommends: ii libautodie-perl 2.22-1 ii libperlio-gzip-perl 0.18-1build3 ii perl-modules [libautodie-perl] 5.18.1-5 Versions of packages lintian suggests: ii binutils-multiarch 2.24-2ubuntu2 ii dpkg-dev 1.17.1ubuntu1 ii libhtml-parser-perl3.71-1build1 ii libtext-template-perl 1.46-1 pn libyaml-perl ii xz-utils 5.1.1alpha+20120614-2ubuntu1 -- no debconf information Kind regards, James. -- James Hunt #upstart on freenode http://upstart.ubuntu.com/cookbook https://lists.ubuntu.com/mailman/listinfo/upstart-devel -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Bug#728684: procenv 0.27-2 FTBFS on kfreebsd, powerpc and sparc
Package: procenv Version: 0.27-2 Severity: normal Dear Maintainer, As shown below, procenv 0.27-2 FTBFS for sid on kFreeBSD, PowerPC and Sparc architectures: https://buildd.debian.org/status/package.php?p=procenv&suite=sid I've requested access to porter boxes to attempt to resolve this issue. Note that my local kFreeBSD system (Debian 7.1.0) builds this version correctly. -- System Information: Debian Release: wheezy/sid APT prefers trusty-updates APT policy: (500, 'trusty-updates'), (500, 'trusty-security'), (500, 'trusty') Architecture: i386 (i686) Kernel: Linux 3.12.0-1-generic (SMP w/4 CPU cores) Locale: LANG=fr_FR.UTF-8, LC_CTYPE=fr_FR.UTF-8 (charmap=UTF-8) Shell: /bin/sh linked to /bin/dash Versions of packages procenv depends on: ii libc6 2.17-93ubuntu4 procenv recommends no packages. procenv suggests no packages. -- no debconf information -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Bug#742470: patch: Incorrect usage causes a segmentation fault
Attached is a patch to resolve this issue. See: https://bugs.launchpad.net/ubuntu/+source/patch/+bug/1306412 Kind regards, James -- James Hunt --- src/patch.c 2014-04-11 11:24:37.0 +0100 +++ /tmp/patch-2.7.1.ORIG/src/patch.c 2014-04-11 11:24:00.589844724 +0100 @@ -1975,5 +1975,6 @@ cleanup (void) remove_if_needed (TMPOUTNAME, &TMPOUTNAME_needs_removal); remove_if_needed (TMPPATNAME, &TMPPATNAME_needs_removal); remove_if_needed (TMPREJNAME, &TMPREJNAME_needs_removal); - forget_output_files (); + if (files_to_output) + forget_output_files (); } signature.asc Description: Digital signature
Bug#690776: debconf can crash if gtk2 fails to import into Perl
Package: debconf
Version: 1.5.46
Severity: normal
Tags: patch
User: ubuntu-de...@lists.ubuntu.com
Usertags: origin-ubuntu quantal ubuntu-patch
Dear Maintainers,
The attached patch, although too late to apply to Ubuntu Quantal,
addresses the issue shown in Ubuntu bug 1060249:
https://bugs.launchpad.net/ubuntu/+source/debconf/+bug/1060249
It will stop debconf from crashing should 'use Gtk2;' fail, and allow it
to gracefully fall back to other frontends as designed.
* Debconf/Frontend/Gnome.pm:init(): Only 'use Gtk2;' once it has been
proven to be safe to avoid rogue Perl module behaviour from crashing
debconf (#LP: 1060249).
Thanks for considering the patch.
-- System Information:
Debian Release: wheezy/sid
APT prefers quantal-updates
APT policy: (500, 'quantal-updates'), (500, 'quantal-security'), (500,
'quantal')
Architecture: i386 (i686)
Kernel: Linux 3.5.0-17-generic (SMP w/4 CPU cores)
Locale: LANG=C, LC_CTYPE=en_GB.UTF-8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/dash
Kind regards,
James.
--
James Hunt
http://upstart.ubuntu.com/cookbook
http://upstart.ubuntu.com/cookbook/upstart_cookbook.pdf
=== modified file 'Debconf/FrontEnd/Gnome.pm'
--- Debconf/FrontEnd/Gnome.pm 2011-06-23 06:52:13 +
+++ Debconf/FrontEnd/Gnome.pm 2012-10-12 19:35:29 +
@@ -14,13 +14,6 @@
use Debconf::Encoding qw(to_Unicode);
use base qw{Debconf::FrontEnd};
-# Catch this so as not to confuse the poor users if Gtk or Gnome are not
-# installed.
-eval q{
- use Gtk2;
-};
-die "Unable to load Gtk -- is libgtk2-perl installed?\n" if $@;
-
=head1 DESCRIPTION
This FrontEnd is a Gnome UI for Debconf.
@@ -121,11 +114,35 @@
}
}
else {
+ # Catch scenario where Gtk/Gnome are not installed.
+ use Gtk2;
+
@ARGV=@ARGV_for_gnome; # temporary change at first
Gtk2->init;
+
+ # Create a window, but don't show it.
+ #
+ # This has the effect of exercising gtk a bit in an
+ # attempt to force an error either in the gtk bindings
+ # themselves, but hopefully also in
+ # gtk/glib/gsettings/etc. There is no guarantee that
+ # this alone will provoke an error, but it's a
+ # relatively safe and reasonable operation to perform
+ # and further reduces the chance of the parent debconf
+ # process ending up in an unrecoverable state.
+ my $window = Gtk2::Window->new('toplevel');
+
exit(0); # success
}
+ # Only load Gtk after the child has successfully proved it can do
+ # the same. This avoids the problem where a module calls into a
+ # native library and causes the perl interpreter to crash. When
+ # we get to here, we know that the child didn't crash, so it
+ # should be safe for us to attempt it.
+ eval q{use Gtk2;};
+ die "Unable to load Gtk -- is libgtk2-perl installed?\n" if $@;
+
my @gnome_sucks=@ARGV;
@ARGV=@ARGV_for_gnome;
Gtk2->init;
=== modified file 'debian/changelog'
Bug#607844: weird interaction between setsid and SIGTSTP
I've just committed a fix for this bug. See: https://bugs.launchpad.net/upstart/+bug/888910 Kind regards, James. -- James Hunt http://upstart.ubuntu.com/cookbook http://upstart.ubuntu.com/cookbook/upstart_cookbook.pdf -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Bug#693966: Patch.
Kind regards,
James.
--
James Hunt
http://upstart.ubuntu.com/cookbook
http://upstart.ubuntu.com/cookbook/upstart_cookbook.pdf
--- util-linux-2.20.1/term-utils/script.c 2011-10-18 12:22:26.0
+
+++ util-linux-2.20.1-NEW/term-utils/script.c 2012-11-22 11:50:44.566001705
+
@@ -359,7 +359,8 @@
#endif
tvec = time((time_t *)NULL);
my_strftime(obuf, sizeof obuf, "%c\n", localtime(&tvec));
- fprintf(fscript, _("Script started on %s"), obuf);
+if (!qflg)
+ fprintf(fscript, _("Script started on %s"), obuf);
do {
if (die && flgs == 0) {
--- util-linux-2.20.1/term-utils/script.1 2011-10-18 05:41:47.0
+
+++ util-linux-2.20.1-NEW/term-utils/script.1 2012-11-22 12:03:52.841128173
+
@@ -90,7 +90,8 @@
Allow the default output destination, i.e. the typescript file, to be a
hard or symbolic link. The command will follow a symbolic link.
.It Fl q, Fl Fl quiet
-Be quiet.
+Be quiet (do not write start and done messages to either standard output
+or the typescript file).
.It Fl t, Fl Fl timing Ns Op Ns = Ns Ar file
Output timing data to standard error, or to \fIfile\fR when given.
This data contains two fields, separated by a space.
Bug#684058: libjson0: json-c version 0.10 is now available.
On 15/08/12 18:07, Fabien Boucher wrote: > Hi, > > Ok, I'm currently working on it. Thanks Steve for that proposal but I > already > have a sponsor. Be sure you'll see the 0.10 version uploaded on > mentors.debian.net <http://mentors.debian.net> > rapidly. Thanks Fabien! Kind regards, James. -- James Hunt http://upstart.ubuntu.com/cookbook http://upstart.ubuntu.com/cookbook/upstart_cookbook.pdf -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Bug#684058: libjson0: json-c version 0.10 is now available.
Hi Fabien, I've asked on the json-c mailing list today [1] about the possibility of a json-c-0.10.tar.gz release file since both Debian and Ubuntu's watch files are currently looking for releases at http://oss.metaparadigm.com/json-c/ (it also appears that Fedora is looking in the same place). Hopefully, the JSON-C project will create this file. But if this is not the file you intend to package, please could you comment on the download location you are intending to use so we can avoid any sync issues. Kind regards, James. [1] - https://groups.google.com/group/json-c/msg/e1a69f9ce42165e6?dmode=source&output=gplain&noredirect (requires Google login I think). On 15/08/12 18:07, Fabien Boucher wrote: > Hi, > > Ok, I'm currently working on it. Thanks Steve for that proposal but I > already > have a sponsor. Be sure you'll see the 0.10 version uploaded on > mentors.debian.net <http://mentors.debian.net> > rapidly. > > Bests Regards, > Fabien > > > 2012/8/14 Steve Langasek <mailto:steve.langa...@ubuntu.com>> > > Hi Fabien, > > On Sat, Aug 11, 2012 at 10:29:27AM +0200, Fabien Boucher wrote: > > Thanks for the informations, i will update the package in a couple > of days. > > That's great to hear! Do you need a sponsor for your upload? The next > version of json-c is important as a dependency for some feature work > in the > next release of upstart, so it's important to have 0.10 in for > Ubuntu 12.10. > I'd be happy to sponsor the upload to Debian if that would be > helpful to you. > > -- > Steve Langasek Give me a lever long enough and a > Free OS > Debian Developer to set it on, and I can move the > world. > Ubuntu Developer > http://www.debian.org/ > slanga...@ubuntu.com <mailto:slanga...@ubuntu.com> > vor...@debian.org <mailto:vor...@debian.org> > > > 2012/8/6 James Hunt <mailto:james.h...@ubuntu.com>> > > > > > Package: libjson0 > > > Version: 0.9-1ubuntu1 > > > Severity: normal > > > > > > Dear Maintainer, > > > > > > There is a new version of json-c (0.10) available from github: > > > > > >https://github.com/json-c/json-c/tarball/json-c-0.10-20120530 > > > > > > Amongst other things, this release includes internal 64-bit integer > > > handling. > > > > > > Note that there is no tar file available (yet?) on the site below: > > > > > > http://oss.metaparadigm.com/json-c/ > > > > > > See my query to the json-c mailing list for further information > > > (requires a google account to access): > > > > > > > https://groups.google.com/forum/?fromgroups#!topic/json-c/QZPATk-20v0 > <https://groups.google.com/forum/?fromgroups#%21topic/json-c/QZPATk-20v0> > > > > > > Kind regards, > > > > > > James. > > > > > > -- System Information: > > > Debian Release: wheezy/sid > > > APT prefers precise-updates > > > APT policy: (500, 'precise-updates'), (500, > 'precise-security'), (500, > > > 'precise') > > > Architecture: i386 (i686) > > > > > > Kernel: Linux 3.2.0-27-generic-pae (SMP w/4 CPU cores) > > > Locale: LANG=fr_CA.UTF8, LC_CTYPE=fr_CA.UTF8 (charmap=UTF-8) > > > Shell: /bin/sh linked to /bin/dash > > > > > > Versions of packages libjson0 depends on: > > > ii libc6 2.15-0ubuntu10 > > > ii multiarch-support 2.15-0ubuntu10 > > > > > > libjson0 recommends no packages. > > > > > > libjson0 suggests no packages. > > > > > > -- no debconf information > > > > > > -- > > > Kind regards, > > > > > > James. > > > -- > > > James Hunt > > > > > > http://upstart.ubuntu.com/cookbook > > > http://upstart.ubuntu.com/cookbook/upstart_cookbook.pdf > > > > > -- James Hunt http://upstart.ubuntu.com/cookbook http://upstart.ubuntu.com/cookbook/upstart_cookbook.pdf -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Bug#684058: libjson0: json-c version 0.10 is now available.
Package: libjson0 Version: 0.9-1ubuntu1 Severity: normal Dear Maintainer, There is a new version of json-c (0.10) available from github: https://github.com/json-c/json-c/tarball/json-c-0.10-20120530 Amongst other things, this release includes internal 64-bit integer handling. Note that there is no tar file available (yet?) on the site below: http://oss.metaparadigm.com/json-c/ See my query to the json-c mailing list for further information (requires a google account to access): https://groups.google.com/forum/?fromgroups#!topic/json-c/QZPATk-20v0 Kind regards, James. -- System Information: Debian Release: wheezy/sid APT prefers precise-updates APT policy: (500, 'precise-updates'), (500, 'precise-security'), (500, 'precise') Architecture: i386 (i686) Kernel: Linux 3.2.0-27-generic-pae (SMP w/4 CPU cores) Locale: LANG=fr_CA.UTF8, LC_CTYPE=fr_CA.UTF8 (charmap=UTF-8) Shell: /bin/sh linked to /bin/dash Versions of packages libjson0 depends on: ii libc6 2.15-0ubuntu10 ii multiarch-support 2.15-0ubuntu10 libjson0 recommends no packages. libjson0 suggests no packages. -- no debconf information -- Kind regards, James. -- James Hunt http://upstart.ubuntu.com/cookbook http://upstart.ubuntu.com/cookbook/upstart_cookbook.pdf -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Bug#682032: init_is_upstart warnings if upstart installed but
Michael - I think you may have misread Didiers query; note the distinction between: # simply prints the version of *initctl*. /sbin/initctl --version # talks to upstart over D-Bus and prints the version of *Upstart* /sbin/initctl version Didier - yes, Upstart does indeed return the string "upstart" in the latter scenario. On an Ubuntu Precise system: $ /sbin/initctl version 2>/dev/null init (upstart 1.5) Kind regards, James. -- James Hunt http://upstart.ubuntu.com/cookbook http://upstart.ubuntu.com/cookbook/upstart_cookbook.pdf -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Bug#688046: ITP: out -- utility for producing UTF-8 output to standard streams and terminal.
On 18/09/12 17:50, Bastien ROUCARIES wrote:
> The name out is too generic. Care to rename? Something like utfout or
> outseq .
Presumably, you are referring to the package name here as opposed to the
name of the command?
Maybe this issue could be side-stepped by putting the binary into the
'moreutils' package which already has /usr/bin/isutf8?
I agree that 'utfout' would be a less ambiguous binary name so if
moreutils isn't appropriate, I'll rename binary+package to utfout.
Kind regards,
James.
>
> Bastien
>
> Le 18 sept. 2012 18:42, "James" <mailto:james.h...@canonical.com>> a écrit :
>
> Package: wnpp
> Severity: wishlist
> Owner: James Hunt mailto:james.h...@ubuntu.com>>
>
> * Package name: out
> Version : 0.0.1
> Upstream Author : James Hunt <mailto:james.h...@ubuntu.com>>
> * URL : https://code.launchpad.net/~jamesodhunt/+junk/out
> * License : GPL-3.0+
> Programming Lang: C
> Description : Utility for producing UTF-8 output to standard
> streams and terminal.
>
> Command-line tool that can produce UTF-8 (Unicode) strings in
> various ways and
> direct them to standard output, standard error or direct to the terminal
> without the need for shell support. Strings can be repeated, delayed,
> randomly-generated, written to arbitrary file descriptors,
> interspersed with
> other characters and generated using ranges. Printf(1)-style escape
> sequences are supported along with extended escape sequences.
>
> This utility sits somewhere between echo(1) and printf(1) in
> functionality with a dash of seq(1) thrown in.
>
> Many example are given in the manual page ('man/out.1') and in the
> initial blog post:
>
> http://ifdeflinux.blogspot.co.uk/2012/09/out-output-utility.html
>
>
> --
> To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org
> <mailto:debian-devel-requ...@lists.debian.org>
> with a subject of "unsubscribe". Trouble? Contact
> listmas...@lists.debian.org <mailto:listmas...@lists.debian.org>
> Archive:
>
> http://lists.debian.org/20120918100046.4971.93756.reportbug@localhost6.localdomain6
>
--
James Hunt
http://upstart.ubuntu.com/cookbook
http://upstart.ubuntu.com/cookbook/upstart_cookbook.pdf
--
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Bug#688046: ITP: out -- utility for producing UTF-8 output to standard streams and terminal.
On 18/09/12 19:04, Neil Williams wrote: > On Tue, 18 Sep 2012 11:00:46 +0100 > "James" wrote: > >> Package: wnpp >> Severity: wishlist >> Owner: James Hunt >> >> * Package name: out > > Terrible choice of name - utf8-out ? > >> Command-line tool that can produce UTF-8 (Unicode) strings in various ways >> and >> direct them to standard output, standard error or direct to the terminal >> without the need for shell support. Strings can be repeated, delayed, >> randomly-generated, written to arbitrary file descriptors, interspersed with >> other characters and generated using ranges. Printf(1)-style escape >> sequences are supported along with extended escape sequences. >> >> This utility sits somewhere between echo(1) and printf(1) in >> functionality with a dash of seq(1) thrown in. > > If that's true, why hasn't this been included into shells already? Well, zsh and bash do provide similar facilities but out/utf8out/foo/whatever would be able to provide a shell-agnostic facility to produce such output (like echo(1) and printf(1) which some shells have I appreciated added as built-ins). > > Struggling to see a use-case for it, TBH, which doesn't help with an > alternative name and may indicate that the long description doesn't > really describe why it would be useful. Is this just for test suites? > What needs this utility? > Its striving to be a generic facility to produce output filling the gap between echo and printf as alluded to. Admittedly, I did write this as a test tool, but have attempted to make it as general-purpose as possible. Personally, I can imagine quite a few use cases for this utility in shell scripts. See the examples in the man page for some ideas. Kind regards, James. -- James Hunt http://upstart.ubuntu.com/cookbook http://upstart.ubuntu.com/cookbook/upstart_cookbook.pdf -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Bug#688046: Info received (Bug#688046: ITP: out -- utility for producing UTF-8 output to standard streams and terminal.)
Please would you consider adding the 'out' utility to the moreutils package. If you agree that this might be an appropriate home for it, I'd be happy to rework the source repository (if required) to give the binary a less generic name (like maybe 'utfout'?). Kind regards, James. -- James Hunt http://upstart.ubuntu.com/cookbook http://upstart.ubuntu.com/cookbook/upstart_cookbook.pdf -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Bug#688046: Info received (Bug#688046: ITP: out -- utility for producing UTF-8 output to standard streams and terminal.)
On 19/09/12 18:07, Joey Hess wrote: > James Hunt wrote: >> Please would you consider adding the 'out' utility to the moreutils package. >> >> If you agree that this might be an appropriate home for it, I'd be happy >> to rework the source repository (if required) to give the binary a less >> generic name (like maybe 'utfout'?). > > This doesn't feel very appropriate for moreutils. Most of its behavior can, > as far as I can see, be obtained by various combinations of existing shell > tools. It doesn't do one pure function that was not previously available > at the shell prompt. > Hi Joey, Yes, that is a fair comment. In fact, that was its original purpose - to provide certain features without requiring a shell. However, thanks for considering. I'll continue to package it as a single binary. Kind regards, James. -- James Hunt http://upstart.ubuntu.com/cookbook http://upstart.ubuntu.com/cookbook/upstart_cookbook.pdf -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Bug#705917: pbuilder: Add basic DEP-8 autopkgtest
Package: pbuilder
Version: 0.213
Severity: normal
Tags: patch
User: ubuntu-de...@lists.ubuntu.com
Usertags: origin-ubuntu raring ubuntu-patch
Dear Maintainer,
The attached debdiff adds a basic DEP-8 autopkgtest for pbuilder.
See:
- https://code.launchpad.net/~jamesodhunt/ubuntu/raring/pbuilder/dep8-procenv
-
https://code.launchpad.net/~jamesodhunt/ubuntu/raring/pbuilder/dep8-procenv/+merge/159690
In Ubuntu, the attached patch was generated to achieve the following:
* ability to detect if pbuilder is able to:
- create a chroot for the current release.
- build a package.
* check that the resulting package is installable and the binary runnable.
Thanks for considering the patch.
-- System Information:
Debian Release: wheezy/sid
APT prefers raring-updates
APT policy: (500, 'raring-updates'), (500, 'raring-security'), (500, 'raring')
Architecture: i386 (i686)
Kernel: Linux 3.8.0-19-generic (SMP w/4 CPU cores)
Locale: LANG=fr_FR.UTF-8, LC_CTYPE=fr_FR.UTF-8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/dash
Kind regards,
James.
--
James Hunt
#upstart on freenode
http://upstart.ubuntu.com/cookbook
https://lists.ubuntu.com/mailman/listinfo/upstart-devel
=== modified file 'debian/changelog'
=== modified file 'debian/control'
--- debian/control 2012-11-05 23:27:25 +
+++ debian/control 2013-04-22 07:41:08 +
@@ -18,6 +18,7 @@
Standards-Version: 3.9.1
XS-Debian-Vcs-Git: git://git.debian.org/git/pbuilder/pbuilder.git
XS-Debian-Vcs-Browser: http://git.debian.org/?p=pbuilder/pbuilder.git
+XS-Testsuite: autopkgtest
Package: pbuilder
Architecture: all
=== added directory 'debian/tests'
=== added file 'debian/tests/build_procenv'
--- debian/tests/build_procenv 1970-01-01 00:00:00 +
+++ debian/tests/build_procenv 2013-04-18 18:02:15 +
@@ -0,0 +1,47 @@
+#!/bin/sh -e
+
+# The package we'll ask pbuilder to build.
+#
+# The advantage of this package being that it runs itself as part of its
+# build, which has the nice side-effect of exposing the full pbuilder
+# environment to those perusing the autopkgtest logs.
+pkg=procenv
+
+dir=/var/cache/pbuilder/result
+
+# Avoid conflict with ADT
+unset TMPDIR
+
+# Have to redirect stderr to avoid ADT thinking the test has failed
+# (the return code is still being checked, so this seems reasonable).
+echo "INFO: Creating pbuilder buildd chroot for current release"
+pbuilder --create --debug --debootstrapopts --variant=buildd 2>&1
+
+if [ ! -d "$dir" ]; then
+ echo "ERROR: cannot find directory $dir" >&2
+ exit 1
+fi
+
+# '--download-only' to avoid unpack which generates a signature
+# warning to stderr, causing this test to fail.
+echo "INFO: Downloading source for package '$pkg'"
+apt-get source --download-only "$pkg"
+
+dsc=$(ls ${pkg}*.dsc)
+
+echo "INFO: Building package '$pkg' from '$dsc'"
+pbuilder --build --debug "$dsc" 2>&1
+
+arch=$(dpkg --print-architecture)
+
+pkg_and_version=$(echo "$dsc"|sed 's/\.dsc$//g')
+deb=${dir}/${pkg_and_version}_${arch}.deb
+
+echo "INFO: Installing package '$pkg' from '$deb'"
+dpkg -i ${dir}/${pkg}*.deb
+
+# run the command to prove the build worked but also to expose the
+# auto-package-test environment used for this test.
+cmd=$pkg
+echo "INFO: Showing AutoPkgTest environment by running '$cmd' from package
'$pkg'"
+"$cmd"
=== added file 'debian/tests/control'
--- debian/tests/control1970-01-01 00:00:00 +
+++ debian/tests/control2013-04-18 18:02:15 +
@@ -0,0 +1,3 @@
+Tests: build_procenv
+Depends: @, debian-keyring
+Restrictions: needs-root
Bug#705926: sbuild: Add basic DEP-8 autopkgtest
Package: sbuild
Version: 0.63.2-1.1
Severity: normal
Tags: patch
User: ubuntu-de...@lists.ubuntu.com
Usertags: origin-ubuntu raring ubuntu-patch
Dear Maintainer,
The attached debdiff adds a basic DEP-8 autopkgtest for sbuild. See:
- https://code.launchpad.net/~jamesodhunt/ubuntu/raring/sbuild/dep8-procenv
-
https://code.launchpad.net/~jamesodhunt/ubuntu/raring/sbuild/dep8-procenv/+merge/159596
In Ubuntu, the attached patch was generated to achieve the following:
* ability to detect if sbuilder is able to:
- create a chroot for the current release.
- build a package.
* check that the resulting package is installable and the binary runnable.
Thanks for considering the patch.
Kind regards,
James.
--
James Hunt
#upstart on freenode
http://upstart.ubuntu.com/cookbook
https://lists.ubuntu.com/mailman/listinfo/upstart-devel
=== modified file 'debian/changelog'
=== modified file 'debian/control'
--- debian/control 2012-06-23 22:27:58 +
+++ debian/control 2013-04-22 10:00:27 +
@@ -8,6 +8,7 @@
Standards-Version: 3.9.1
Vcs-Browser: http://git.debian.org/?p=buildd-tools/sbuild.git
Vcs-Git: git://git.debian.org/git/buildd-tools/sbuild
+XS-Testsuite: autopkgtest
Package: libsbuild-perl
Architecture: all
=== added directory 'debian/tests'
=== added file 'debian/tests/build_procenv'
--- debian/tests/build_procenv 1970-01-01 00:00:00 +
+++ debian/tests/build_procenv 2013-04-22 10:00:05 +
@@ -0,0 +1,132 @@
+#!/bin/sh -e
+#-
+# DEP-8 test for sbuild.
+#
+# Creates a sbuild chroot, builds a package, installs the resulting
+# .deb, then runs the command provided by the .deb.
+#-
+
+die()
+{
+msg="$*"
+echo "ERROR: $msg" >&2
+exit 1
+}
+
+# The package we'll ask sbuild to build (we know its buildable since
+# it's already in the archive :-)
+#
+# The advantage of choosing this particular package being that it runs
+# *itself* at the end of its build, which has the nice side-effect of
+# exposing the full sbuild environment to those perusing the autopkgtest
+# logs.
+pkg=procenv
+
+release=$(lsb_release -c|cut -d: -f2|awk '{print $1}')
+dir="$ADTTMP/schroot-$release"
+arch=$(dpkg --print-architecture)
+chroot="${release}-${arch}-sbuild"
+
+distro=$(lsb_release --id|cut -d: -f2-|awk '{print $1}'|tr '[A-Z]' '[a-z]')
+
+if [ "$distro" = ubuntu ]
+then
+url=http://archive.ubuntu.com/ubuntu
+else
+url=http://archive.debian.org/debian
+fi
+
+# schroot does not allow a chroot name to be specified at creation time.
+# As such, we must take care to avoid stomping on a developers chroots.
+# If we find any that match the chroot we're about to try and make, exit
+# with a message.
+#
+# Note that we are very cautious in the grep check below; we purposely
+# don't match on the _type_ of schroot in case new schroot types are
+# added and this test is not updated to take account of the new types.
+
+schroots=$(schroot -l 2>/dev/null)
+
+if [ -n "$schroots" ] && echo "$schroots"|grep -q ":${release}-${arch}"
+then
+echo "INFO:"
+echo "INFO: Existing schroots detected for current release ($release)"
+echo "INFO: and architecture ($arch)"
+echo "INFO:"
+echo "INFO: Not continuing."
+echo "INFO:"
+
+# exit successfully
+exit 0
+fi
+
+echo "INFO: Creating sbuild chroot '$chroot' for release '$release' in
directory '$dir' from url '$url'"
+sbuild-createchroot "$release" "$dir" "$url"
+
+echo "INFO: Checking chroot '$chroot' is known"
+schroot --list --all-chroots|grep "^chroot:${chroot}$"
+
+echo "INFO: Displaying information about chroot '$chroot'"
+schroot --info "${chroot}"
+
+# '--download-only' to avoid unpack which generates a signature
+# warning to stderr, causing this test to fail.
+echo "INFO: Downloading source for package '$pkg'"
+apt-get source --download-only "$pkg"
+
+dsc=$(ls ${pkg}*.dsc)
+
+# crucial
+echo "INFO: Creating sbuild key pair"
+sbuild-update --keygen 2>&1
+
+echo "INFO: Building package '$pkg' for release '$release' from '$dsc'"
+sbuild -A -d "$release" $dsc 2>&1
+
+pkg_and_version=$(echo "$dsc"|sed 's/\.dsc$//g')
+deb=${pkg_and_version}_${arch}.deb
+
+echo "INFO: Displaying sbuild log"
+cat ${pkg_and_version}_${arch}*.build
+
+echo "INFO: Installing package '$pkg' from '$deb'"
+
Bug#705934: lxc: Add basic DEP-8 autopkgtest
Package: lxc
Version: 0.8.0~rc1-4
Severity: normal
Tags: patch
User: ubuntu-de...@lists.ubuntu.com
Usertags: origin-ubuntu raring ubuntu-patch
Dear Maintainer,
The attached debdiff adds a basic DEP-8 autopkgtest for lxc. See:
- https://code.launchpad.net/~jamesodhunt/ubuntu/raring/lxc/dep-8-tests
-
https://code.launchpad.net/~jamesodhunt/ubuntu/raring/lxc/dep-8-tests/+merge/157938
In Ubuntu, the attached patch was generated to achieve the following:
* Ensure a container can be created, started, stopped and cloned.
This is a very basic test, but I plan to add further checks soon.
Thanks for considering the patch.
-- System Information:
Debian Release: wheezy/sid
APT prefers raring-updates
APT policy: (500, 'raring-updates'), (500, 'raring-security'), (500, 'raring')
Architecture: i386 (i686)
Kernel: Linux 3.8.0-19-generic (SMP w/4 CPU cores)
Locale: LANG=C, LC_CTYPE=C (charmap=ANSI_X3.4-1968)
Shell: /bin/sh linked to /bin/dash
Kind regards,
James.
--
James Hunt
#upstart on freenode
http://upstart.ubuntu.com/cookbook
https://lists.ubuntu.com/mailman/listinfo/upstart-devel
=== modified file 'debian/changelog'
=== modified file 'debian/control'
--- debian/control 2013-02-18 22:12:13 +
+++ debian/control 2013-04-12 13:31:57 +
@@ -16,6 +16,7 @@
Standards-Version: 3.9.4
Homepage: http://lxc.sourceforge.net/
X-Python3-Version: >= 3.2
+XS-Testsuite: autopkgtest
Package: lxc
Architecture: linux-any
=== added directory 'debian/tests'
=== added file 'debian/tests/control'
--- debian/tests/control1970-01-01 00:00:00 +
+++ debian/tests/control2013-04-12 13:32:11 +
@@ -0,0 +1,3 @@
+Tests: exercise
+Depends: @, cgroup-lite
+Restrictions: needs-root
=== added file 'debian/tests/exercise'
--- debian/tests/exercise 1970-01-01 00:00:00 +
+++ debian/tests/exercise 2013-04-22 13:22:33 +
@@ -0,0 +1,80 @@
+#!/bin/sh
+#-
+# Some very basic tests to run in a DEP-8 environment.
+#-
+
+template_dir=/usr/share/lxc/templates
+
+# Exit with error message.
+#
+# @msg: message to display.
+die()
+{
+msg="$*"
+echo "ERROR: $msg" >&2
+exit 1
+}
+
+# seconds to wait for container to be running/stopped
+boot_secs=10
+shutdown_secs=10
+
+distro=$(lsb_release --id|cut -d: -f2-|awk '{print $1}'|tr '[A-Z]' '[a-z]')
+[ -z "$distro" ] && die "failed to determine distro"
+
+[ ! -d "$template_dir" ] && die "template directory does not exist"
+
+file=$(ls "${template_dir}/lxc-${distro}" 2>/dev/null)
+[ -z "$file" ] && die "template does not exist for distro '$distro'"
+template="$distro"
+
+release=$(lsb_release -c|awk '{print $2}')
+[ -z "$release" ] && die "failed to establish release"
+
+orig_name="${release}-dep8"
+new_name="${orig_name}-clone"
+
+name="$orig_name"
+
+# flush cache to ensure we always get the latest bootstrap image
+lxc-create -n "$name" -t "$template" -- "$release" --flush-cache || \
+die "failed to create container '$name' using template '$template' for
release '$release'"
+
+lxc-ls -1 | grep -q "^${name}$" || \
+die "container not known"
+
+lxc-start -n "$name" --daemon || die "failed to initiate container start"
+
+lxc-wait -n "$name" -s RUNNING -t $boot_secs || \
+die "container $name: did not start after $boot_secs seconds"
+
+lxc-stop -n "$name" || die "container $name: failed to initiate shutdown"
+
+lxc-wait -n "$name" -s STOPPED -t $shutdown_secs || \
+die "container $name: did not stop within $shutdown_secs seconds"
+
+lxc-clone -o "$orig_name" -n "$new_name" || \
+die "failed to clone container '$orig_name' to '$new_name'"
+
+# switch attention to the clone
+name="$new_name"
+
+lxc-start -n "$name" --daemon || die "container $name: failed to initiate
start"
+
+lxc-wait -n "$name" -s RUNNING -t $boot_secs || \
+die "container $name: did not start after $boot_secs seconds"
+
+lxc-stop -n "$name" || die "container $new_nam: failed to initiate shutdown"
+
+lxc-wait -n "$name" -s STOPPED -t $shutdown_secs || \
+die "container $name: did not stop within $shutdown_secs seconds"
+
+# clean up
+for name in "$orig_name" "$new_name"
+do
+lxc-destroy -n "$name" || die "container: $name: cannot delete"
+done
+
+echo SUCCESS
+
+exit 0
Bug#705926: Updated debdiff
I raised a similar bug for a pbuilder DEP-8 test:
http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=705917
Some concerns were raised on that bug so please find attached an updated sbuild
debdiff which take those concerns into account.
Kind regards,
James.
--
James Hunt
#upstart on freenode
http://upstart.ubuntu.com/cookbook
https://lists.ubuntu.com/mailman/listinfo/upstart-devel
diff -Nru sbuild-0.63.2/debian/changelog sbuild-0.63.2/debian/changelog
--- sbuild-0.63.2/debian/changelog 2013-01-17 14:00:53.0 +
+++ sbuild-0.63.2/debian/changelog 2013-04-23 10:06:31.0 +0100
@@ -1,3 +1,9 @@
+sbuild (0.63.2-1.1ubuntu2) UNRELEASED; urgency=low
+
+ * Added DEP-8 autopkgtest to build procenv.
+
+ -- James Hunt Tue, 23 Apr 2013 10:06:24 +0100
+
sbuild (0.63.2-1.1ubuntu1) raring; urgency=low
* Merge from Debian unstable. Remaining changes:
diff -Nru sbuild-0.63.2/debian/control sbuild-0.63.2/debian/control
--- sbuild-0.63.2/debian/control2012-11-28 12:42:59.0 +
+++ sbuild-0.63.2/debian/control2013-04-22 11:00:27.0 +0100
@@ -8,6 +8,7 @@
Standards-Version: 3.9.1
Vcs-Browser: http://git.debian.org/?p=buildd-tools/sbuild.git
Vcs-Git: git://git.debian.org/git/buildd-tools/sbuild
+XS-Testsuite: autopkgtest
Package: libsbuild-perl
Architecture: all
diff -Nru sbuild-0.63.2/debian/tests/build_procenv
sbuild-0.63.2/debian/tests/build_procenv
--- sbuild-0.63.2/debian/tests/build_procenv1970-01-01 01:00:00.0
+0100
+++ sbuild-0.63.2/debian/tests/build_procenv2013-04-23 10:00:42.0
+0100
@@ -0,0 +1,150 @@
+#!/bin/sh -e
+#-
+# DEP-8 test for sbuild.
+#
+# Creates a sbuild chroot, builds a package, installs the resulting
+# .deb, then runs the command provided by the .deb.
+#-
+
+die()
+{
+msg="$*"
+echo "ERROR: $msg" >&2
+exit 1
+}
+
+# The package we'll ask sbuild to build (we know its buildable since
+# it's already in the archive :-)
+#
+# The advantage of choosing this particular package being that it runs
+# *itself* at the end of its build, which has the nice side-effect of
+# exposing the full sbuild environment to those perusing the autopkgtest
+# logs.
+pkg=procenv
+
+# os-release is more portable than lsb_release
+distro=$(grep ^ID= /etc/os-release 2>/dev/null | cut -d= -f2- || :)
+[ -z "$distro" ] && die "cannot establish distribution"
+
+if [ "$distro" = ubuntu ]
+then
+# Build chroot for latest release.
+release=$(distro-info --devel)
+
+url=http://archive.ubuntu.com/ubuntu
+elif [ "$distro" = debian ]
+then
+# Build chroot for latest stable release since
+# sid may not be buildable on a particular day.
+release=$(distro-info --stable)
+
+url=http://archive.debian.org/debian
+else
+die "need to know where archive is for distro '$distro'"
+fi
+
+arch=$(dpkg --print-architecture 2>/dev/null)
+[ -z "$arch" ] && die "cannot establish architecture"
+
+[ -z "$ADTTMP" ] && die "ADTTMP not set"
+dir="$ADTTMP/schroot-$release"
+
+# The expected name of the schroot that sbuild-createchroot will create
+chroot="${release}-${arch}-sbuild"
+
+# schroot does not allow a chroot name to be specified at creation time.
+# As such, we must take care to avoid stomping on a developers chroots.
+# If we find any that match the chroot we're about to try and make, exit
+# with a message.
+#
+# Note that we are very cautious in the grep check below; we purposely
+# don't match on the _type_ of schroot in case new schroot types are
+# added and this test is not updated to take account of the new types.
+
+schroots=$(schroot -l 2>/dev/null)
+
+if [ -n "$schroots" ] && echo "$schroots"|grep -q ":${release}-${arch}"
+then
+echo "INFO:"
+echo "INFO: Existing schroots detected for current release ($release)"
+echo "INFO: and architecture ($arch)"
+echo "INFO:"
+echo "INFO: Not continuing."
+echo "INFO:"
+
+# exit successfully
+exit 0
+fi
+
+# Use '--download-only' to avoid unpack which generates a
+# signature warning to stderr, causing this test to fail.
+# Take care to download the package version for the release we will
+# create the chroot for.
+echo "INFO: Downloading source for package '$pkg'"
+apt-get source --download-only "$pkg/$release"
+
+dsc=$(ls ${pkg}*.dsc)
+
+echo "INFO: Creating sbuild chroot '$chroot' for release '$release' in
directory '$dir' from url '$url'"
+sbuild-createchroot "$releas
Bug#705917: pbuilder: Add basic DEP-8 autopkgtest
Note that this bug is analogous to the sbuild bug I raised here: http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=705926 Kind regards, James. -- James Hunt #upstart on freenode http://upstart.ubuntu.com/cookbook https://lists.ubuntu.com/mailman/listinfo/upstart-devel -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Bug#705926: sbuild: Add basic DEP-8 autopkgtest
An updated debdiff with the following changes:
* debian/tests/build_procenv:
- Revert to using lsb-release: os-release is available by default, but
does not provide release name in a portable format.
- Only install the built .deb if the release built for matches the
running release.
- Perform checks on .deb since we might not be able to install it.
* debian/tests/control:
- Removed breaks-testbed as this stops the test even running on
Ubuntu.
- Added lsb-release dependency.
Kind regards,
James.
--
James Hunt
#upstart on freenode
http://upstart.ubuntu.com/cookbook
https://lists.ubuntu.com/mailman/listinfo/upstart-devel
diff -Nru sbuild-0.63.2/debian/changelog sbuild-0.63.2/debian/changelog
--- sbuild-0.63.2/debian/changelog 2013-01-17 14:00:53.0 +
+++ sbuild-0.63.2/debian/changelog 2013-04-23 22:30:44.0 +0100
@@ -1,3 +1,9 @@
+sbuild (0.63.2-1.1ubuntu2) UNRELEASED; urgency=low
+
+ * Added DEP-8 autopkgtest to build procenv.
+
+ -- James Hunt Tue, 23 Apr 2013 22:30:35 +0100
+
sbuild (0.63.2-1.1ubuntu1) raring; urgency=low
* Merge from Debian unstable. Remaining changes:
diff -Nru sbuild-0.63.2/debian/control sbuild-0.63.2/debian/control
--- sbuild-0.63.2/debian/control2012-11-28 12:42:59.0 +
+++ sbuild-0.63.2/debian/control2013-04-22 11:00:27.0 +0100
@@ -8,6 +8,7 @@
Standards-Version: 3.9.1
Vcs-Browser: http://git.debian.org/?p=buildd-tools/sbuild.git
Vcs-Git: git://git.debian.org/git/buildd-tools/sbuild
+XS-Testsuite: autopkgtest
Package: libsbuild-perl
Architecture: all
diff -Nru sbuild-0.63.2/debian/tests/build_procenv
sbuild-0.63.2/debian/tests/build_procenv
--- sbuild-0.63.2/debian/tests/build_procenv1970-01-01 01:00:00.0
+0100
+++ sbuild-0.63.2/debian/tests/build_procenv2013-04-23 22:01:46.0
+0100
@@ -0,0 +1,170 @@
+#!/bin/sh -e
+#-
+# DEP-8 test for sbuild.
+#
+# Creates a sbuild chroot, builds a package, installs the resulting
+# .deb, then runs the command provided by the .deb.
+#-
+
+die()
+{
+msg="$*"
+echo "ERROR: $msg" >&2
+exit 1
+}
+
+# The package we'll ask sbuild to build (we know its buildable since
+# it's already in the archive :-)
+#
+# The advantage of choosing this particular package being that it runs
+# *itself* at the end of its build, which has the nice side-effect of
+# exposing the full sbuild environment to those perusing the autopkgtest
+# logs.
+pkg=procenv
+
+distro=$(lsb_release --id|cut -d: -f2-|awk '{print $1}'|tr '[A-Z]' '[a-z]')
+[ -z "$distro" ] && die "cannot establish distribution"
+
+host_release=$(lsb_release --codename|cut -d: -f2-|awk '{print $1}')
+[ -z "$host_release" ] && die "cannot establish release running on host"
+
+if [ "$distro" = ubuntu ]
+then
+# Build chroot for latest release.
+release=$(distro-info --devel)
+
+url=http://archive.ubuntu.com/ubuntu
+elif [ "$distro" = debian ]
+then
+# Build chroot for latest stable release since
+# sid may not be buildable on a particular day.
+release=$(distro-info --stable)
+
+url=http://archive.debian.org/debian
+else
+die "need to know where archive is for distro '$distro'"
+fi
+
+arch=$(dpkg --print-architecture 2>/dev/null)
+[ -z "$arch" ] && die "cannot establish architecture"
+
+[ -z "$ADTTMP" ] && die "ADTTMP not set"
+dir="$ADTTMP/schroot-$release"
+
+# The expected name of the schroot that sbuild-createchroot will create
+chroot="${release}-${arch}-sbuild"
+
+# schroot does not allow a chroot name to be specified at creation time.
+# As such, we must take care to avoid stomping on a developers chroots.
+# If we find any that match the chroot we're about to try and make, exit
+# with a message.
+#
+# Note that we are very cautious in the grep check below; we purposely
+# don't match on the _type_ of schroot in case new schroot types are
+# added and this test is not updated to take account of the new types.
+
+schroots=$(schroot -l 2>/dev/null)
+
+if [ -n "$schroots" ] && echo "$schroots"|grep -q ":${release}-${arch}"
+then
+echo "INFO:"
+echo "INFO: Existing schroots detected for current release ($release)"
+echo "INFO: and architecture ($arch)"
+echo "INFO:"
+echo "INFO: Not continuing."
+echo "INFO:"
+
+# exit successfully
+exit 0
+fi
+
+# Use '--download-only' to avoid unpack which generates a
+# signature warning to stderr, causing this te
Bug#705917: pbuilder: Add basic DEP-8 autopkgtest
Updated debdiff attached that makes the following changes:
* debian/tests/build_procenv: Aligned with sbuild DEP-8 script:
- Use appropriate archive based on distribution.
- Build package for specified distribution (stable for debian to
ensure chroot is creatable).
- Download package source for specified release.
- Only install build .deb if host environment matches the release the
package was built for.
- Perform basic checks on built .dep in case it is not possible to
install it.
* debian/tests/control: Add distro-info and lsb-release dependencies.
Adding breaks-testbed stops this test running in the Ubuntu test environment so
I'd like to understand further how this option is used (or is intended to be
used) in Debian.
Kind regards,
James.
--
James Hunt
#upstart on freenode
http://upstart.ubuntu.com/cookbook
https://lists.ubuntu.com/mailman/listinfo/upstart-devel
diff -Nru pbuilder-0.213ubuntu1/debian/changelog
pbuilder-0.213ubuntu2/debian/changelog
--- pbuilder-0.213ubuntu1/debian/changelog 2012-11-07 16:11:40.0
+
+++ pbuilder-0.213ubuntu2/debian/changelog 2013-04-24 09:34:41.0
+0100
@@ -1,3 +1,9 @@
+pbuilder (0.213ubuntu2) UNRELEASED; urgency=low
+
+ * Added basic DEP-8 autopkgtest.
+
+ -- James Hunt Wed, 24 Apr 2013 09:34:33 +0100
+
pbuilder (0.213ubuntu1) raring; urgency=low
* Merge from Debian unstable (LP: #1075347, LP: #1060101)
diff -Nru pbuilder-0.213ubuntu1/debian/control
pbuilder-0.213ubuntu2/debian/control
--- pbuilder-0.213ubuntu1/debian/control2012-11-07 16:11:40.0
+
+++ pbuilder-0.213ubuntu2/debian/control2013-04-22 08:41:08.0
+0100
@@ -18,6 +18,7 @@
Standards-Version: 3.9.1
XS-Debian-Vcs-Git: git://git.debian.org/git/pbuilder/pbuilder.git
XS-Debian-Vcs-Browser: http://git.debian.org/?p=pbuilder/pbuilder.git
+XS-Testsuite: autopkgtest
Package: pbuilder
Architecture: all
diff -Nru pbuilder-0.213ubuntu1/debian/tests/build_procenv
pbuilder-0.213ubuntu2/debian/tests/build_procenv
--- pbuilder-0.213ubuntu1/debian/tests/build_procenv1970-01-01
01:00:00.0 +0100
+++ pbuilder-0.213ubuntu2/debian/tests/build_procenv2013-04-24
08:51:33.0 +0100
@@ -0,0 +1,113 @@
+#!/bin/sh -e
+
+#-
+# DEP-8 test for sbuild.
+#
+# Creates a sbuild chroot, builds a package, installs the resulting
+# .deb, then runs the command provided by the .deb.
+#-
+
+die()
+{
+msg="$*"
+echo "ERROR: $msg" >&2
+exit 1
+}
+
+# The package we'll ask sbuild to build (we know its buildable since
+# it's already in the archive :-)
+#
+# The advantage of choosing this particular package being that it runs
+# *itself* at the end of its build, which has the nice side-effect of
+# exposing the full sbuild environment to those perusing the autopkgtest
+# logs.
+pkg=procenv
+
+dir=/var/cache/pbuilder/result
+
+# Avoid conflict with ADT
+unset TMPDIR
+
+distro=$(lsb_release --id|cut -d: -f2-|awk '{print $1}'|tr '[A-Z]' '[a-z]' ||
:)
+[ -z "$distro" ] && die "cannot establish distribution"
+
+host_release=$(lsb_release --codename|cut -d: -f2-|awk '{print $1}' || :)
+[ -z "$host_release" ] && die "cannot establish release running on host"
+
+if [ "$distro" = ubuntu ]
+then
+# Build chroot for latest release.
+release=$(distro-info --devel)
+
+url=http://archive.ubuntu.com/ubuntu
+elif [ "$distro" = debian ]
+then
+# Build chroot for latest stable release since
+# sid may not be buildable on a particular day.
+release=$(distro-info --stable)
+
+url=http://archive.debian.org/debian
+else
+die "need to know where archive is for distro '$distro'"
+fi
+
+# Have to redirect stderr to avoid ADT thinking the test has failed
+# (the return code is still being checked, so this seems reasonable).
+echo "INFO: Creating pbuilder buildd chroot for release '$release'"
+pbuilder --create --debug \
+--distribution "$release" \
+--mirror "$url" \
+--debootstrapopts --variant=buildd 2>&1
+
+if [ ! -d "$dir" ]
+then
+echo "ERROR: cannot find directory $dir" >&2
+exit 1
+fi
+
+# Use '--download-only' to avoid unpack which generates a
+# signature warning to stderr, causing this test to fail.
+# Take care to download the package version for the release we will
+# create the chroot for.
+echo "INFO: Downloading source for package '$pkg' release '$release'"
+apt-get source --download-only "$pkg/$release"
+
+dsc=$(ls ${pkg}*.dsc)
+
+echo "INFO: Building package '$pkg' for release
Bug#710955: automake: how to force a verbose build with automake 1.13?
Package: automake Version: 1:1.11.6-1ubuntu1 Severity: normal Dear Maintainer, I maintain the procenv application which is slightly unusual in that as part of its package build (make check as handled by dh_auto_test), it runs itself to allow those viewing the build logs to see the sort of environment that, for example, a buildd provides. This has proved to be extremely useful in the past debugging issues with _other_ packages build failures caused by erroneous assumptions made about the build environment. Indeed, I've added links to various sets of build logs on the tools project page as and aid to others: https://launchpad.net/procenv/ All used to work perfectly until (I think) the recent change to automake 1.13. Now, although the tests run, the build log just shows the test summary, like this: Testsuite summary for procenv 0.23 # TOTAL: 2 # PASS: 2 # SKIP: 0 # XFAIL: 0 # FAIL: 0 # XPASS: 0 # ERROR: 0 I've tried various hacks to force the old verbose/non-silent "make check" build behaviour including: 1) Explicitly specifying `AM_SILENT_RULES([yes])' in configure.ac and then forcing `--disable-silent-rules' in debian/rules via override_dh_auto_configure. 2) Forcing the setting of V=1 and VERBOSE=1 in various places. Is there a way to re-enable a verbose build that displays more than the test summary such as the following? I suspect (2) is the more official way to control automakes verbosity, but I haven't yet found the magic incantation that works in a Debian build environment. -- System Information: Debian Release: wheezy/sid APT prefers raring-updates APT policy: (500, 'raring-updates'), (500, 'raring-security'), (500, 'raring') Architecture: i386 (i686) Kernel: Linux 3.8.0-23-generic (SMP w/4 CPU cores) Locale: LANG=fr_FR.UTF-8, LC_CTYPE=fr_FR.UTF-8 (charmap=UTF-8) Shell: /bin/sh linked to /bin/dash Versions of packages automake depends on: ii autoconf 2.69-1ubuntu1 ii autotools-dev 20120608.1 ii dpkg 1.16.10ubuntu1 ii install-info 4.13a.dfsg.1-10ubuntu4 automake recommends no packages. automake suggests no packages. -- no debconf information Kind regards, James. -- James Hunt #upstart on freenode http://upstart.ubuntu.com/cookbook https://lists.ubuntu.com/mailman/listinfo/upstart-devel -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Bug#709357: abi-compliance-checker: DEP-8 test
Package: abi-compliance-checker Version: 1.97.7-1 Severity: normal Dear Maintainer, Please consider adding the patch below to provide a DEP-8 test for abi-compliance-checker. The need for one was seen today when the self-test failed on Ubuntu Saucy (separate bug to follow). -- System Information: Debian Release: wheezy/sid APT prefers raring-updates APT policy: (500, 'raring-updates'), (500, 'raring-security'), (500, 'raring') Architecture: i386 (i686) Kernel: Linux 3.8.0-21-generic (SMP w/4 CPU cores) Locale: LANG=fr_FR.UTF-8, LC_CTYPE=fr_FR.UTF-8 (charmap=UTF-8) Shell: /bin/sh linked to /bin/dash Versions of packages abi-compliance-checker depends on: ii binutils 2.23.2-2ubuntu1 ii gcc 4:4.7.3-1ubuntu10 ii perl 5.14.2-21 abi-compliance-checker recommends no packages. Versions of packages abi-compliance-checker suggests: ii doc-base 0.10.4 -- no debconf information Kind regards, James. -- James Hunt #upstart on freenode http://upstart.ubuntu.com/cookbook https://lists.ubuntu.com/mailman/listinfo/upstart-devel === modified file 'debian/changelog' --- debian/changelog 2012-05-31 14:21:07 + +++ debian/changelog 2013-05-22 16:33:04 + @@ -1,3 +1,9 @@ +abi-compliance-checker (1.97.7-2) unstable; urgency=low + + * Added DEP-8 self-test. + + -- James Hunt Wed, 22 May 2013 17:32:23 +0100 + abi-compliance-checker (1.97.7-1) unstable; urgency=low * New upstream === modified file 'debian/control' --- debian/control 2012-05-31 14:21:07 + +++ debian/control 2013-05-22 15:01:44 + @@ -9,6 +9,7 @@ Homepage: http://ispras.linux-foundation.org/index.php/ABI_compliance_checker Vcs-Svn: svn://svn.debian.org/collab-maint/deb-maint/abi-compliance-checker/trunk Vcs-Browser: http://anonscm.debian.org/viewvc/collab-maint/deb-maint/abi-compliance-checker/trunk/ +XS-Testsuite: autopkgtest Package: abi-compliance-checker Architecture: all === added directory 'debian/tests' === added file 'debian/tests/control' --- debian/tests/control 1970-01-01 00:00:00 + +++ debian/tests/control 2013-05-22 15:02:49 + @@ -0,0 +1,2 @@ +Tests: self_test +Depends: @ === added file 'debian/tests/self_test' --- debian/tests/self_test 1970-01-01 00:00:00 + +++ debian/tests/self_test 2013-05-22 15:05:37 + @@ -0,0 +1,3 @@ +#!/bin/sh -e + +abi-compliance-checker -test
Bug#709357: abi-compliance-checker: DEP-8 test
Updated diff attached where abi-compliance-checker is run with additional option '-debug' to allow analysis should test fail. Kind regards, James. -- James Hunt #upstart on freenode http://upstart.ubuntu.com/cookbook https://lists.ubuntu.com/mailman/listinfo/upstart-devel === modified file 'debian/changelog' --- debian/changelog 2012-05-31 14:21:07 + +++ debian/changelog 2013-05-23 07:40:10 + @@ -1,3 +1,9 @@ +abi-compliance-checker (1.97.7-2) unstable; urgency=low + + * Added DEP-8 self-test. + + -- James Hunt Thu, 23 May 2013 08:39:57 +0100 + abi-compliance-checker (1.97.7-1) unstable; urgency=low * New upstream === modified file 'debian/control' --- debian/control 2012-05-31 14:21:07 + +++ debian/control 2013-05-22 15:01:44 + @@ -9,6 +9,7 @@ Homepage: http://ispras.linux-foundation.org/index.php/ABI_compliance_checker Vcs-Svn: svn://svn.debian.org/collab-maint/deb-maint/abi-compliance-checker/trunk Vcs-Browser: http://anonscm.debian.org/viewvc/collab-maint/deb-maint/abi-compliance-checker/trunk/ +XS-Testsuite: autopkgtest Package: abi-compliance-checker Architecture: all === added directory 'debian/tests' === added file 'debian/tests/control' --- debian/tests/control 1970-01-01 00:00:00 + +++ debian/tests/control 2013-05-22 15:02:49 + @@ -0,0 +1,2 @@ +Tests: self_test +Depends: @ === added file 'debian/tests/self_test' --- debian/tests/self_test 1970-01-01 00:00:00 + +++ debian/tests/self_test 2013-05-23 07:39:27 + @@ -0,0 +1,3 @@ +#!/bin/sh -e + +abi-compliance-checker -test -debug
Bug#717806: abi-compliance-checker: self-check fails with error "can't compile header(s)"
Package: abi-compliance-checker Version: 1.99.1-1 Severity: important Dear Maintainer, The self-test mode in abi-compliance-checker now fails: __ $ abi-compliance-checker -test -debug Using GCC 4.8 (i486-linux-gnu) verifying detectable C library changes running perl /usr/bin/abi-compliance-checker -l libsample_c -d1 libsample_c/v1.xml -d2 libsample_c/v2.xml -debug preparation, please wait ... Using GCC 4.8 (i486-linux-gnu) checking header(s) 1.0 ... Detected code: "int testCppKeywords1(int class," Using C++ compatibility mode ERROR: can't compile header(s) ERROR: analysis has failed __ Error is also seen in Ubuntu Saucy. -- System Information: Debian Release: jessie/sid APT prefers unstable APT policy: (500, 'unstable') Architecture: i386 (i686) Kernel: Linux 3.10.0-5-generic (SMP w/4 CPU cores) Locale: LANG=C, LC_CTYPE=C (charmap=ANSI_X3.4-1968) Shell: /bin/sh linked to /bin/dash Versions of packages abi-compliance-checker depends on: ii binutils 2.23.52.20130722-1 ii gcc 4:4.8.1-2 ii perl 5.14.2-21 Versions of packages abi-compliance-checker recommends: ii exuberant-ctags [ctags] 1:5.9~svn20110310-6 Versions of packages abi-compliance-checker suggests: pn doc-base pn icheck -- no debconf information Kind regards, James. -- James Hunt #upstart on freenode http://upstart.ubuntu.com/cookbook https://lists.ubuntu.com/mailman/listinfo/upstart-devel -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Bug#710955: automake: how to force a verbose build with automake 1.13?
I raised this issue with automake upstream: http://lists.gnu.org/archive/html/automake/2013-06/msg00048.html ... however, the "fix" suggested in the thread doesn't appear to work. Kind regards, James. -- James Hunt #upstart on freenode http://upstart.ubuntu.com/cookbook https://lists.ubuntu.com/mailman/listinfo/upstart-devel -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Bug#710955: automake: how to force a verbose build with automake 1.13?
Stefano Lattarini [1] solved the problem: > Anyway, my suggestion is: just start requiring Automake >= 1.12 (or even > >= 1.13), and use the 'serial-tests' option unconditionally. > > Or better again: add the parallel-tests option (supported since automake > 1.11, and the default since automake 1.13) and convert your testsuite to > be able to run in parellel mode (that is usually a good idea regardless). > Then add a custom target (to be run after "make check") that displays the > content of all the testsuite log files on stdout; e.g.: > > display-testsuite-logs: > for f in $(TEST_LOGS); do printf "\n=== $$f ===\n\n"; cat $$f; done For procenv, I did the following: configure.ac: # automake-1.13 defaults to running tests in parallel. As a consequence, # it also disables verbose output meaning that procenv output is not # visible in build logs. Therefore, force old behaviour. AM_INIT_AUTOMAKE([serial-tests]) Aside - I had though that the following should also work, but it didn't seem to for me: Makefile.am: # Force old behaviour to retain traditional test semantics. AUTOMAKE_OPTIONS = serial-tests -- Kind regards, James. [1] - https://lists.gnu.org/archive/html/automake/2013-06/msg00051.html -- James Hunt #upstart on freenode http://upstart.ubuntu.com/cookbook https://lists.ubuntu.com/mailman/listinfo/upstart-devel -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Bug#664787: debconf-doc: debconf-devel(7) does not document entire protocol
Package: debconf-doc Version: 1.5.42ubuntu1 Severity: normal Dear Maintainer, Missing commands in debconf-devel(7) are: DATA, PREVIOUS_MODULE, TEXT, INFO and PROGRESS. Observed by comparing man page to commands listed in `/usr/share/debconf/confmodule'. Kind regards, James -- System Information: Debian Release: wheezy/sid APT prefers precise-updates APT policy: (500, 'precise-updates'), (500, 'precise-security'), (500, 'precise') Architecture: i386 (i686) Kernel: Linux 3.2.0-19-generic-pae (SMP w/4 CPU cores) Locale: LANG=C, LC_CTYPE=C (charmap=ANSI_X3.4-1968) Shell: /bin/sh linked to /bin/dash debconf-doc depends on no packages. debconf-doc recommends no packages. Versions of packages debconf-doc suggests: pn debian-policy -- no debconf information -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Bug#671009: debian-policy: does not mention debtags
Package: debian-policy Severity: normal Dear Maintainer, I recently came across a package (casper) which included a "Tag:" field in 'apt-cache show' output. Obtaining the source showed this field in the debian/control file, but there is no mention of "Tag:" field in the Debian Policy. After fruitless Googling, I asked on IRC and was pointed to: http://wiki.debian.org/Debtags ... which provided the information I was looking for. However, since "Tag:" is an optional binary package control file field (like Depends, Essential, etc) and since it is used by packages in stable Debian releases, surely this should be documented directly in the Debian Policy manual? (or at the very least mentioned in passing to avoid confusion)? Kind regards, James Hunt -- System Information: Debian Release: wheezy/sid APT prefers precise-updates APT policy: (500, 'precise-updates'), (500, 'precise-security'), (500, 'precise') Architecture: i386 (i686) Kernel: Linux 3.2.0-24-generic-pae (SMP w/4 CPU cores) Locale: LANG=fr_CA.UTF8, LC_CTYPE=fr_CA.UTF8 (charmap=UTF-8) Shell: /bin/sh linked to /bin/dash -- James Hunt http://upstart.ubuntu.com/cookbook http://upstart.ubuntu.com/cookbook/upstart_cookbook.pdf -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Bug#764461: debhelper: [PATCH] Add '--show-progress' option.
Package: debhelper
Version: 9.20140613ubuntu2
Severity: normal
Dear Maintainer,
dh currently offers a '--no-act' option that will show the commands that
would be run. It also displays the name of the command that failed in
the event of error.
This patch adds a '--show-progress' option (along with
"DH_SHOW_PROGRESS" environment variable) that:
1) Displays the list (or plan) of commands that are going to be run (like
'--no-act').
2) Adds two output tags for each dh_* command called, one is emitted
before the command is run and the other is emitted after the command has
completed successfully.
For example, if 'dh build --show-progress' is called for a program
whose tests fail, a 'grep ^dh build.log' will show:
dh plan: dh_testdir
dh plan: dh_auto_configure
dh plan: dh_auto_build
dh plan: dh_auto_test
dh -> dh_testdir
dh <- dh_testdir
dh -> dh_auto_configure
dh <- dh_auto_configure
dh -> dh_auto_build
dh <- dh_auto_build
dh -> dh_auto_test
dh_auto_test: make -j1 check returned exit code 2
This makes it clear which commands dh was planning to run ('dh plan:')
and which command failed (du_auto_test).
The '->' ("about to call") and '<-' ("called") output is not strictly
necessary to determine the failing step but are useful to watch build
progress. They could also be used by other tools to create expandable
"twisties" for example for a web version of a build log.
This idea was inspired by perusing the excellent FTBFS page at [1].
Currently, we can see easily which packages FTBFS, but the build
comprises multiple steps and having the ability to display and summarise
the precise stage of the build that is failing would be highly
desirable.
[1] -
http://people.ubuntuwire.org/~wgrant/rebuild-ftbfs-test/test-rebuild-20140914-utopic.html
-- System Information:
Debian Release: jessie/sid
APT prefers utopic-updates
APT policy: (500, 'utopic-updates'), (500, 'utopic-security'), (500,
'utopic'), (100, 'utopic-backports')
Architecture: amd64 (x86_64)
Foreign Architectures: i386
armhf
Kernel: Linux 3.16.0-20-generic (SMP w/4 CPU cores)
Locale: LANG=fr_FR.UTF-8, LC_CTYPE=fr_FR.UTF-8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/dash
Versions of packages debhelper depends on:
ii binutils 2.24.51.20141001-1ubuntu2
ii dh-apparmor 2.8.96~2652-0ubuntu7
ii dpkg 1.17.13ubuntu1
ii dpkg-dev 1.17.13ubuntu1
ii file 1:5.19-1ubuntu1.1
ii man-db 2.7.0.2-1
ii perl 5.20.0-6
ii po-debconf 1.0.16+nmu3
debhelper recommends no packages.
Versions of packages debhelper suggests:
ii dh-make 1.20140617
-- no debconf information
--
Kind regards,
James
--
James Hunt
From c25e8750c4b1fc9335376063efcbf0b3dc6797a8 Mon Sep 17 00:00:00 2001
From: James Hunt
Date: Wed, 8 Oct 2014 11:30:42 +0100
Subject: [PATCH] Add '--show-progress' option.
Signed-off-by: James Hunt
---
Debian/Debhelper/Dh_Getopt.pm | 2 ++
Debian/Debhelper/Dh_Lib.pm| 4 +++
debhelper.pod | 4 +++
dh| 63 +++
4 files changed, 68 insertions(+), 5 deletions(-)
diff --git a/Debian/Debhelper/Dh_Getopt.pm b/Debian/Debhelper/Dh_Getopt.pm
index e4f3e47..c236ce9 100644
--- a/Debian/Debhelper/Dh_Getopt.pm
+++ b/Debian/Debhelper/Dh_Getopt.pm
@@ -80,6 +80,8 @@ sub getoptions {
"verbose" => \$dh{VERBOSE},
"no-act" => \$dh{NO_ACT},
+
+ "show-progress" => \$dh{SHOW_PROGRESS},
"i" => \&AddPackage,
"indep" => \&AddPackage,
diff --git a/Debian/Debhelper/Dh_Lib.pm b/Debian/Debhelper/Dh_Lib.pm
index 6a79c9c..8b9607e 100644
--- a/Debian/Debhelper/Dh_Lib.pm
+++ b/Debian/Debhelper/Dh_Lib.pm
@@ -70,6 +70,10 @@ sub init {
$dh{NO_ACT}=1;
}
+ if (defined $ENV{SHOW_PROGRESS} && $ENV{SHOW_PROGRESS} ne "") {
+ $dh{SHOW_PROGRESS}=1;
+ }
+
# Get the name of the main binary package (first one listed in
# debian/control). Only if the main package was not set on the
# command line.
diff --git a/debhelper.pod b/debhelper.pod
index 659c4a3..e23de14 100644
--- a/debhelper.pod
+++ b/debhelper.pod
@@ -663,6 +663,10 @@ overriding any value in F.
Set to B<1> to enable no-act mode.
+=item B
+
+Set to B<1> to enable show-progress mode.
+
=item B
Anything in this variable will be prepended to the command line arguments
diff --git a/dh b/dh
index f3bd321..f11f64e 100755
--- a/dh
+++ b/dh
@@ -68,6 +68,15 @@ Prints commands that would run for a given sequence, but does not run them.
Note that dh normally skips running commands that it knows will do nothing.
With --no-act, the full list of commands in a sequence is printed.
+=item B<--show-progress>
+
+Displays commands that are about to be run ("
Bug#764461: debhelper: [PATCH] Add '--show-progress' option.
Thanks Jakub.
Updated patch attached.
Kind regards,
James
--
James Hunt
From 073d5b01e5f62fcdf998603d4fc49cc67d823db8 Mon Sep 17 00:00:00 2001
From: James Hunt
Date: Wed, 8 Oct 2014 11:30:42 +0100
Subject: Add '--show-progress' option.
Signed-off-by: James Hunt
---
Debian/Debhelper/Dh_Getopt.pm | 2 ++
Debian/Debhelper/Dh_Lib.pm| 4 +++
debhelper.pod | 4 +++
dh| 63 +++
4 files changed, 68 insertions(+), 5 deletions(-)
diff --git a/Debian/Debhelper/Dh_Getopt.pm b/Debian/Debhelper/Dh_Getopt.pm
index e4f3e47..c236ce9 100644
--- a/Debian/Debhelper/Dh_Getopt.pm
+++ b/Debian/Debhelper/Dh_Getopt.pm
@@ -80,6 +80,8 @@ sub getoptions {
"verbose" => \$dh{VERBOSE},
"no-act" => \$dh{NO_ACT},
+
+ "show-progress" => \$dh{SHOW_PROGRESS},
"i" => \&AddPackage,
"indep" => \&AddPackage,
diff --git a/Debian/Debhelper/Dh_Lib.pm b/Debian/Debhelper/Dh_Lib.pm
index 6a79c9c..24c9b89 100644
--- a/Debian/Debhelper/Dh_Lib.pm
+++ b/Debian/Debhelper/Dh_Lib.pm
@@ -70,6 +70,10 @@ sub init {
$dh{NO_ACT}=1;
}
+ if (defined $ENV{DH_SHOW_PROGRESS} && $ENV{DH_SHOW_PROGRESS} ne "") {
+ $dh{SHOW_PROGRESS}=1;
+ }
+
# Get the name of the main binary package (first one listed in
# debian/control). Only if the main package was not set on the
# command line.
diff --git a/debhelper.pod b/debhelper.pod
index 659c4a3..e23de14 100644
--- a/debhelper.pod
+++ b/debhelper.pod
@@ -663,6 +663,10 @@ overriding any value in F.
Set to B<1> to enable no-act mode.
+=item B
+
+Set to B<1> to enable show-progress mode.
+
=item B
Anything in this variable will be prepended to the command line arguments
diff --git a/dh b/dh
index f3bd321..f11f64e 100755
--- a/dh
+++ b/dh
@@ -68,6 +68,15 @@ Prints commands that would run for a given sequence, but does not run them.
Note that dh normally skips running commands that it knows will do nothing.
With --no-act, the full list of commands in a sequence is printed.
+=item B<--show-progress>
+
+Displays commands that are about to be run ("the plan", same output as B<--no-act>)
+and then proceeds to run the commands as usual, printing a start and end tag for
+every command run.
+
+This can be useful for scripts that parse the output to determine more easily which
+step failed, and how far through the build process the failing step was.
+
=back
Other options passed to B are passed on to each command it runs. This
@@ -440,6 +449,9 @@ else {
# Additional command options
my %command_opts;
+# array of commands to run
+my @commands;
+
# sequence addon interface
sub _insert {
my $offset=shift;
@@ -524,6 +536,26 @@ sub list_addons {
exit 0;
}
+# If show_plan is set, show commands that would be run (NO_ACT),
+# else run the commands.
+sub run_commands {
+ my $commands=shift;
+ my $show_plan=shift;
+
+ foreach (@{$commands}) {
+
+ my $command = @{$_}{command};
+ my $options = @{$_}{options};
+
+ if ($show_plan) {
+ print "dh plan: ".escape_shell($command, @{$options})."\n";
+ next;
+ }
+
+ run($command, @{$options});
+ }
+}
+
# Load addons, which can modify sequences.
foreach my $addon (@{$dh{WITH}}) {
my $mod="Debian::Debhelper::Sequence::$addon";
@@ -570,7 +602,7 @@ while (@ARGV_orig) {
shift @ARGV_orig;
next;
}
- elsif ($opt =~ /^--?(no-act|remaining|(after|until|before|with|without)=)/) {
+ elsif ($opt =~ /^--?(no-act|show-progress|remaining|(after|until|before|with|without)=)/) {
next;
}
elsif ($opt=~/^-/) {
@@ -700,9 +732,17 @@ foreach my $i (0..$stoppoint) {
}
}
- run($command, @opts);
+ # Save commands to run
+ push @commands, {command => $command, options => \@opts};
+}
+
+if ($dh{SHOW_PROGRESS}) {
+ # show the commands before we actually start
+ run_commands (\@commands, 1);
}
+run_commands (\@commands, 0);
+
sub run {
my $command=shift;
my @options=@_;
@@ -713,10 +753,19 @@ sub run {
# 3 space indent lines the command being run up under the
# sequence name after "dh ".
- print " ".escape_shell($command, @options)."\n";
+ if ($dh{SHOW_PROGRESS}) {
+ print "dh -> ".escape_shell($command, @options)."\n";
+ } else {
+ print " ".escape_shell($command, @options)."\n";
+ }
+
+ if ($dh{NO_ACT}) {
+ print "dh <- ".escape_shell($command, @options)."\n"
+ if $dh{SHOW_PROGRESS};
+
+ return;
+ }
- return if $dh{NO_ACT};
-
my $ret=system($command, @options);
if ($ret >> 8 != 0) {
exit $ret >> 8;
@@ -724,6 +773,10 @@ sub run {
elsif ($ret) {
exit 1;
}
+
+ if ($dh{SHOW_PROGRESS}) {
+ print "dh <- ".escape_shell($command, @options)."\n";
+ }
}
# Tries to run an override target for a command. Returns the list of
--
2.1.0
signature.asc
Description: Digital signature
Bug#760979: python-isodate: Version 0.5.0 available and debdiff for packaging changes.
Package: python-isodate
Version: 0.4.6-1
Severity: normal
Dear Maintainer,
There is a new version of isodate (0.5.0) available from:
https://pypi.python.org/pypi/isodate
I noticed this whilst working on the Ubuntu bug below:
http://bugs.launchpad.net/bugs/1349899
This bug is to introduce isodate into Ubuntu's main component. As a
result of this, I've made some packaging changes which are attached as a
debdiff.
-- System Information:
Debian Release: jessie/sid
APT prefers utopic-updates
APT policy: (500, 'utopic-updates'), (500, 'utopic-security'), (500,
'utopic'), (100, 'utopic-backports')
Architecture: amd64 (x86_64)
Foreign Architectures: i386
Kernel: Linux 3.16.0-13-generic (SMP w/1 CPU core)
Locale: LANG=en_GB.UTF-8, LC_CTYPE=en_GB.UTF-8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/dash
Versions of packages python-isodate depends on:
ii python 2.7.8-1
ii python2.7 2.7.8-6ubuntu1
python-isodate recommends no packages.
python-isodate suggests no packages.
-- no debconf information
--
Kind regards,
James Hunt.
=== modified file 'debian/changelog'
--- debian/changelog 2012-01-22 12:09:40 +
+++ debian/changelog 2014-09-09 16:13:20 +
@@ -1,3 +1,29 @@
+isodate (0.4.6-1ubuntu3) UNRELEASED; urgency=medium
+
+ * debian/control: Add Vcs-Git: and Vcs-Browser:.
+ * debian/copyright: Convert to DEP-5 format.
+
+ -- James Hunt Tue, 09 Sep 2014 17:12:34 +0100
+
+isodate (0.4.6-1ubuntu2) utopic; urgency=medium
+
+ * debian/tests/run-tests: Redirect stderr to stdout to avoid adt
+ infrastructure marking tests as failed.
+
+ -- James Hunt Tue, 09 Sep 2014 15:28:41 +0100
+
+isodate (0.4.6-1ubuntu1) utopic; urgency=medium
+
+ * debian/control:
+- Add XS-Testsuite for DEP-8 tests.
+- Update standards version.
+- Make python 2 and python 3 Descriptions unique to appease lintian.
+ * debian/tests/: Basic DEP-8 test to run test suite for python 2 and
+python 3.
+ * debian/watch: Added missing watch file.
+
+ -- James Hunt Tue, 09 Sep 2014 11:44:49 +0100
+
isodate (0.4.6-1) unstable; urgency=low
* Initial release. (Closes: #656919)
=== modified file 'debian/control'
--- debian/control 2012-01-22 12:09:40 +
+++ debian/control 2014-09-09 15:10:07 +
@@ -1,20 +1,24 @@
Source: isodate
-Maintainer: Maximiliano Curia
+Maintainer: Ubuntu Developers
+XSBC-Original-Maintainer: Maximiliano Curia
Section: python
Priority: optional
Build-Depends: python-all (>= 2.6.6-3~), python3-all (>= 3.1.2-6~), debhelper (>= 7.0.50~)
X-Python-Version: >= 2.5
X-Python3-Version: >= 3.0
-Standards-Version: 3.9.2
+Standards-Version: 3.9.5
+Vcs-Git: https://github.com/gweis/isodate.git
+Vcs-Browser: https://github.com/gweis/isodate
+XS-Testsuite: autopkgtest
Package: python-isodate
Architecture: all
Depends: ${misc:Depends}, ${python:Depends}
Description: ISO 8601 date/time/duration parser and formatter (Python module)
- This module implements ISO 8601 date, time and duration parsing.
+ This python module implements ISO 8601 date, time and duration parsing.
Package: python3-isodate
Architecture: all
Depends: ${misc:Depends}, ${python3:Depends}
Description: ISO 8601 date/time/duration parser and formatter (Python 3 module)
- This module implements ISO 8601 date, time and duration parsing.
+ This python 3 module implements ISO 8601 date, time and duration parsing.
=== modified file 'debian/copyright'
--- debian/copyright 2012-01-22 12:09:40 +
+++ debian/copyright 2014-09-09 16:08:34 +
@@ -1,93 +1,84 @@
-This package was debianized by Maximiliano Curia on
-Sun, 22 Jan 2012 13:23:26 -0300.
-
-It was downloaded from http://hg.proclos.com/isodate
-
-Upstream Author: Gerhard Weis
-
-Copyright 2009, Gerhard Weis
-
-License:
-
-# Redistribution and use in source and binary forms, with or without
-# modification, are permitted provided that the following conditions are met:
-#
-# * Redistributions of source code must retain the above copyright notice,
-#this list of conditions and the following disclaimer.
-# * Redistributions in binary form must reproduce the above copyright notice,
-#this list of conditions and the following disclaimer in the documentation
-#and/or other materials provided with the distribution.
-# * Neither the name of the authors nor the names of its contributors
-#may be used to endorse or promote products derived from this software
-#without specific prior written permission.
-#
-# THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
-# AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
-# IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
-# ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
-# LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
-# CONSEQUENTIAL DAMAGES (INCLUDING, BUT N
Bug#761177: python-sparqlwrapper: New version 1.6.4 available
Package: python-sparqlwrapper Version: 1.6.0-2 Severity: normal Dear Maintainer, Please consider updating the version available in Debian to the latest 1.6.4 release: https://github.com/RDFLib/sparqlwrapper/releases The current 1.6.0-2 version has the python3 tests disabled (syntax errors are flagged), whereas with version 1.6.4, the python3 tests pass using 'nosetests3'. -- System Information: Debian Release: jessie/sid APT prefers unstable APT policy: (500, 'unstable') Architecture: amd64 (x86_64) Kernel: Linux 3.16.0-14-generic (SMP w/4 CPU cores) Locale: LANG=C, LC_CTYPE=C (charmap=ANSI_X3.4-1968) Shell: /bin/sh linked to /bin/dash Versions of packages python-sparqlwrapper depends on: ii python-rdflib 4.1.2-3 pn python:any python-sparqlwrapper recommends no packages. python-sparqlwrapper suggests no packages. -- no debconf information Kind regards, James -- James Hunt signature.asc Description: Digital signature
