Bug#785968: cfflib: diff for NMU version 2.0.5-1.1
Control: tags 785968 + patch
I've prepared an NMU for cfflib (versioned as 2.0.5-1.1).
But I haven't tested or uploaded it, as you're caught up in the
libstdc++ transition.
I do intend to NMU this when it's buildable, though.
Unless you object :)
Regards,
SR
diff -Nru cfflib-2.0.5/debian/changelog cfflib-2.0.5/debian/changelog
--- cfflib-2.0.5/debian/changelog 2011-05-10 17:45:48.0 +0200
+++ cfflib-2.0.5/debian/changelog 2015-08-20 14:36:49.0 +0200
@@ -1,3 +1,12 @@
+cfflib (2.0.5-1.1) UNRELEASED; urgency=medium
+
+ * Non-maintainer upload.
+ * Port from python-support to dh_python2 (Closes: #785968).
+ * Drop unused and unnecessary Provides: ${python:Provides}.
+ * Replace pyversions with X-P-V.
+
+ -- Stefano Rivera Thu, 20 Aug 2015 14:00:01 +0200
+
cfflib (2.0.5-1) unstable; urgency=low
* Initial release (Closes: #624456).
diff -Nru cfflib-2.0.5/debian/control cfflib-2.0.5/debian/control
--- cfflib-2.0.5/debian/control 2011-05-10 17:45:48.0 +0200
+++ cfflib-2.0.5/debian/control 2015-08-20 14:36:33.0 +0200
@@ -2,7 +2,7 @@
Section: python
Priority: extra
Maintainer: Stephan Gerhard
-Build-Depends: debhelper (>= 7.2.18), python-all, python-support,
+Build-Depends: debhelper (>= 7.2.18), dh-python, python-all,
python-numpy,
python-sphinx,
python-nose,
@@ -12,6 +12,7 @@
Homepage: http://cmtk.org/cfflib
Vcs-Git: git://github.com/LTS5/cfflib.git
Vcs-Browser: http://github.com/LTS5/cfflib
+X-Python-Version: >= 2.6
Package: python-cfflib
Architecture: all
@@ -24,7 +25,6 @@
python-sphinx,
python-tables,
python-h5py
-Provides: ${python:Provides}
Description: Multi-modal connectome and metadata management and integration
The Connectome File Format Library (cfflib) is a Python module for
multi-modal neuroimaging connectome data and metadata management and
diff -Nru cfflib-2.0.5/debian/pyversions cfflib-2.0.5/debian/pyversions
--- cfflib-2.0.5/debian/pyversions 2011-05-10 17:45:48.0 +0200
+++ cfflib-2.0.5/debian/pyversions 1970-01-01 01:00:00.0 +0100
@@ -1 +0,0 @@
-2.6-
diff -Nru cfflib-2.0.5/debian/rules cfflib-2.0.5/debian/rules
--- cfflib-2.0.5/debian/rules 2011-05-10 17:45:48.0 +0200
+++ cfflib-2.0.5/debian/rules 2015-08-20 14:36:16.0 +0200
@@ -9,7 +9,7 @@
PYTHON=$(shell pyversions -d)
%:
- dh $@
+ dh $@ --with python2
override_dh_auto_test:
: # Do not test just after build, lets install and then test
Bug#785972: chirp: diff for NMU version 0.4.1-1.1
Control: tags 785972 + patch Control: tags 785972 + pending Dear maintainer, I've prepared an NMU for chirp (versioned as 0.4.1-1.1) and uploaded it to DELAYED/2. Please feel free to tell me if I should delay it longer. Regards, SR diff -Nru chirp-0.4.1/debian/changelog chirp-0.4.1/debian/changelog --- chirp-0.4.1/debian/changelog 2014-10-27 22:42:34.0 +0100 +++ chirp-0.4.1/debian/changelog 2015-08-20 15:10:55.0 +0200 @@ -1,3 +1,11 @@ +chirp (0.4.1-1.1) unstable; urgency=medium + + * Non-maintainer upload. + * Drop python-support Build-Depend in favour of dh-python, it's already been +ported to dh_python2 (Closes: #785972). + + -- Stefano Rivera Thu, 20 Aug 2015 15:10:55 +0200 + chirp (0.4.1-1.0) unstable; urgency=medium * Non-maintainer upload diff -Nru chirp-0.4.1/debian/control chirp-0.4.1/debian/control --- chirp-0.4.1/debian/control 2014-10-27 22:42:34.0 +0100 +++ chirp-0.4.1/debian/control 2015-08-20 15:07:36.0 +0200 @@ -3,7 +3,7 @@ Priority: optional Maintainer: Debian Hamradio Maintainers Uploaders: Steve Conklin , Kamal Mostafa , Bdale Garbee -Build-Depends: debhelper (>= 9), python-support, python (>=2.7), python-libxml2, python-serial +Build-Depends: debhelper (>= 9), dh-python, python (>=2.7), python-libxml2, python-serial Standards-Version: 3.9.5 Homepage: http://chirp.danplanet.com/ Vcs-Git: git://anonscm.debian.org/collab-maint/chirp.git
Bug#785973: crmsh: diff for NMU version 1.2.6+git+e77add-1.4
Control: tags 785973 + patch
Control: tags 785973 + pending
Dear maintainer,
I've prepared an NMU for crmsh (versioned as 1.2.6+git+e77add-1.4) and
uploaded it to DELAYED/2. Please feel free to tell me if I
should delay it longer.
Regards,
SR
diff -Nru crmsh-1.2.6+git+e77add/debian/changelog crmsh-1.2.6+git+e77add/debian/changelog
--- crmsh-1.2.6+git+e77add/debian/changelog 2014-11-22 13:12:41.0 +0100
+++ crmsh-1.2.6+git+e77add/debian/changelog 2015-08-20 15:31:26.0 +0200
@@ -1,3 +1,10 @@
+crmsh (1.2.6+git+e77add-1.4) unstable; urgency=medium
+
+ * Non-maintainer upload.
+ * Port from python-support to dh_python2 (Closes: #785973)
+
+ -- Stefano Rivera Thu, 20 Aug 2015 15:31:26 +0200
+
crmsh (1.2.6+git+e77add-1.3) unstable; urgency=low
* Non-maintainer upload.
diff -Nru crmsh-1.2.6+git+e77add/debian/control crmsh-1.2.6+git+e77add/debian/control
--- crmsh-1.2.6+git+e77add/debian/control 2014-11-22 13:12:53.0 +0100
+++ crmsh-1.2.6+git+e77add/debian/control 2015-08-20 15:23:44.0 +0200
@@ -3,13 +3,13 @@
Priority: extra
Maintainer: Debian HA Maintainers
Uploaders: Martin Loschwitz
-Build-Depends: debhelper (>= 9.0.0), cluster-glue-dev, hardening-wrapper, python-all-dev, python-support, dh-autoreconf, libcib3-dev, libcrmcluster4-dev, libcrmcommon3-dev (>= 1.1.8+git20121010), libpe-rules2-dev, libpe-status4-dev, libpengine4-dev (>= 1.1.8+git20121010), libstonithd2-dev, libtransitioner2-dev, liblrmd1-dev, libcrmservice1-dev, python-support, pkg-config, asciidoc, libxml2-utils, docbook-xml, xsltproc, docbook-xsl
+Build-Depends: debhelper (>= 9.0.0), cluster-glue-dev, hardening-wrapper, python-all-dev, dh-autoreconf, dh-python, libcib3-dev, libcrmcluster4-dev, libcrmcommon3-dev (>= 1.1.8+git20121010), libpe-rules2-dev, libpe-status4-dev, libpengine4-dev (>= 1.1.8+git20121010), libstonithd2-dev, libtransitioner2-dev, liblrmd1-dev, libcrmservice1-dev, pkg-config, asciidoc, libxml2-utils, docbook-xml, xsltproc, docbook-xsl
Standards-Version: 3.9.5
Homepage: http://savannah.nongnu.org/projects/crmsh/
Package: crmsh
Architecture: any
-Depends: ${shlibs:Depends}, ${misc:Depends}, python (>= 2.6.6-9), python-support (>= 0.90), python-lxml, python-yaml
+Depends: ${shlibs:Depends}, ${misc:Depends}, ${python:Depends}, python-lxml, python-yaml
Recommends: pacemaker (>= 1.1.8-1)
Replaces: pacemaker (<< 1.1.8-1)
Breaks: pacemaker (<< 1.1.8-1)
diff -Nru crmsh-1.2.6+git+e77add/debian/rules crmsh-1.2.6+git+e77add/debian/rules
--- crmsh-1.2.6+git+e77add/debian/rules 2012-10-09 17:54:18.0 +0200
+++ crmsh-1.2.6+git+e77add/debian/rules 2015-08-20 15:22:13.0 +0200
@@ -18,7 +18,7 @@
#export DH_VERBOSE=1
%:
- dh $@ --with autoreconf
+ dh $@ --with autoreconf,python2
override_dh_installdocs:
dh_installdocs
Bug#792631: python-librtmp: diff for NMU version 0.2.2-1.1
Control: tags 792631 + patch
Control: tags 792631 + pending
Dear maintainer,
I've prepared an NMU for python-librtmp (versioned as 0.2.2-1.1) and
uploaded it to DELAYED/2. Please feel free to tell me if I
should delay it longer.
Regards,
SR
diff -Nru python-librtmp-0.2.2/debian/changelog python-librtmp-0.2.2/debian/changelog
--- python-librtmp-0.2.2/debian/changelog 2015-05-12 19:38:08.0 +0200
+++ python-librtmp-0.2.2/debian/changelog 2015-08-21 12:55:23.0 +0200
@@ -1,3 +1,12 @@
+python-librtmp (0.2.2-1.1) UNRELEASED; urgency=medium
+
+ * Non-maintainer upload.
+ * Build-Depend on python{,3}-cffi-backend-dbg instead of
+python{,3}-cffi-dbg, it was renamed without a transition package
+(Closes: #792631).
+
+ -- Stefano Rivera Fri, 21 Aug 2015 12:54:19 +0200
+
python-librtmp (0.2.2-1) unstable; urgency=low
* New upstream version
diff -Nru python-librtmp-0.2.2/debian/control python-librtmp-0.2.2/debian/control
--- python-librtmp-0.2.2/debian/control 2015-05-12 19:38:08.0 +0200
+++ python-librtmp-0.2.2/debian/control 2015-08-21 12:54:16.0 +0200
@@ -12,8 +12,8 @@
python3-all-dbg,
python-cffi,
python3-cffi,
- python-cffi-dbg,
- python3-cffi-dbg,
+ python-cffi-backend-dbg,
+ python3-cffi-backend-dbg,
python-setuptools,
python3-setuptools,
python-singledispatch,
Bug#792631: python-librtmp: diff for NMU version 0.2.2-1.1
Hi Stefan (2015.08.21_14:20:57_+0200) > no, please continue with the upload. Rescheduled to day 0. > Thank you all for debugging and patching the issue. I missed the > original mails because they were marked as spam. I'll look into > improving that, so that I am more responsive in the future. I could have done something sooner, too. I've been ignoring this :) SR -- Stefano Rivera http://tumbleweed.org.za/ +1 415 683 3272
Bug#785947: balder2d: diff for NMU version 1.0-1.2
Dear maintainer,
The first NMU failed, because we hit a lintian autoreject on the
bundled libtinyxml. I've had a shot at unbundling it.
I've prepared an NMU for balder2d (versioned as 1.0-1.2) and
uploaded it to DELAYED/2. Please feel free to tell me if I
should delay it longer.
Regards,
SR
diff -u balder2d-1.0/SConstruct balder2d-1.0/SConstruct
--- balder2d-1.0/SConstruct
+++ balder2d-1.0/SConstruct
@@ -12,7 +12,7 @@
import glob
files = glob.glob("src/*.cpp") + glob.glob("src/*/*.cpp") + glob.glob("src/menu/*/*.cpp")+ glob.glob("src/*/*.c")
-libs = Split('guichan_sdl guichan SDL SDL_image SDL_mixer SDL_gfx physfs') + ['python%d.%d' % sys.version_info[:2]]
+libs = Split('guichan_sdl guichan SDL SDL_image SDL_mixer SDL_gfx physfs tinyxml') + ['python%d.%d' % sys.version_info[:2]]
env.ParseConfig('sdl-config --cflags --libs')
env.Program('bin/balder2d',files, LIBS=libs)
diff -u balder2d-1.0/debian/control balder2d-1.0/debian/control
--- balder2d-1.0/debian/control
+++ balder2d-1.0/debian/control
@@ -2,13 +2,13 @@
Section: games
Priority: optional
Maintainer: Bjørn Hansen
-Build-Depends: debhelper (>= 4.0.0), scons, libphysfs-dev, libguichan-dev, libsdl1.2-dev, libsdl-gfx1.2-dev, libsdl-image1.2-dev, libsdl-mixer1.2-dev, libstdc++6, python-dev (>= 2.4), python-support
+Build-Depends: debhelper (>= 4.0.0), scons, libphysfs-dev, libguichan-dev, libsdl1.2-dev, libsdl-gfx1.2-dev, libsdl-image1.2-dev, libsdl-mixer1.2-dev, libstdc++6, python-dev (>= 2.4), libtinyxml-dev
Standards-Version: 3.8.0
Homepage: http://balder.sourceforge.net/balder2d/
Package: balder2d
Architecture: any
-Depends: ${shlibs:Depends}, ${misc:Depends}, balder2d-data
+Depends: ${shlibs:Depends}, ${misc:Depends}, balder2d-data
Description: A 2D shooter in zero gravity
In Balder2D, players control small probes which shoot tiny projectiles with
which they try to destroy each other. It features a 2D overhead view of the
diff -u balder2d-1.0/debian/changelog balder2d-1.0/debian/changelog
--- balder2d-1.0/debian/changelog
+++ balder2d-1.0/debian/changelog
@@ -1,3 +1,11 @@
+balder2d (1.0-1.2) unstable; urgency=medium
+
+ * Non-maintainer upload.
+ * Remove python-support, not actually used (Closes: #785947)
+ * Use Debian's tinyxml rather than the bundled one.
+
+ -- Stefano Rivera Tue, 18 Aug 2015 17:39:53 +0200
+
balder2d (1.0-1.1) unstable; urgency=low
* Non-maintainer upload, to drop Python 2.4; thanks to Luca Falavigna for the
diff -u balder2d-1.0/debian/rules balder2d-1.0/debian/rules
--- balder2d-1.0/debian/rules
+++ balder2d-1.0/debian/rules
@@ -24,6 +24,7 @@
configure-stamp:
dh_testdir
# Add here commands to configure the package.
+ rm -rf src/tinyxml
touch configure-stamp
only in patch2:
unchanged:
--- balder2d-1.0.orig/src/configmanager.cpp
+++ balder2d-1.0/src/configmanager.cpp
@@ -20,7 +20,7 @@
#include "configmanager.h"
#include "renderer.h"
-#include "tinyxml/tinyxml.h"
+#include "tinyxml.h"
#include
using namespace Balder;
only in patch2:
unchanged:
--- balder2d-1.0.orig/src/mapparser.cpp
+++ balder2d-1.0/src/mapparser.cpp
@@ -21,7 +21,7 @@
#include "log.h"
#include "mapparser.h"
#include "gamemanager.h"
-#include "tinyxml/tinyxml.h"
+#include "tinyxml.h"
#include "filepathmanager.h"
using namespace Balder;
Bug#785969: command-not-found: diff for NMU version 0.2.38-1.1
Control: tags 785969 + patch
Control: tags 785969 + pending
Dear maintainer,
I've prepared an NMU for command-not-found (versioned as 0.2.38-1.1) and
uploaded it to DELAYED/5. Please feel free to tell me if I
should delay it longer.
Regards,
SR
diff -Nru command-not-found-0.2.38/debian/changelog command-not-found-0.2.38/debian/changelog
--- command-not-found-0.2.38/debian/changelog 2009-11-02 13:17:39.0 -0800
+++ command-not-found-0.2.38/debian/changelog 2015-08-25 17:21:33.0 -0700
@@ -1,3 +1,11 @@
+command-not-found (0.2.38-1.1) unstable; urgency=medium
+
+ * Non-maintainer upload.
+ * Port from python-support to dh_python2 (Closes: #785969)
+ * Drop XB-Python-Version, unused.
+
+ -- Stefano Rivera Tue, 25 Aug 2015 17:20:09 -0700
+
command-not-found (0.2.38-1) unstable; urgency=low
* Update to latest release
diff -Nru command-not-found-0.2.38/debian/control command-not-found-0.2.38/debian/control
--- command-not-found-0.2.38/debian/control 2009-11-02 12:00:37.0 -0800
+++ command-not-found-0.2.38/debian/control 2015-08-25 17:21:55.0 -0700
@@ -2,15 +2,13 @@
Section: admin
Priority: optional
Maintainer: Julian Andres Klode
-Build-Depends: debhelper (>= 7.0.50)
-Build-Depends-Indep: python-support
-XS-Python-Version: >= 2.5
+Build-Depends: debhelper (>= 7.0.50), dh-python, python
+X-Python-Version: >= 2.5
Standards-Version: 3.8.3
Package: command-not-found
Architecture: all
Depends: ${python:Depends}, apt-file, python-gdbm, lsb-release
-XB-Python-Version: ${python:Versions}
Description: Suggest installation of packages in interactive bash sessions
This package will install handler for command_not_found that looks up
programs not currently installed but available from the repositories.
diff -Nru command-not-found-0.2.38/debian/rules command-not-found-0.2.38/debian/rules
--- command-not-found-0.2.38/debian/rules 2009-11-02 13:14:56.0 -0800
+++ command-not-found-0.2.38/debian/rules 2015-08-25 17:15:12.0 -0700
@@ -28,7 +28,7 @@
dh_installdocs README
dh_installman debian/update-command-not-found.8
dh_installchangelogs
- dh_pysupport
+ dh_python2
dh_compress
dh_fixperms
dh_installdeb
Bug#799467: lintian: Update home of dh_gstscancodecs
Package: lintian Version: 2.5.37 Severity: normal Tags: patch This is bogus: > E: gst-plugins-dvswitch source: missing-build-dependency-for-dh_-command > dh_gstscancodecs => libgstreamer0.10-dev We're migrating to GStreamer 1.0 (#785198). So, here's a patch. SR diff --git a/data/debhelper/dh_commands b/data/debhelper/dh_commands index 774300e..a3c298e 100644 --- a/data/debhelper/dh_commands +++ b/data/debhelper/dh_commands @@ -49,7 +49,7 @@ dh_gnome=gnome-pkg-tools dh_gnome_clean=gnome-pkg-tools dh_gnustep=gnustep-make dh_golang=dh-golang -dh_gstscancodecs=libgstreamer0.10-dev +dh_gstscancodecs=libgstreamer1.0-dev dh_gtkmodules=libgtk2.0-dev dh_haskell_blurbs=haskell-devscripts dh_haskell_depends=haskell-devscripts diff --git a/data/debhelper/dh_packages b/data/debhelper/dh_packages index 62b6209..c115433 100644 --- a/data/debhelper/dh_packages +++ b/data/debhelper/dh_packages @@ -43,7 +43,7 @@ haskell-devscripts libaqbanking-dev libdbi-perl libgcj-common -libgstreamer0.10-dev +libgstreamer1.0-dev libgtk2.0-dev libimager-perl libpango1.0-dev
Bug#799485: header files moved to /usr/lib/pypy
Hi Jelmer (2015.09.19_20:04:34_+0200) > With 2.6.1+dfsg-1, header files from pypy appear to have moved from > /usr/lib/pypy/include to /usr/lib/pypy. This breaks module building, as > it still happens with the old include path specified. Oops. There's definitely my mistake in b56a15c7955711728af0e0dd80bf76c5302bfb53 SR -- Stefano Rivera http://tumbleweed.org.za/ +1 415 683 3272
Bug#825702: python-pypy.sandbox: Upgrading past 4.0.1 causes pypy-sandbox to crash on startup
Hi raylu (2016.05.29_01:02:42_+0200) > $ pypy-sandbox > 'import site' failed > debug: OperationError: > debug: operror-type: RuntimeError > debug: operror-value: None > [Subprocess exit code: 1] Running it with --verbose showed that the problem was an un-handled mkdir call. It's trying to create __pycache__ directories. Those don't exist when the pypy package isn't installed. This is easy enough to fix. SR -- Stefano Rivera http://tumbleweed.org.za/ +1 415 683 3272
Bug#858096: [mate-panel?] Mate panel unstable, notification area icons disappear without warning
Control: severity -1 normal > I have been doing some more on this issue, even though it appears no one > else is interested, and have found that it happens when the panel properties > > background is changed from "None (use system theme" to "Solid colour" and > the style is adjusted to make the panel transparent. If the panel is left > Opaque the icons stay as they should but any transparency causes the icons > to disappear. Yep. I can reproduce that. And I don't think it qualifies as an RC bug, it is easily worked around. SR -- Stefano Rivera http://tumbleweed.org.za/ +1 415 683 3272
Bug#864076: unblock: distro-info-data/0.36
Package: release.debian.org Severity: normal User: release.debian@packages.debian.org Usertags: unblock Please unblock package distro-info-data This is a pre-upload unblock request for distro-info-data, now that the Jessie release date has been announced. While I was here, I realised that we didn't have EOL dates for Jessie or Wheezy yet :( We have a long-standing bug of not including LTS dates (#782685) so I've maintained the status-quo and did that for these two as well. Alternatively, I could just extend the support dates out to include LTS, but that seems like another bad idea :/ So, are you OK with this patch-set, and would you consider allowing it in, for Stretch? unblock distro-info-data/0.36 Thanks, SR diff --git a/debian.csv b/debian.csv index c1f0962..b476031 100644 --- a/debian.csv +++ b/debian.csv @@ -10,10 +10,10 @@ version,codename,series,created,release,eol 4.0,Etch,etch,2005-06-06,2007-04-08,2010-02-15 5.0,Lenny,lenny,2007-04-08,2009-02-14,2012-02-06 6.0,Squeeze,squeeze,2009-02-14,2011-02-06,2014-05-31 -7,Wheezy,wheezy,2011-02-06,2013-05-04 -8,Jessie,jessie,2013-05-04,2015-04-25 -9,Stretch,stretch,2015-04-25 -10,Buster,buster,2018-07-01 +7,Wheezy,wheezy,2011-02-06,2013-05-04,2016-04-26 +8,Jessie,jessie,2013-05-04,2015-04-25,2018-06-06 +9,Stretch,stretch,2015-04-25,2017-06-17 +10,Buster,buster,2017-06-17 11,Bullseye,bullseye,2020-11-05 ,Sid,sid,1993-08-16 ,Experimental,experimental,1993-08-16 diff --git a/debian/changelog b/debian/changelog index cec721c..130df23 100644 --- a/debian/changelog +++ b/debian/changelog @@ -1,3 +1,14 @@ +distro-info-data (0.36) UNRELEASED; urgency=medium + + * Set EOL date for Debian Wheezy. This excludes LTS, which we haven't +supported in distro-info yet, for Debian, but matches what we did for +Squeeze. + * Set (provisional) EOL date for Debian Jessie. + * Set release date for Stretch (and matching creation date for Buster). It +has been announced. + + -- Stefano Rivera Sat, 03 Jun 2017 18:07:40 -0700 + distro-info-data (0.35) unstable; urgency=medium * Correct Ubuntu Zesty release date.
Bug#810592: python3-cffi: Does not depend on py3-dev ?
Hi Goirand (2016.01.10_05:15:52_-0800) > > Installing python3-dev fixed this. Thus I conclude that there's a > > missing dependency here. There is a Suggests: python3-dev. This goes back to earlier days in cffi. Using the verifier requires -dev at build time, but not runtime. In the new out-of-line modes, runtime use will have the cffi-backend package installed, so we can strengthen the dependency to Recommends or Depends. That said, you need -dev when building ABI-level extensions, but not API-level ones. So maybe Recommends is the right thing to use... > If that's true, then the build-depends should be python-dev-all, not > just python-dev. Yes, I think for packages using ABI-level extensions should be build-depending on -dev-all. > # [ starting a crypto program ] > /usr/lib/python3/dist-packages/cryptography/hazmat/bindings/__pycache__/_cffi__x7af6e22bxb4fc9075.c:2:20: > fatal error: Python.h: No such file or directory > #include That shouldn't happen. Packages using cffi should build extensions like that at build time, not run time. Tell me more about this python3-cryptography package you're using? SR -- Stefano Rivera http://tumbleweed.org.za/ +1 415 683 3272
Bug#814930: jessie-pu: package hplip/3.15.11+repack0-1
6 ui4/pqdiagdialog_base.py|4 ui4/printdialog.py | 155 -- ui4/printdialog_base.py |6 ui4/printernamecombobox.py |9 ui4/printsettingsdialog.py | 10 ui4/printsettingsdialog_base.py |4 ui4/printsettingstoolbox.py | 167 +- ui4/printtestpagedialog.py | 14 ui4/printtestpagedialog_base.py |4 ui4/queuesconf.py |8 ui4/readonlyradiobutton.py |2 ui4/sendfaxdialog.py| 110 - ui4/sendfaxdialog_base.py |4 ui4/settingsdialog.py |9 ui4/settingsdialog_base.py |4 ui4/setupdialog.py | 215 ++- ui4/setupdialog_base.py | 51 ui4/systemtray.py | 68 - ui4/systrayframe.py | 18 ui4/ui_utils.py | 247 ++-- ui4/upgradedialog.py|8 ui4/wifisetupdialog.py | 83 - ui4/wifisetupdialog_base.py |2 uninstall.py|8 unload.py | 179 +-- upgrade.py | 46 wificonfig.py | 14 241 files changed, 8496 insertions(+), 6187 deletions(-) This was mostly putting a feeler out, as Didier thought you may be interested in a stable update, that supported new hardware. It seems to not be the case, so maybe I should just do a backport. I'm not particularly invested in this. I just made a backport that is probably useful to others, and am trying to find the right place to put it. SR -- Stefano Rivera http://tumbleweed.org.za/ +1 415 683 3272
Bug#816043: pypy: bsddb module is broken
Control: tag -1 upstream Control: retitle -1 PyPy has no _bsddb module Hi Tristan (2016.02.26_13:35:29_-0800) > The extension module seems to be missing: Yep, upstream. SR -- Stefano Rivera http://tumbleweed.org.za/ +1 415 683 3272
Bug#814930: jessie-pu: package hplip/3.15.11+repack0-1
Package: release.debian.org Severity: normal Tags: jessie User: release.debian@packages.debian.org Usertags: pu I asked the printing people how they felt about an backport of hplip, and OdyX suggested [0]: > As far as I remember (but could never take the time to actively > check), the Debian Stable Managers were open to update packages in > Stable for hardware support (and "new HP Printer" would qualify). I > haven't checked the hplip code to see whether a full new upstream > release would make sense over backporting specific parts though. > tl;dr: I'd check with the SRMs first. How would you feel about a wholesale backport of hplip, to stable? No debdiff attached, because it's scary huge. Not even a diffstat, because: > 4362 files changed, 1703256 insertions(+), 17230 deletions(-) [0]: https://lists.debian.org/3588455.xzku8qg...@odyx.org SR
Bug#827568: marked as done (twine: please ship README)
Control: reopen -1 Bah. We have the HTML, but I would like to ship the RST too. Oh well, 1.11 is coming out next week, it can be fixed in that upload. SR -- Stefano Rivera http://tumbleweed.org.za/ +1 415 683 3272
Bug#886456: python-pip: segfault when running 'pip install --upgrade ...'
Control: tags -1 moreinfo > I was trying to upgrade ansible (which I initially installed with pip). > The result was, that the program segfaulted. I tried this with --user and > without, but it makes no difference. Can you reproduce this failure? I can't. SR -- Stefano Rivera http://tumbleweed.org.za/ +1 415 683 3272
Bug#798148: new upstream available
Hi Aurelien (2016.06.01_07:23:41_+1000) I've been prodded by timvideos people to try to get an updated usb.ids in Debian. > Anyway I have uploaded the version 008 to experimental so that people > who really need the new version can use it. Is there anything still blocking this from unstable? I don't see any relevant open bugs. > Note that it poses additional issues: > - the udeb package doesn't work anymore as it can't access the systemd > database I see 1:009-2 dropped the udeb. Is the functionality still needed? > - some packages depends on usbutils to be able to access usb.ids. It is > not provided anymore in this version, so they will break. I'll try to > identify the affected packages and start to fill bugs. I guess this is the crux of the issue. I couldn't see any of those, if they've been filed, I guess they should have blocks on this. SR -- Stefano Rivera http://tumbleweed.org.za/ +1 415 683 3272
Bug#895638: ITP: ruby-iso8601 -- Ruby parser to work with ISO 8601 dateTimes and durations
Package: wnpp Severity: wishlist Owner: Stefano Rivera * Package name: ruby-iso8601 Version : 0.10.1 Upstream Author : Arnau Siches * URL : https://github.com/arnau/ISO8601 * License : Expat Programming Lang: Ruby Description : Ruby parser to work with ISO 8601 dateTimes and durations ISO8601 is a simple implementation in Ruby of the ISO 8601 (Data elements and interchange formats - Information interchange - Representation of dates and times) standard.
Bug#895641: ITP: ruby-tomlrb -- A racc based toml parser
Package: wnpp Severity: wishlist Owner: Stefano Rivera * Package name: ruby-tomlrb Version : 1.2.6 Upstream Author : Francois Bernier * URL : https://github.com/fbernier/tomlrb * License : Expat Programming Lang: Ruby Description : A racc based toml parser A Racc based TOML Ruby parser supporting the 0.4.0 version of the spec. This is a dependency of the Chef stack. I intend to package it within the ruby team.
Bug#895641: ITP: ruby-tomlrb -- A racc based toml parser
Hi Pirate (2018.04.14_07:44:39_+0200) > There is already a ruby-toml package, > https://tracker.debian.org/pkg/ruby-toml > > Can this be used instead? Yeah, I considered patching it to use that. But I think Chef recipes can reasonably assume that tomlrb is present, when Chef itself uses it. From what I see on the TOML wiki [0] the tomlrb library is in better shape that ruby-toml. I see an issue about merging the many ruby toml implementations [1], but crickets so far... [0]: https://github.com/toml-lang/toml/wiki [1]: https://github.com/jm/toml/issues/53 SR -- Stefano Rivera http://tumbleweed.org.za/ +1 415 683 3272
Bug#882632: Bug #882632: libgit2: debian/copyright refers to CC0 by URL
Hi Nicolas (2018.01.14_11:07:19_-0800) > Please consider uploading a new version; I will take the liberty to do > a NMU if there is no reply within a week. I've sponsored an NMU to the delayed/0 queue, given there was several months of notice here. SR -- Stefano Rivera http://tumbleweed.org.za/ +1 415 683 3272
Bug#885454: closed by Holger Levsen (Bug#885454: fixed in munin 2.0.34-2)
Control: reopen -1
Control: notfixed 2.0.34-2
>* Partly revert f1f48dee so that directories are again setup correctly
> on initial installations. (Closes: #885454)
You did that for munin, but not munin-node or munin-async.
I suggest this patch.
SR
--
Stefano Rivera
http://tumbleweed.org.za/
+1 415 683 3272
From e129c85f209e903db10827642110b831424422a0 Mon Sep 17 00:00:00 2001
From: Stefano Rivera
Date: Thu, 10 May 2018 23:14:14 -0700
Subject: [PATCH] Revert more of f1f48dee so that directories are again setup
correctly on initial installations of munin-node and munin-async. (Closes:
#885454)
---
debian/changelog| 2 ++
debian/munin-async.postinst | 7 +++
debian/munin-node.postinst | 11 +++
3 files changed, 20 insertions(+)
diff --git a/debian/changelog b/debian/changelog
index 7ac9f6a8..905c6753 100644
--- a/debian/changelog
+++ b/debian/changelog
@@ -1,6 +1,8 @@
munin (2.0.37-2) UNRELEASED; urgency=medium
* Bump Standards-Version: to 4.1.4, no changes needed.
+ * Revert more of f1f48dee so that directories are again setup correctly on
+initial installations of munin-node and munin-async. (Closes: #885454)
-- Holger Levsen Thu, 05 Apr 2018 17:47:11 +
diff --git a/debian/munin-async.postinst b/debian/munin-async.postinst
index a940e507..cf104824 100644
--- a/debian/munin-async.postinst
+++ b/debian/munin-async.postinst
@@ -15,9 +15,16 @@ add_munin_async_user() {
fi
}
+initperms() {
+ chown munin-async:munin-async /var/lib/munin-async
+}
+
case "$1" in
configure)
add_munin_async_user
+ if [ -z "$2" ]; then
+ initperms
+ fi
;;
abort-upgrade|abort-deconfigure|abort-remove)
:
diff --git a/debian/munin-node.postinst b/debian/munin-node.postinst
index 845e9424..4f54d6f9 100644
--- a/debian/munin-node.postinst
+++ b/debian/munin-node.postinst
@@ -4,6 +4,14 @@ set -e
prevver="$2"
+initperms() {
+ chown munin:adm /var/log/munin
+ chmod 755 /var/log/munin
+
+ chown root:munin /etc/munin/plugin-conf.d
+ chmod 750 /etc/munin/plugin-conf.d
+}
+
init_plugins() {
TMPFILE=`mktemp /tmp/munin-node.configure.XX`
TMPFILE_STDERR=`mktemp /tmp/munin-node.configure.err.XX`
@@ -41,6 +49,9 @@ init_plugins() {
case "$1" in
configure)
+ if [ -z "$2" ]; then
+ initperms
+ fi
init_plugins
;;
triggered)
--
2.17.0
Bug#881549: munkres: Please migrate away from Epydoc if possible
Hi Kenneth (2017.11.12_16:14:10_-0800) > If possible, please consider moving away from the use of Epydoc in your > package. Epydoc is basically unmaintained upstream. Also, it is only > supported for Python 2, so it will reach its end of life along with > Python 2 sometime in 2020. Upstream has migrated to something called pdoc, which seems to be an epydoc replacement that supports Python 3. https://github.com/BurntSushi/pdoc That's not currently packaged for Debian, so I'm just going to drop these docs, for now. SR -- Stefano Rivera http://tumbleweed.org.za/ +1 415 683 3272
Bug#891087: beautifulsoup has been replaced by bs4
Source: beautifulsoup Severity: important We really should remove beautifulsoup, it has been replaced by bs4, many years ago. I've just been lazy, and haven't hassled consumers to migrate... This bug is for tracking the transition to bs4. SR
Bug#891090: archmage: python-beautifulsoup is replaced with python-bs4
Package: archmage Version: 1:0.3.1-3 Severity: important Control: block 891087 by -1 beautifulsoup version 4 was replaced as a new package, bs4, which has been in Debian for over 5 years now. beautfiulsoup (3.x) hasn't seen any maintenance since then. It's high time to remove it from the archive. Most code written against Beautiful Soup 3 will work against Beautiful Soup 4 with one simple change. All you should have to do is change the package name from BeautifulSoup to bs4, and depend on python-bs4 instead of python-beautifulsoup. There is some documentation on the migration here: https://www.crummy.com/software/BeautifulSoup/bs4/doc/#porting-code-to-bs4 Thanks, SR
Bug#891094: calibre: python-beautifulsoup is replaced with python-bs4
Package: calibre Version: 3.17.0+dfsg-1 Severity: important Control: block 891087 by -1 beautifulsoup version 4 was replaced as a new package, bs4, which has been in Debian for over 5 years now. beautfiulsoup (3.x) hasn't seen any maintenance since then. It's high time to remove it from the archive. Most code written against Beautiful Soup 3 will work against Beautiful Soup 4 with one simple change. All you should have to do is change the package name from BeautifulSoup to bs4, and depend on python-bs4 instead of python-beautifulsoup. There is some documentation on the migration here: https://www.crummy.com/software/BeautifulSoup/bs4/doc/#porting-code-to-bs4 Thanks, SR
Bug#891101: python-pyth: python-beautifulsoup is replaced with python-bs4
Package: python-pyth Version: 0.6.0-1 Severity: important Control: block 891087 by -1 beautifulsoup version 4 was replaced as a new package, bs4, which has been in Debian for over 5 years now. beautfiulsoup (3.x) hasn't seen any maintenance since then. It's high time to remove it from the archive. Most code written against Beautiful Soup 3 will work against Beautiful Soup 4 with one simple change. All you should have to do is change the package name from BeautifulSoup to bs4, and depend on python-bs4 instead of python-beautifulsoup. There is some documentation on the migration here: https://www.crummy.com/software/BeautifulSoup/bs4/doc/#porting-code-to-bs4 Thanks, SR
Bug#891097: python-freevo: python-beautifulsoup is replaced with python-bs4
Package: python-freevo Version: 1.9.2b2-4.3 Severity: important Control: block 891087 by -1 beautifulsoup version 4 was replaced as a new package, bs4, which has been in Debian for over 5 years now. beautfiulsoup (3.x) hasn't seen any maintenance since then. It's high time to remove it from the archive. Most code written against Beautiful Soup 3 will work against Beautiful Soup 4 with one simple change. All you should have to do is change the package name from BeautifulSoup to bs4, and depend on python-bs4 instead of python-beautifulsoup. There is some documentation on the migration here: https://www.crummy.com/software/BeautifulSoup/bs4/doc/#porting-code-to-bs4 Thanks, SR
Bug#891098: python-gumbo: python-beautifulsoup is replaced with python-bs4
Package: python-gumbo Version: 0.10.1+dfsg-2.1 Severity: important Control: block 891087 by -1 beautifulsoup version 4 was replaced as a new package, bs4, which has been in Debian for over 5 years now. beautfiulsoup (3.x) hasn't seen any maintenance since then. It's high time to remove it from the archive. Most code written against Beautiful Soup 3 will work against Beautiful Soup 4 with one simple change. All you should have to do is change the package name from BeautifulSoup to bs4, and depend on python-bs4 instead of python-beautifulsoup. There is some documentation on the migration here: https://www.crummy.com/software/BeautifulSoup/bs4/doc/#porting-code-to-bs4 Thanks, SR
Bug#891099: python-pattern: python-beautifulsoup is replaced with python-bs4
Package: python-pattern Version: 2.6+git20150109-2 Severity: important Control: block 891087 by -1 beautifulsoup version 4 was replaced as a new package, bs4, which has been in Debian for over 5 years now. beautfiulsoup (3.x) hasn't seen any maintenance since then. It's high time to remove it from the archive. Most code written against Beautiful Soup 3 will work against Beautiful Soup 4 with one simple change. All you should have to do is change the package name from BeautifulSoup to bs4, and depend on python-bs4 instead of python-beautifulsoup. There is some documentation on the migration here: https://www.crummy.com/software/BeautifulSoup/bs4/doc/#porting-code-to-bs4 Thanks, SR
Bug#891096: planet-venus: python-beautifulsoup is replaced with python-bs4
Package: planet-venus Version: 0~git9de2109-4 Severity: important Control: block 891087 by -1 beautifulsoup version 4 was replaced as a new package, bs4, which has been in Debian for over 5 years now. beautfiulsoup (3.x) hasn't seen any maintenance since then. It's high time to remove it from the archive. Most code written against Beautiful Soup 3 will work against Beautiful Soup 4 with one simple change. All you should have to do is change the package name from BeautifulSoup to bs4, and depend on python-bs4 instead of python-beautifulsoup. There is some documentation on the migration here: https://www.crummy.com/software/BeautifulSoup/bs4/doc/#porting-code-to-bs4 Thanks, SR
Bug#891093: guake-indicator: python-beautifulsoup is replaced with python-bs4
Package: guake-indicator Version: 1.1-2 Severity: important Control: block 891087 by -1 beautifulsoup version 4 was replaced as a new package, bs4, which has been in Debian for over 5 years now. beautfiulsoup (3.x) hasn't seen any maintenance since then. It's high time to remove it from the archive. Most code written against Beautiful Soup 3 will work against Beautiful Soup 4 with one simple change. All you should have to do is change the package name from BeautifulSoup to bs4, and depend on python-bs4 instead of python-beautifulsoup. There is some documentation on the migration here: https://www.crummy.com/software/BeautifulSoup/bs4/doc/#porting-code-to-bs4 Thanks, SR
Bug#891100: gourmet: python-beautifulsoup is replaced with python-bs4
Package: gourmet Version: 0.17.4-5 Severity: important Control: block 891087 by -1 beautifulsoup version 4 was replaced as a new package, bs4, which has been in Debian for over 5 years now. beautfiulsoup (3.x) hasn't seen any maintenance since then. It's high time to remove it from the archive. Most code written against Beautiful Soup 3 will work against Beautiful Soup 4 with one simple change. All you should have to do is change the package name from BeautifulSoup to bs4, and depend on python-bs4 instead of python-beautifulsoup. There is some documentation on the migration here: https://www.crummy.com/software/BeautifulSoup/bs4/doc/#porting-code-to-bs4 Thanks, SR
Bug#891107: geneagrapher: python-beautifulsoup is replaced with python-bs4
Package: geneagrapher Version: 1.0c2+git20120704-2 Severity: important Control: block 891087 by -1 beautifulsoup version 4 was replaced as a new package, bs4, which has been in Debian for over 5 years now. beautfiulsoup (3.x) hasn't seen any maintenance since then. It's high time to remove it from the archive. Most code written against Beautiful Soup 3 will work against Beautiful Soup 4 with one simple change. All you should have to do is change the package name from BeautifulSoup to bs4, and depend on python-bs4 instead of python-beautifulsoup. There is some documentation on the migration here: https://www.crummy.com/software/BeautifulSoup/bs4/doc/#porting-code-to-bs4 Thanks, SR
Bug#891095: ibid: python-beautifulsoup is replaced with python-bs4
Package: ibid Version: 0.1.1+dfsg-4 Severity: important Control: block 891087 by -1 beautifulsoup version 4 was replaced as a new package, bs4, which has been in Debian for over 5 years now. beautfiulsoup (3.x) hasn't seen any maintenance since then. It's high time to remove it from the archive. Most code written against Beautiful Soup 3 will work against Beautiful Soup 4 with one simple change. All you should have to do is change the package name from BeautifulSoup to bs4, and depend on python-bs4 instead of python-beautifulsoup. There is some documentation on the migration here: https://www.crummy.com/software/BeautifulSoup/bs4/doc/#porting-code-to-bs4 Thanks, SR
Bug#891106: websploit: python-beautifulsoup is replaced with python-bs4
Package: websploit Version: 3.0.0-1 Severity: important Control: block 891087 by -1 beautifulsoup version 4 was replaced as a new package, bs4, which has been in Debian for over 5 years now. beautfiulsoup (3.x) hasn't seen any maintenance since then. It's high time to remove it from the archive. Most code written against Beautiful Soup 3 will work against Beautiful Soup 4 with one simple change. All you should have to do is change the package name from BeautifulSoup to bs4, and depend on python-bs4 instead of python-beautifulsoup. There is some documentation on the migration here: https://www.crummy.com/software/BeautifulSoup/bs4/doc/#porting-code-to-bs4 Thanks, SR
Bug#891103: uicilibris: python-beautifulsoup is replaced with python-bs4
Package: uicilibris Version: uicilibris Severity: important Control: block 891087 by -1 beautifulsoup version 4 was replaced as a new package, bs4, which has been in Debian for over 5 years now. beautfiulsoup (3.x) hasn't seen any maintenance since then. It's high time to remove it from the archive. Most code written against Beautiful Soup 3 will work against Beautiful Soup 4 with one simple change. All you should have to do is change the package name from BeautifulSoup to bs4, and depend on python-bs4 instead of python-beautifulsoup. There is some documentation on the migration here: https://www.crummy.com/software/BeautifulSoup/bs4/doc/#porting-code-to-bs4 Thanks, SR
Bug#891102: python-tvrage: python-beautifulsoup is replaced with python-bs4
Package: python-tvrage Version: 0.4.1-1 Severity: important Control: block 891087 by -1 beautifulsoup version 4 was replaced as a new package, bs4, which has been in Debian for over 5 years now. beautfiulsoup (3.x) hasn't seen any maintenance since then. It's high time to remove it from the archive. Most code written against Beautiful Soup 3 will work against Beautiful Soup 4 with one simple change. All you should have to do is change the package name from BeautifulSoup to bs4, and depend on python-bs4 instead of python-beautifulsoup. There is some documentation on the migration here: https://www.crummy.com/software/BeautifulSoup/bs4/doc/#porting-code-to-bs4 Thanks, SR
Bug#891105: webcheck: python-beautifulsoup is replaced with python-bs4
Package: webcheck Version: 1.10.4-1 Severity: important Control: block 891087 by -1 beautifulsoup version 4 was replaced as a new package, bs4, which has been in Debian for over 5 years now. beautfiulsoup (3.x) hasn't seen any maintenance since then. It's high time to remove it from the archive. Most code written against Beautiful Soup 3 will work against Beautiful Soup 4 with one simple change. All you should have to do is change the package name from BeautifulSoup to bs4, and depend on python-bs4 instead of python-beautifulsoup. There is some documentation on the migration here: https://www.crummy.com/software/BeautifulSoup/bs4/doc/#porting-code-to-bs4 Thanks, SR
Bug#891104: wapiti: python-beautifulsoup is replaced with python-bs4
Package: wapiti Version: 2.3.0+dfsg-6 Severity: important Control: block 891087 by -1 beautifulsoup version 4 was replaced as a new package, bs4, which has been in Debian for over 5 years now. beautfiulsoup (3.x) hasn't seen any maintenance since then. It's high time to remove it from the archive. Most code written against Beautiful Soup 3 will work against Beautiful Soup 4 with one simple change. All you should have to do is change the package name from BeautifulSoup to bs4, and depend on python-bs4 instead of python-beautifulsoup. There is some documentation on the migration here: https://www.crummy.com/software/BeautifulSoup/bs4/doc/#porting-code-to-bs4 Thanks, SR
Bug#891185: transition: re2
Package: release.debian.org Severity: normal User: release.debian@packages.debian.org Usertags: transition Hi, re2 is C++ and likes to have transitions. Not many reverse-deps, though :) It's in experimental. I've test built the reverse-depends, and didn't see any new failures. I can't get chromium-browser to build before or after the transition, but presumably it's fine, Google would be targeting the latest re2 anyway. Reportbug Ben file: title = "re2"; is_affected = .depends ~ "libre2-3" | .depends ~ "libre2-4"; is_good = .depends ~ "libre2-4"; is_bad = .depends ~ "libre2-3"; https://release.debian.org/transitions/html/auto-re2.html Looks good, though. SR
Bug#860864: unblock: distro-info-data/0.35
Package: release.debian.org Severity: normal User: release.debian@packages.debian.org Usertags: unblock Please unblock package distro-info-data Ubuntu 17.04 has now released, and we need to add 17.10. Of course there will still need to be one more update of distro-info-data once stretch has a release date (that could be after the release). unblock distro-info-data/0.35 Thanks, SR diff -Nru distro-info-data-0.33/debian/changelog distro-info-data-0.35/debian/changelog --- distro-info-data-0.33/debian/changelog 2017-01-15 15:53:52.0 -0800 +++ distro-info-data-0.35/debian/changelog 2017-04-20 19:43:47.0 -0700 @@ -1,3 +1,15 @@ +distro-info-data (0.35) unstable; urgency=medium + + * Correct Ubuntu Zesty release date. + + -- Stefano Rivera Thu, 20 Apr 2017 19:43:47 -0700 + +distro-info-data (0.34) unstable; urgency=medium + + * Add Ubuntu 17.10, Artful Aardvark. + + -- Stefano Rivera Thu, 20 Apr 2017 16:42:23 -0700 + distro-info-data (0.33) unstable; urgency=medium * Add Debian 11 codename (with provisional creation date) (Closes: #851447) diff -Nru distro-info-data-0.33/ubuntu.csv distro-info-data-0.35/ubuntu.csv --- distro-info-data-0.33/ubuntu.csv2016-10-21 15:48:30.0 -0700 +++ distro-info-data-0.35/ubuntu.csv2017-04-20 19:43:47.0 -0700 @@ -24,4 +24,5 @@ 15.10,Wily Werewolf,wily,2015-04-23,2015-10-22,2016-07-22 16.04 LTS,Xenial Xerus,xenial,2015-10-22,2016-04-21,2021-04-21 16.10,Yakkety Yak,yakkety,2016-04-21,2016-10-13,2017-07-20 -17.04,Zesty Zapus,zesty,2016-10-13,2017-04-20,2018-01-25 +17.04,Zesty Zapus,zesty,2016-10-13,2017-04-13,2018-01-25 +17.10,Artful Aardvark,artful,2017-04-13,2017-10-19,2018-07-19
Bug#828517: OpenSSL transition severity
Hi Kurt (2016.10.27_00:35:51_-0700) > I just took a pass at it, and I think, did most of the port in an > afternoon and evening. Landed upstream, it'll be in the next upstream release. SR -- Stefano Rivera http://tumbleweed.org.za/ +1 415 683 3272
Bug#835937: beets: FTBFS randomly (test_nonexistent_file fails)
Hi Santiago (2016.08.29_06:44:11_-0700) > The failure happens randomly: It always happen in a certain > autobuilder I have, and it does never happen on the others. Yeah I saw this when I uploaded to the archive, and have been meaning to dig deeper. Thanks for doing the legwork :) SR -- Stefano Rivera http://tumbleweed.org.za/ +1 415 683 3272
Bug#834545: transition: re2
Hi Emilio (2016.08.29_09:58:46_+0200) > Go ahead. Would you mind if I held back for the next release, due on the 1st? SR -- Stefano Rivera http://tumbleweed.org.za/ +1 415 683 3272
Bug#834545: transition: re2
Hi Emilio (2016.08.31_00:35:21_+0200) > > Would you mind if I held back for the next release, due on the 1st? > > 1st of September? > > That'd be fine. Yep. Upstream does monthly snapshots, rather than releases. And seem to be moving rather fast atm. SR -- Stefano Rivera http://tumbleweed.org.za/ +1 415 683 3272
Bug#834545: transition: re2
Hi Emilio (2016.08.31_00:35:21_+0200) > > Would you mind if I held back for the next release, due on the 1st? So, that is staged in git and ready to go. It will require a 1-line patch to ocaml-re2 (inserting an std::), and ruby-re2 should be binnmuable. chromium-browser, libphonenumber, and hhvm all have unrelated FTBFSs at the moment. SR -- Stefano Rivera http://tumbleweed.org.za/ +1 415 683 3272
Bug#836907: src:hidapi-cffi: Missing dependencies on python{3,}-cffi
Package: src:hidapi-cffi
Version: 0.2.1-1
Severity: serious
Tags: patch
Justification: Policy 7.2
Upstream only declares
> setup_requires=['cffi >= 0.8']
But that's insufficient, it actually requires cffi at runtime, the first
line is:
> from cffi import FFI
It's worth noting that in Debian we split the cffi package from the
cffi-backend package. We assume that cffi packages are using the
out-of-line modes, and they only need the cffi-backend. So, when a
package declares cffi in install_requires, we generate a dependency on
the cffi-backend package.
But this package is using the in-line ABI mode (see [0] for an
explanation of the modes). So it will need an explicit dependency on
cffi, even if upstream fixes setup.py
[0]: https://cffi.readthedocs.io/en/latest/overview.html
Patch attached: dependencies.patch
Also, I noticed a bunch of things in debian/rules that pybulid can do
for you, so there's a second patch: pybulid.patch.
SR
diff -Nru hidapi-cffi-0.2.1/debian/control hidapi-cffi-0.2.1/debian/control
--- hidapi-cffi-0.2.1/debian/control 2015-06-25 13:35:38.0 -0700
+++ hidapi-cffi-0.2.1/debian/control 2016-09-06 22:04:59.0 -0700
@@ -7,14 +7,14 @@
Package: python-hidapi
Architecture: any
-Depends: ${misc:Depends}, ${python:Depends}, libhidapi-hidraw0 | libhidapi-libusb0
+Depends: ${misc:Depends}, ${python:Depends}, libhidapi-hidraw0 | libhidapi-libusb0, python-cffi (>= 0.8)
Description: Python bindings for the HID API
Python bindings for libhidapi for working with Human Interface Devices
such as mouses and keyboards.
Package: python3-hidapi
Architecture: any
-Depends: ${misc:Depends}, ${python3:Depends}, libhidapi-hidraw0 | libhidapi-libusb0
+Depends: ${misc:Depends}, ${python3:Depends}, libhidapi-hidraw0 | libhidapi-libusb0, python3-cffi (>= 0.8)
Description: Python bindings for the HID API
Python bindings for libhidapi for working with Human Interface Devices
such as mouses and keyboards.
diff -Nru hidapi-cffi-0.2.1/debian/clean hidapi-cffi-0.2.1/debian/clean
--- hidapi-cffi-0.2.1/debian/clean 1969-12-31 16:00:00.0 -0800
+++ hidapi-cffi-0.2.1/debian/clean 2016-09-06 22:04:59.0 -0700
@@ -0,0 +1 @@
+hidapi_cffi.egg-info/*
diff -Nru hidapi-cffi-0.2.1/debian/rules hidapi-cffi-0.2.1/debian/rules
--- hidapi-cffi-0.2.1/debian/rules 2015-06-25 13:08:50.0 -0700
+++ hidapi-cffi-0.2.1/debian/rules 2016-09-06 22:04:59.0 -0700
@@ -1,26 +1,6 @@
#!/usr/bin/make -f
-# This file was automatically generated by stdeb 0.8.2 at
-# Thu, 09 Oct 2014 22:55:33 +0300
-PYTHONS:=$(shell pyversions -vr)
-PYTHON3S:=$(shell py3versions -vr)
export PYBUILD_NAME=hidapi
+
%:
dh $@ --with python2,python3 --buildsystem=pybuild
-
-override_dh_clean:
- dh_clean -O--buildsystem=pybuild
- rm -rf build
- rm -rf __pycache__
-
-override_dh_install:
- set -e ; for pyvers in $(PYTHONS); do \
- python$$pyvers setup.py install --install-layout=deb \
- --root $(CURDIR)/debian/python-hidapi; \
- done
- set -e ; for pyvers in $(PYTHON3S); do \
- python$$pyvers setup.py install --install-layout=deb \
- --root $(CURDIR)/debian/python3-hidapi; \
- done
- rm -rf $(CURDIR)/debian/python*-hidapi/usr/lib/python*/dist-packages/*.pth
- rm -rf $(CURDIR)/debian/python*-hidapi/usr/lib/python3.*
Bug#828929: cairocffi: FTBFS with gcc-6 and glibc 2.23
Hi Jean-Christophe (2016.07.02_11:15:36_-0700)
> I don’t think it has anything to do with gcc/glibc.
I agree. I suggest doing a build with a debug python interpreter, then
you get a very helpful error message:
$ LC_ALL=C.UTF-8 xvfb-run -a --server-args="-screen 0 1024x768x24" python3-dbg
-m pytest
test session starts
platform linux -- Python 3.5.2+, pytest-3.0.2, py-1.4.31, pluggy-0.3.1
rootdir: /build/cairocffi-op2T2E/cairocffi-0.7.2, inifile:
collected 44 items
cairocffi/test_cairo.py ..
cairocffi/test_pixbuf.py
cairocffi/test_xcb.py FF
= FAILURES
==
__ test_xcb_pixmap
__
xcb_conn =
def test_xcb_pixmap(xcb_conn):
width = 10
height = 10
# create a new window
wid = create_window(xcb_conn, width, height)
# create the pixmap used to draw with cairo
pixmap = create_pixmap(xcb_conn, wid, width, height)
# create graphics context to copy pixmap on window
gc = create_gc(xcb_conn)
# create XCB surface on pixmap
root_visual = find_root_visual(xcb_conn)
> surface = XCBSurface(xcb_conn, pixmap, root_visual, width, height)
cairocffi/test_xcb.py:128:
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
_ _ _ _ _ _ _
self =
conn = , drawable = 2097153
visual = , width = 10,
height = 10
def __init__(self, conn, drawable, visual, width, height):
c_visual = visualtype_to_c_struct(visual)
p = cairo.cairo_xcb_surface_create(
> conn._conn, drawable, c_visual, width, height)
E TypeError: initializer for ctype 'xcb_connection_t *' appears indeed to
be 'xcb_connection_t *', but the types are different (check that you are not
e.g. mixing up different ffi instances)
cairocffi/xcb.py:38: TypeError
__ test_xcb_window
__
xcb_conn =
@pytest.mark.xfail(cairo_version() < 11200,
reason="Cairo version too low")
def test_xcb_window(xcb_conn):
width = 10
height = 10
# create a new window used to draw with cairo
wid = create_window(xcb_conn, width, height)
# map the window and wait for it to appear
xcb_conn.core.MapWindow(wid)
xcb_conn.flush()
start = time.time()
while time.time() < start + 10:
event = xcb_conn.wait_for_event()
if isinstance(event, xcffib.xproto.ExposeEvent):
break
else:
pytest.fail("Never received ExposeEvent")
# create XCB surface on window
root_visual = find_root_visual(xcb_conn)
> surface = XCBSurface(xcb_conn, wid, root_visual, width, height)
cairocffi/test_xcb.py:192:
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
_ _ _ _ _ _ _
self =
conn = , drawable = 2097152
visual = , width = 10,
height = 10
def __init__(self, conn, drawable, visual, width, height):
c_visual = visualtype_to_c_struct(visual)
p = cairo.cairo_xcb_surface_create(
> conn._conn, drawable, c_visual, width, height)
E TypeError: initializer for ctype 'xcb_connection_t *' appears indeed to
be 'xcb_connection_t *', but the types are different (check that you are not
e.g. mixing up different ffi instances)
cairocffi/xcb.py:38: TypeError
======== 2 failed, 42 passed in 0.87 seconds
--
Stefano Rivera
http://tumbleweed.org.za/
+1 415 683 3272
Bug#828517: OpenSSL transition severity
Hi Kurt (2016.10.26_10:50:40_-0700) > severity 828517 serious FYI, I chatted about this with the upstream core developers, last month. They're estimating that it's 3 months' work, and haven't started on it, yet. (The OpenSSL binding needs to be entirely re-implemented to support the now opaque data structures.) So I have a pretty low expectation of this happening in time. SR -- Stefano Rivera http://tumbleweed.org.za/ +1 415 683 3272
Bug#828517: OpenSSL transition severity
Hi Kurt (2016.10.26_21:22:48_+0200) > Having really no idea what you're talking about, the cases that > are problematic is that they created a function like "rsa_set_p" > and we have a function RSA_set0_factors() that allows to set p but > you need to pass q at the same time and it's not allowed to set it > to NULL. It does bind at a pretty low level (lots of structs reproduced), but you're right that I don't think it's as bad as they thought. I just took a pass at it, and I think, did most of the port in an afternoon and evening. It passes its own tests, and translates, but the resulting binary segfaults on some stdlib tests. SR -- Stefano Rivera http://tumbleweed.org.za/ +1 415 683 3272
Bug#843992: ITP: ruby-knife-acl -- Knife plugin to manupulate Chef server access control lists
Package: wnpp Severity: wishlist Owner: Stefano Rivera * Package name: ruby-knife-acl Version : 1.0.3 Upstream Author : Seth Falcon , Jeremiah Snapp * URL : https://github.com/chef/knife-acl * License : Apache-2.0 Programming Lang: Ruby Description : Knife plugin to manupulate Chef server access control lists
Bug#840673: dput missing a dependency on python setuptools library [and 1 more messages]
Hi Matthias (2016.11.09_21:34:14_+) > it's surprising that dput would need a dependency on the setuptools egg > instead > of the pkg_resources egg. A dependency on the setuptools egg just sounds > plain > wrong. I think what's going on here is that the egg-info for setuptools is in the setuptools package, not the pkg-resources package. pkg-resources is refusing to do anything unless it can resolve the requirements of the package in question, which in this case includes setuptools. So even though it has everything it needs on hand, it isn't doing anything. If we're rewriting dependencies from setuptools to pkg-recources, we should ship the setuptools egg-info in pkg-resources. SR -- Stefano Rivera http://tumbleweed.org.za/ +1 415 683 3272
Bug#844735: libusb-1.0: Build fxload package from libusb-1.0 examples
Source: libusb-1.0 Affects: fxload Severity: normal Tags: patch fxload upstream seems to have stalled [0]. [0]: http://linux-hotplug.cvs.sourceforge.net/viewvc/linux-hotplug/fxload/ I think the best maintained version of fxload these days is in the libusb examples [1,2]. [1]: https://github.com/libusb/libusb/blob/master/examples/fxload.c [2]: https://github.com/libusb/libusb/commits/master/examples/fxload.c For flashing the Opsis [3] boards, which the DebConf Video team uses, we need to use a libusb version of fxload. Currently we're using a patchset [4] that I don't see landing in a dead upstream, any time soon :( [3]: https://opsis.hdmi2usb.tv/getting-started/jtag.html [4]: https://git.launchpad.net/~timvideos/+git/fxload/tree/debian/patches/multios So, the best path forward is probably libusb's fxload, in examples. Can we take over the fxload package with it? SR
Bug#844735: libusb-1.0: Build fxload package from libusb-1.0 examples
Almost forgot. I have a patch.
It doesn't include a manpage, as there isn't one upstream :(
SR
--
Stefano Rivera
http://tumbleweed.org.za/
+1 415 683 3272
diff -Nru libusb-1.0-1.0.21/debian/control libusb-1.0-1.0.21/debian/control
--- libusb-1.0-1.0.21/debian/control2016-10-26 17:53:41.0 +0200
+++ libusb-1.0-1.0.21/debian/control2016-11-18 13:33:48.0 +0100
@@ -56,3 +56,12 @@
of Linux kernel internals.
.
This is a minimal package for use in debian-installer.
+
+Package: fxload
+Section: admin
+Architecture: any
+Depends: ${misc:Depends}
+Description: Firmware download to EZ-USB devices
+ This program is conveniently able to download firmware into FX and FX2 ez-usb
+ devices. It is intended to be invoked by hotplug scripts when the unprogrammed
+ device appears on the bus.
diff -Nru libusb-1.0-1.0.21/debian/fxload.install
libusb-1.0-1.0.21/debian/fxload.install
--- libusb-1.0-1.0.21/debian/fxload.install 1970-01-01 01:00:00.0
+0100
+++ libusb-1.0-1.0.21/debian/fxload.install 2016-11-18 13:33:49.0
+0100
@@ -0,0 +1 @@
+/usr/sbin/fxload
diff -Nru libusb-1.0-1.0.21/debian/patches/install-fxload
libusb-1.0-1.0.21/debian/patches/install-fxload
--- libusb-1.0-1.0.21/debian/patches/install-fxload 1970-01-01
01:00:00.0 +0100
+++ libusb-1.0-1.0.21/debian/patches/install-fxload 2016-11-18
13:33:49.0 +0100
@@ -0,0 +1,12 @@
+--- a/examples/Makefile.am
b/examples/Makefile.am
+@@ -1,7 +1,8 @@
+ AM_CPPFLAGS = -I$(top_srcdir)/libusb
+ LDADD = ../libusb/libusb-1.0.la
+
+-noinst_PROGRAMS = listdevs xusb fxload hotplugtest testlibusb
++noinst_PROGRAMS = listdevs xusb hotplugtest testlibusb
++sbin_PROGRAMS = fxload
+
+ if HAVE_SIGACTION
+ noinst_PROGRAMS += dpfp
diff -Nru libusb-1.0-1.0.21/debian/patches/series
libusb-1.0-1.0.21/debian/patches/series
--- libusb-1.0-1.0.21/debian/patches/series 2016-08-25 10:30:40.0
+0200
+++ libusb-1.0-1.0.21/debian/patches/series 2016-11-18 13:33:49.0
+0100
@@ -1 +1,2 @@
gnu-hurd-stub.diff
+install-fxload
diff -Nru libusb-1.0-1.0.21/debian/rules libusb-1.0-1.0.21/debian/rules
--- libusb-1.0-1.0.21/debian/rules 2016-10-26 17:43:38.0 +0200
+++ libusb-1.0-1.0.21/debian/rules 2016-11-18 13:33:49.0 +0100
@@ -20,10 +20,12 @@
override_dh_auto_build-arch:
dh_auto_build --builddirectory build-deb
+ dh_auto_build --builddirectory build-deb/examples
dh_auto_build --builddirectory build-udeb
override_dh_auto_install-arch:
dh_auto_install --builddirectory build-deb
--destdir=$(CURDIR)/debian/tmp-deb
+ dh_auto_install --builddirectory build-deb/examples
--destdir=$(CURDIR)/debian/tmp-deb
# Move the library to /lib
mkdir -p $(CURDIR)/debian/tmp-deb/lib/$(DEB_HOST_MULTIARCH)/
mv
$(CURDIR)/debian/tmp-deb/usr/lib/$(DEB_HOST_MULTIARCH)/libusb-1.0.so.* \
Bug#844735: libusb-1.0: Build fxload package from libusb-1.0 examples
Hi Debian (2016.11.18_15:36:47_+0100) > For flashing the Opsis [3] boards, which the DebConf Video team uses, we > need to use a libusb version of fxload. Bah, I don't know where I got that idea from. (Well I do, but it was confused.) That said, libusb still has the better maintained fxload :) SR -- Stefano Rivera http://tumbleweed.org.za/ +1 415 683 3272
Bug#844748: openocd: Please package upstream snapshot with jtagspi
Package: openocd Version: 0.9.0-1+b1 Severity: normal Is there any chance of getting an upstream snapshot that includes the jtagspi driver? It was added in d25355473da9a925a696183a9947aac292cd2f60 upstream (Jul 2015). We (DebConf video team) need it to flash our HDMI capture boards. So, having a package (even in experimental) would be useful. SR
Bug#845044: voctomix-gui: Missing .desktop file
Package: voctomix-gui Version: 0.4-1 Severity: minor It'd be really nice if voctogui had a .desktop file, and appeared in menus. It'll need a logo or something... SR
Bug#845107: ITP: hdmi2usb-mode-switch -- Configuration and firmware tool for HDMI2USB devices
Package: wnpp Severity: wishlist Owner: Stefano Rivera * Package name: hdmi2usb-mode-switch Version : 0.0.0+git20161016-1 Upstream Author : Tim 'mithro' Ansell * URL : https://github.com/timvideos/HDMI2USB-mode-switch * License : Apache-2.0 Programming Lang: Python 3 Description : Configuration and firmware tool for HDMI2USB devices This is the tool for flashing and configuring the HDMI2USB devices. It can load a runtime firmware, and write firmware to the device's flash. https://hdmi2usb.tv/ is an open hardware and software project for capturing HDMI video with an FPGA board. This package supports the Digilent Atlys and Numato Opsis boards.
Bug#845120: RM: gst-plugins-dvswitch -- ROM; dvswitch is dead
Package: ftp.debian.org Severity: normal We took over the package, and cleaned it up, this time last year, so that we could use HDMI2USB boards with dvswitch. dvswitch itself was already dead, and we were using it on ancient wheezy installs. The DebConf Video team has switched to voctomix, so this can now die. SR
Bug#780169: jessie-pu: package youtube-dl/2014.08.05-1jessie0.1
Package: release.debian.org Severity: normal Tags: jessie User: release.debian@packages.debian.org Usertags: pu youtube-dl 2014.08.05-1 requires SSLv3 in https requests, and doesn't use protocol negotiation. See #780059. This, besides being bad, and not very future-proof has caused complete incompatibility with Python 2.7.9, which dropped the PROTOCOL_SSLv3 attribute from the ssl module. This bug has been fixed, by the upstream, in unstable. But at this point in the freeze, I doubt you'd consider letting the latest upstream version migrate to testing. So, please consider this t-p-u upload: diff -Nru youtube-dl-2014.08.05/debian/changelog youtube-dl-2014.08.05/debian/changelog --- youtube-dl-2014.08.05/debian/changelog 2014-08-06 11:43:31.0 -0700 +++ youtube-dl-2014.08.05/debian/changelog 2015-03-09 17:15:30.0 -0700 @@ -1,3 +1,11 @@ +youtube-dl (2014.08.05-1jessie0.1) testing; urgency=medium + + * Non-maintainer upload. + * Use SSL protocol negotiation, rather than requiring SSLv3 (which is no +longer supported in python 2.7.9). Closes: #780059. + + -- Stefano Rivera Mon, 09 Mar 2015 17:14:45 -0700 + youtube-dl (2014.08.05-1) unstable; urgency=medium * Imported Upstream version 2014.08.05. diff -Nru youtube-dl-2014.08.05/debian/patches/no-sslv3 youtube-dl-2014.08.05/debian/patches/no-sslv3 --- youtube-dl-2014.08.05/debian/patches/no-sslv3 1969-12-31 16:00:00.0 -0800 +++ youtube-dl-2014.08.05/debian/patches/no-sslv3 2015-03-09 17:09:54.0 -0700 @@ -0,0 +1,34 @@ +Description: Support Python 2.7.9, which removed PROTOCOL_SSLv3 + In fact, don't try to force an SSL version at all. Debian OpenSSL doesn't + support insecure versions. + Upstream use Python's default SSL handshake since + https://github.com/rg3/youtube-dl/commit/0db261ba567cb5370455d67c4398e11e5e2119f8 + And switches to TLSv1 in legacy paths in + https://github.com/rg3/youtube-dl/commit/d79323136fabc2cd72afc7c124e17797e32df514 +Author: Stefano Rivera +Bug-Debian: https://bugs.debian.org/780059 +Forwarded: not-needed +Last-Update: 2015-03-08 + +--- a/youtube_dl/utils.py b/youtube_dl/utils.py +@@ -588,17 +588,14 @@ + if getattr(self, '_tunnel_host', False): + self.sock = sock + self._tunnel() +-try: +-self.sock = ssl.wrap_socket(sock, self.key_file, self.cert_file, ssl_version=ssl.PROTOCOL_SSLv3) +-except ssl.SSLError: +-self.sock = ssl.wrap_socket(sock, self.key_file, self.cert_file, ssl_version=ssl.PROTOCOL_SSLv23) ++self.sock = ssl.wrap_socket(sock, self.key_file, self.cert_file, ssl_version=ssl.PROTOCOL_SSLv23) + + class HTTPSHandlerV3(compat_urllib_request.HTTPSHandler): + def https_open(self, req): + return self.do_open(HTTPSConnectionV3, req) + return HTTPSHandlerV3(**kwargs) + else: +-context = ssl.SSLContext(ssl.PROTOCOL_SSLv3) ++context = ssl.SSLContext(ssl.PROTOCOL_SSLv23) + context.verify_mode = (ssl.CERT_NONE +if opts_no_check_certificate +else ssl.CERT_REQUIRED) diff -Nru youtube-dl-2014.08.05/debian/patches/series youtube-dl-2014.08.05/debian/patches/series --- youtube-dl-2014.08.05/debian/patches/series 1969-12-31 16:00:00.0 -0800 +++ youtube-dl-2014.08.05/debian/patches/series 2015-03-08 13:43:36.0 -0700 @@ -0,0 +1 @@ +no-sslv3 -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Bug#768426: developers-reference: version scheme in §5.13.3 (“Direct updates to testing”)
Control: tags -1 + patch Here's a patch, implementing this change. I just proposed a t-p-u upload, following the scheme in dev-ref, and was reprimanded :) SR -- Stefano Rivera http://tumbleweed.org.za/ +1 415 683 3272 diff --git a/pkgs.dbk b/pkgs.dbk index 5a34b95..a21c24b 100644 --- a/pkgs.dbk +++ b/pkgs.dbk @@ -2741,11 +2741,11 @@ updates through unstable are possible, if the upload via unstable does not pull in any new dependencies. -Version numbers are usually selected by adding the codename of the -testing distribution and a running number, like -1.2squeeze1 for the first upload through -testing-proposed-updates of package version -1.2. +Version numbers are usually selected by appending ++debXu1, +where X is the major release number and +Y is a counter starting at 1. +e.g. 1:2.4.3-4+deb8u1. Please make sure you didn't miss any of these items in your upload:
Bug#780243: t-p-u: Use the release code name as the target distribution
Package: developers-reference Version: 3.4.14 Severity: normal Tags: patch In a t-p-u (#780169), the release team asked me to use a code name as the upload target distribution, rather than "testing" or "testing-proposed-updates". This patch implements that page in developers-reference. SR diff --git a/pkgs.dbk b/pkgs.dbk index 5a34b95..ca5847f 100644 --- a/pkgs.dbk +++ b/pkgs.dbk @@ -2770,8 +2770,9 @@ Make sure that you included an appropriate explanation in the changelog; -Make sure that you've written testing or -testing-proposed-updates into your target distribution; +Make sure that you've written the testing +code name (e.g. jessie) +into your target distribution;
Bug#780169: jessie-pu: package youtube-dl/2014.08.05-1jessie0.1
Control: tags -1 - moreinfo > Please use 2014.08.05-1+deb8u1 as version number. Lintian doesn't think that's correct for an NMU, but using it anyway. Also, submitted a patch to developers-reference #768426. > and jessie as a distribution (instead of testing). Submitted #780243 (with a patch) to developers-reference. SR -- Stefano Rivera http://tumbleweed.org.za/ +1 415 683 3272 -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Bug#780059: youtube-dl: Forces SSLv3, incompatible with Python 2.7.9
Package: youtube-dl Version: 2014.08.05-1 Severity: grave Tags: patch Justification: renders package unusable Control: fixed -1 2015-01-16-1 Upstream is doing some crazy stuff with SSL. Fortunately, they admit this in their git history, and have improved things since the 2014.08.05 release. The protocol is forced to SSLv3, rather than negotiating the latest protocol supported by both sides. There is a fallback path to negotiation, but it doesn't work when PROTOCOL_SSLv3 isn't available in the Python ssl module (as is the case, since 2.7.8-12). The attached patch should fix the issue. SR Description: Support Python 2.7.9, which removed PROTOCOL_SSLv3 In fact, don't try to force an SSL version at all. Debian OpenSSL doesn't support insecure versions. Upstream use Python's default SSL handshake since https://github.com/rg3/youtube-dl/commit/0db261ba567cb5370455d67c4398e11e5e2119f8 And switches to TLSv1 in legacy paths in https://github.com/rg3/youtube-dl/commit/d79323136fabc2cd72afc7c124e17797e32df514 Author: Stefano Rivera Forwarded: not-needed Last-Update: 2015-03-08 --- a/youtube_dl/utils.py +++ b/youtube_dl/utils.py @@ -588,17 +588,14 @@ if getattr(self, '_tunnel_host', False): self.sock = sock self._tunnel() -try: -self.sock = ssl.wrap_socket(sock, self.key_file, self.cert_file, ssl_version=ssl.PROTOCOL_SSLv3) -except ssl.SSLError: -self.sock = ssl.wrap_socket(sock, self.key_file, self.cert_file, ssl_version=ssl.PROTOCOL_SSLv23) +self.sock = ssl.wrap_socket(sock, self.key_file, self.cert_file, ssl_version=ssl.PROTOCOL_SSLv23) class HTTPSHandlerV3(compat_urllib_request.HTTPSHandler): def https_open(self, req): return self.do_open(HTTPSConnectionV3, req) return HTTPSHandlerV3(**kwargs) else: -context = ssl.SSLContext(ssl.PROTOCOL_SSLv3) +context = ssl.SSLContext(ssl.PROTOCOL_SSLv23) context.verify_mode = (ssl.CERT_NONE if opts_no_check_certificate else ssl.CERT_REQUIRED)
Bug#780059: youtube-dl: Forces SSLv3, incompatible with Python 2.7.9
Hi Debian (2015.03.08_14:32:19_-0700)
> The protocol is forced to SSLv3, rather than negotiating the latest protocol
> supported by both sides. There is a fallback path to negotiation, but it
> doesn't work when PROTOCOL_SSLv3 isn't available in the Python ssl module (as
> is the case, since 2.7.8-12).
I forgot to provide a stack trace:
[youtube] Setting language
Traceback (most recent call last):
File "/usr/bin/youtube-dl", line 9, in
load_entry_point('youtube-dl==2014.08.05', 'console_scripts',
'youtube-dl')()
File "/usr/lib/python2.7/dist-packages/youtube_dl/__init__.py", line 890, in
main
_real_main(argv)
File "/usr/lib/python2.7/dist-packages/youtube_dl/__init__.py", line 880, in
_real_main
retcode = ydl.download(all_urls)
File "/usr/lib/python2.7/dist-packages/youtube_dl/YoutubeDL.py", line 1052,
in download
self.extract_info(url)
File "/usr/lib/python2.7/dist-packages/youtube_dl/YoutubeDL.py", line 516, in
extract_info
ie_result = ie.extract(url)
File "/usr/lib/python2.7/dist-packages/youtube_dl/extractor/common.py", line
169, in extract
self.initialize()
File "/usr/lib/python2.7/dist-packages/youtube_dl/extractor/common.py", line
164, in initialize
self._real_initialize()
File "/usr/lib/python2.7/dist-packages/youtube_dl/extractor/youtube.py", line
123, in _real_initialize
if not self._set_language():
File "/usr/lib/python2.7/dist-packages/youtube_dl/extractor/youtube.py", line
50, in _set_language
fatal=False))
File "/usr/lib/python2.7/dist-packages/youtube_dl/extractor/common.py", line
283, in _download_webpage
res = self._download_webpage_handle(url_or_request, video_id, note,
errnote, fatal)
File "/usr/lib/python2.7/dist-packages/youtube_dl/extractor/common.py", line
223, in _download_webpage_handle
urlh = self._request_webpage(url_or_request, video_id, note, errnote, fatal)
File "/usr/lib/python2.7/dist-packages/youtube_dl/extractor/common.py", line
203, in _request_webpage
return self._downloader.urlopen(url_or_request)
File "/usr/lib/python2.7/dist-packages/youtube_dl/YoutubeDL.py", line 1231,
in urlopen
return self._opener.open(req, timeout=self._socket_timeout)
File "/usr/lib/python2.7/urllib2.py", line 431, in open
response = self._open(req, data)
File "/usr/lib/python2.7/urllib2.py", line 449, in _open
'_open', req)
File "/usr/lib/python2.7/urllib2.py", line 409, in _call_chain
result = func(*args)
File "/usr/lib/python2.7/dist-packages/youtube_dl/utils.py", line 598, in
https_open
return self.do_open(HTTPSConnectionV3, req)
File "/usr/lib/python2.7/urllib2.py", line 1194, in do_open
h.request(req.get_method(), req.get_selector(), req.data, headers)
File "/usr/lib/python2.7/httplib.py", line 1001, in request
self._send_request(method, url, body, headers)
File "/usr/lib/python2.7/httplib.py", line 1035, in _send_request
self.endheaders(body)
File "/usr/lib/python2.7/httplib.py", line 997, in endheaders
self._send_output(message_body)
File "/usr/lib/python2.7/httplib.py", line 850, in _send_output
self.send(msg)
File "/usr/lib/python2.7/httplib.py", line 812, in send
self.connect()
File "/usr/lib/python2.7/dist-packages/youtube_dl/utils.py", line 592, in
connect
self.sock = ssl.wrap_socket(sock, self.key_file, self.cert_file,
ssl_version=ssl.PROTOCOL_SSLv3)
AttributeError: 'module' object has no attribute 'PROTOCOL_SSLv3'
SR
--
Stefano Rivera
http://tumbleweed.org.za/
+1 415 683 3272
--
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Bug#803689: pypy: DEFAULT_SOABI at pypy-26 while DEFAULT_MAGIC_TAG = 'pypy-40'
Package: pypy,dh-python Severity: serious dh_pypy assumes that the version in the soabi tag refers to the major and minor version of PyPy, like it does in cPython. This used to be true, until https://bitbucket.org/pypy/pypy/commits/97d59353f34e I didn't realise this until PyPy 4.0 was uploaded, but we now have a situation where the tag on .pyc files is different to the one on .so files. And dh_pypy generates incorrect dependencies for packages containing .so files. We have some options here: 1. Revert this upstream patch, and continue to transition PyPy for each major release. 2. Make the same change to DEFAULT_MAGIC_TAG. As far as I can tell, the bytecode hasn't changed in ages. Convert dh_pypy to depend on pypy-26 (i.e. the tag) rather than pypy (<< 2.7) Thoughts? SR
Bug#803689: Acknowledgement (pypy: DEFAULT_SOABI at pypy-26 while DEFAULT_MAGIC_TAG = 'pypy-40')
an pypy using .pypy-26.so (well .pypy-26-x86_64-linux-gnu.so or similar) and do .pypy-26.pyc too 07:26 < tumbleweed> then revisit this when we need to 07:27 < arigato> if you have some suggestion for upstream, they are welcome 07:27 < arigato> I think that cffi can be kept backward- and forward-compatible 07:27 < tumbleweed> so, it seems the only thing right now, is to expose SOABI in sysconfig 07:28 < arigato> .pyc too 07:28 < arigato> but maybe cpyext will see big changes 07:29 < tumbleweed> I've tried to incorporate cffi's version ranges into Debian package dependencies, but can't quite yet (waiting for a feature in dpkg) 07:31 < tumbleweed> so I can't use the min and max versions yet, but I can just use VERSION (a strict dependency) and that works quite neatly 07:31 < tumbleweed> we could expose that version in the filename, but then I guess importing searching gets painful 07:34 < tumbleweed> and now, bed for me :) SR -- Stefano Rivera http://tumbleweed.org.za/ +1 415 683 3272
Bug#804350: ITP: vizzini -- Kernel driver for Exar XR21V1414 USB UART
Package: wnpp Severity: wishlist Owner: Stefano Rivera * Package name: vizzini Version : 1.0.0 Upstream Author : Exar Corporation, Inc. * URL : https://github.com/mithro/exar-uart-driver * License : GPL-2+ Programming Lang: C Description : Kernel driver for Exar XR21V1414 USB UART The debconf-video team is using a board with this UART on it for HDMI capture.
Bug#804350: ITP: vizzini -- Kernel driver for Exar XR21V1414 USB UART
Hi Ben (2015.11.07_19:33:52_+0200) > What's blocking this from going into mainline? Apparently: 17:49 <@mithro> tumbleweed: but what happened is that the exar guys effectively forked an in tree kernel driver and made it work with their crappy device 17:50 <@mithro> tumbleweed: shenki then fixed it to work with modern kernels 17:51 <@mithro> tumbleweed: It needs to be rewritten as something that can be merged into the upstream driver. 17:51 <@mithro> (The CDC-ACM driver) 17:52 <@mithro> That is never going to happen as nobody has the time or will power to do it Less than ideal for the archive, but useful if you want to muck around with this hardware. SR -- Stefano Rivera http://tumbleweed.org.za/ +1 415 683 3272
Bug#804519: RM: gstreamer0.10-dvswitch -- ROM; Replaced by gst-plugins-dvswitch
Package: ftp.debian.org Severity: normal Replaced by gst-plugins-dvswitch, which is now maintained by me. See #798130, #785835. Thanks, SR
Bug#771148: (pre-upload) unblock: pypy/2.4.0+dfsg-2
Control: tags -1 - moreinfo > As you noted on irc, it still fails after 3 attempts, so a new upload is > probably necessary. Please remove the moreinfo tag once the mipsel issue is > resolved. Another (pair of) give-backs resolved it. SR -- Stefano Rivera http://tumbleweed.org.za/ +1 415 683 3272 -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Bug#802229: python-cffi: Uses versioned Provides
Hi Dmitry (2015.10.18_08:54:33_-0700) > Packages python-cffi-backend and python3-cffi-backend currently have > versioned Provides field: This has been supported in dpkg since 1.17.11. However, we can't safely use it until #786803 is resolved. > However, the Policy § 7.5 (Virtual packages - Provides) [1] says: See #761219 SR -- Stefano Rivera http://tumbleweed.org.za/ +1 415 683 3272
Bug#746946: wheezy-pu: package distro-info-data/0.23~deb7u1
Hi 746946 (2014.11.01_03:02:58_+0200) > We still need to make a decision on this bit. I've left that patch in, > for now. Ping? SR -- Stefano Rivera http://tumbleweed.org.za/ H: +27 21 461 1230 C: +27 72 419 8559 -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Bug#774678: ykneomgr: Missing dependency on pcscd
Package: ykneomgr Version: 0.1.6-1 Severity: serious Justification: Policy 7.2 Without pcscd, ykneomgr displays a fairly unhelpful error: $ ykneomgr -m error: ykneomgr_init (-4): Backend error I think ykneomgr should Depend on pcscd. SR -- System Information: Debian Release: 8.0 APT prefers testing APT policy: (900, 'testing'), (800, 'unstable'), (100, 'experimental') Architecture: amd64 (x86_64) Foreign Architectures: i386 Kernel: Linux 3.16.0-4-amd64 (SMP w/4 CPU cores) Locale: LANG=en_ZA.utf8, LC_CTYPE=en_ZA.utf8 (charmap=UTF-8) Shell: /bin/sh linked to /bin/dash Init: systemd (via /run/systemd/system) Versions of packages ykneomgr depends on: ii libc6 2.19-13 ii libykneomgr0 0.1.6-1 ykneomgr recommends no packages. ykneomgr suggests no packages. -- no debconf information signature.asc Description: Digital signature
Bug#855548: hdmi2usb-fx2-firmware: --mode serial doesn't work
Package: hdmi2usb-fx2-firmware Version: 0.0.0~git20151018-1 Severity: grave Justification: renders package unusable Control: affects -1 hdmi2usb-mode-switch hdmi2usb-mode-switch usually needs to put an opsis into serial mode, before it'll successfully flash an image. Currently, the uart image included in hdmi2usb-fx2-firmware doesn't seem to work. It looks like this was fixed upstream, but we packaged the wrong branch. SR
Bug#855555: unblock: hdmi2usb-fx2-firmware/0.0.0~git20151225-1
..5d5960c 100644 --- a/.travis.yml +++ b/.travis.yml @@ -10,6 +10,12 @@ install: - # Install sdcc - sudo apt-get install --force-yes -y sdcc - sdcc --version + - # doxygen & rubber are needed for generating the documentation + - sudo apt-get install -y doxygen rubber script: - make + - make docs + +after_success: + - ./.travis-push-docs.sh diff --git a/debian/changelog b/debian/changelog index 3541a3a..82797f3 100644 --- a/debian/changelog +++ b/debian/changelog @@ -1,4 +1,12 @@ -hdmi2usb-fx2-firmware (0.0.0~git20151018-1) unstable; urgency=low +hdmi2usb-fx2-firmware (0.0.0~git20151225-1) UNRELEASED; urgency=low + + * New upstream release (different git branch) +- Should actually build a working uart firmware for the opsis. + (Closes: #855548) + + -- Stefano Rivera Mon, 28 Nov 2016 23:35:19 -0800 + +hdmi2usb-fx2-firmware (0.0.0~git20151128-1) unstable; urgency=low * Initial upload. (Closes: #796769) diff --git a/debian/rules b/debian/rules index c4f2158..152525d 100755 --- a/debian/rules +++ b/debian/rules @@ -4,12 +4,13 @@ dh $@ VER=$(shell dpkg-parsechangelog | sed -rne 's/^Version: (.+)-.*/\1/p') +BRANCH=opsis-uart-with-eeprom-serialno get-packaged-orig-source: - git clone https://github.com/mithro/fx2lib -b cdc-usb-serialno-from-eeprom + git clone https://github.com/mithro/fx2lib -b $(BRANCH) set -xe; \ GIT_DATE=$$(dpkg-parsechangelog | sed -rne 's/^Version: .*\~git()(..)(..)-.*/\1-\2-\3 00:00:00 UTC/p'); \ cd fx2lib; \ - GIT_COMMIT=$$(git rev-list -n1 --until="$$GIT_DATE" cdc-usb-serialno-from-eeprom); \ + GIT_COMMIT=$$(git rev-list -n1 --until="$$GIT_DATE" $(BRANCH)); \ git archive $$GIT_COMMIT --prefix=hdmi2usb-fx2-firmware_$(VER).orig/ \ -o ../hdmi2usb-fx2-firmware_$(VER).orig.tar xz -f hdmi2usb-fx2-firmware_$(VER).orig.tar diff --git a/examples/cdc/Makefile b/examples/cdc/Makefile index 57cb825..e9b579c 100644 --- a/examples/cdc/Makefile +++ b/examples/cdc/Makefile @@ -1,4 +1,4 @@ -DIRS=to-uart +DIRS=loopback to-uart .PHONY: dirs $(DIRS) clean diff --git a/examples/cdc/common/dscr.a51 b/examples/cdc/common/dscr.a51 index 285d9f9..533d5ec 100644 --- a/examples/cdc/common/dscr.a51 +++ b/examples/cdc/common/dscr.a51 @@ -42,7 +42,8 @@ ENDPOINT_TYPE_ISO=1 ENDPOINT_TYPE_BULK=2 ENDPOINT_TYPE_INT=3 -.globl _dev_dscr, _dev_qual_dscr, _highspd_dscr, _fullspd_dscr, _dev_strings, _dev_strings_end, _dev_serial +.globl _dev_dscr, _dev_qual_dscr, _highspd_dscr, _fullspd_dscr, _dev_strings, _dev_strings_end +.globl _dev_serial ; These need to be in code memory. If ; they aren't you'll have to manully copy them somewhere ; in code memory otherwise SUDPTRH:L don't work right @@ -57,9 +58,9 @@ _dev_dscr: .db 0x00 ; 5 bDeviceSubclass 1 Subclass code .db 0x00 ; 6 bDeviceProtocol 1 Protocol Code .db 64; 7 bMaxPacketSize0 1 Maximum packet size for endpoint zero - .dw 0xB404; 8 idVendor 2 Vendor ID - .dw 0x0410; 10 idProduct 2 Product ID - .dw 0x0100; 12 bcdDevice 2 Device release number (BCD) + .dw 0x192A; 8 idVendor 2 Vendor ID + .dw 0x4154; 10 idProduct 2 Product ID + .dw 0x0300; 12 bcdDevice 2 Device release number (BCD) .db 1 ; 14 iManufacturer 1 Index of string descriptor for the manufacturer .db 2 ; 15 iProduct 1 Index of string descriptor for the product .db 3 ; 16 iSerialNumber 1 Index of string descriptor for the serial number @@ -107,7 +108,7 @@ highspd_dscr_end: .db 0x02 ; Interface class .db 0x02 ; Interface sub class .db 0x01 ; Interface protocol code class - .db 0x00 ; Interface descriptor string index + .db 0; Interface descriptor string index ;; CDC Header Functional Descriptor .db 0x05 ; Descriptor Size in Bytes (5) @@ -154,7 +155,7 @@ highspd_dscr_end: .db 0x0A ; Interface class .db 0x00 ; Interface sub class .db 0x00 ; Interface protocol code class - .db 0x00 ; Interface descriptor string index + .db 0; Interface descriptor string index ; endpoint 2 out .db DSCR_ENDPOINT_LEN; Descriptor length @@ -195,15 +196,15 @@ fullspd_dscr_end: ; NOTE the default TRM actually has more alt interfaces ; but you can add them back in if you need them. ; here, we just use the default alt setting 1 from the trm - .db DSCR_INTERFACE_LEN - .db DSCR_INTERFACE_TYPE - .db 0 ; index - .db 0 ; alt setting idx - .db 2 ; n endpoints - .db 0x2 ; class - .db 0x2 - .db 0x1 - .db 3
Bug#855555: unblock: hdmi2usb-fx2-firmware/0.0.0~git20151225-1
Control: tags -1 - moreinfo > How soon can we have confirmed whether this upload fixes the issue with > Numato Opsis boards? If we unblock this, I would like to know it at > least fixes the issue we are unblocking it for. It works. I confirmed this yesterday, and with the package, as built in the archive, this morning. Thanks CarlFK for hooking up an Opsis for me :) SR -- Stefano Rivera http://tumbleweed.org.za/ +1 415 683 3272
Bug#846185: ITP: ixo-usb-jtag -- Firmware for USB JTAG programmers
Package: wnpp Severity: wishlist Owner: Stefano Rivera * Package name: ixo-usb-jtag Version : 0.0.0+git20160908 Upstream Author : Tim 'mithro' Ansell * URL : https://github.com/mithro/ixo-usb-jtag * License : GPL-2+ Programming Lang: C Description : Firmware for USB JTAG programmers This firmware allows a USB-capable microcontroller to act like an Altera USB-Blaster JTAG pod. Which in turn may allow you to use tools you'd normally use with the Altera USB-Blaster, including UrJTAG and openocd. Supported hardware: The Cypress FX2 EZ-USB family, or an FTDI FT245 in combination with a CPLD. Builds are included for the hdmi2usb project's boards (Digilet Atlys and Numato Opsis).
Bug#796769: retitle to ITP: hdmi2usb-fx2-firmware -- f/w for micro-controller on HDMI2USB hardware
Control: retitle -1 ITP: hdmi2usb-fx2-firmware -- FX2 firmware for hdmi2usb board development Control: owner -1 ! This package contains the FX2 firmware for several modes of the Numato Opsis board's USB interface. It is used for flashing updates to the board. SR -- Stefano Rivera http://tumbleweed.org.za/ +1 415 683 3272
Bug#850661: Provides line defines versioned virtual packages
Hi martin (2017.01.09_02:36:38_+0200) > Provides defines virtual packages, and those do not (and cannot as > it wouldn't make sense) carry version numbers. They have been supported in dpkg since 1.17.11, and other tools since. Pypy is not the only package to use them. And it waited until they were fairly well supported in archive infrastructure (britney, etc.) before starting to use them. > § 7.5 of the policy says: > > A Provides field may not contain version numbers Pretty sure this is a case of policy lagging behind reality (which happens more often than we'd like :( ). Here's the dpkg maintainer agreeing with me, in an aptitude bug: https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=801216#26 SR -- Stefano Rivera http://tumbleweed.org.za/ +1 415 683 3272
Bug#785559: python-pypy.sandbox: --timeout option to pypy-sandbox is ignored.
Hi Valentin (2015.05.17_12:35:25_-0700) > /usr/bin/pypy-sandbox has an option to kill sandboxed scripts after a > given timeout. > However, this option seems to be ignored. This looks like a packaging error. Sorry. There's a special marshal library that the sandbox needs to use. SR -- Stefano Rivera http://tumbleweed.org.za/ +1 415 683 3272 -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Bug#785554: python-pypy.sandbox: Depends on non-existing module ansiramp
Hi Valentin (2015.05.17_11:19:22_-0700) > With some values of $TERM (when it contains the string “256”), script > /usr/lib/python2.7/dist-packages/pypy/sandbox/rpython/tool/ansi_mandelbrot.py > imports module ansiramp, which is not available in any Debian package. Oops. That does exist in the source, but needs to be bundled in with the sandbox. This breaks --verbose. SR -- Stefano Rivera http://tumbleweed.org.za/ +1 415 683 3272 -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Bug#786738: pypy: Segmentation fault when printing unpickled generator
Control: -1 tags + upstream Control: -1 forwarded https://bitbucket.org/pypy/pypy/issue/2075/pypy-segmentation-fault-when-printing Hi Dan (2015.05.24_20:37:09_-0700) > The following code causes pypy to segfault. Thanks for the report. This is an upstream bug, and I've forwarded it to PyPy's bugtracker. SR -- Stefano Rivera http://tumbleweed.org.za/ +1 415 683 3272 -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Bug#786738: pypy: Segmentation fault when printing unpickled generator
Control: tags -1 + patch And, fixed upstream: https://bitbucket.org/pypy/pypy/commits/a357ba14fe22 SR -- Stefano Rivera http://tumbleweed.org.za/ +1 415 683 3272 -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Bug#735855: python-setuptools: Package setuptools for pypy
Control: tag -1 + patch
Hi Matthias (2015.02.26_10:20:31_-0800)
> please could you provide a patch?
I've had a shot at this.
It was non-trivial, as python-setuptools isn't in the best shape.
There's a little brokenness in the source package, a lot of duplicated
code, and some dead code.
I get the feeling that it originally tried to be entirely
self-supported, with no external dependencies. But over time, it has
gained dependencies on dh_pythonX, without getting a whole lot of
clean-up.
I don't know what the purpose of being entirely self-supported would be.
This is an arch-indep package, that isn't involved in arch
bootstrapping. It's a build-dependency for *many* Python packages, but
otherwise I don't think it's particularly special.
Another source of complication are the version-suffixed easy_install
binaries. I think we can drop those, without anyone caring. This is a
route we've been taking in other packages, with no fallout I've seen, so
far. codesearch.debian.net shows up no direct use of them. And I'm sure
the vast majority of users use pip rather than easy_install.
So, I have two options attached.
minimal.debdiff:
debian/changelog |7 +++
debian/control | 26 +-
debian/rules | 43 ---
3 files changed, 72 insertions(+), 4 deletions(-)
A minimal patch that adds a set of pypy binaries. This increases code
duplication, as /usr/share/python3/python.mk doesn't know anything about
pypy.
python-setuptools_17.0-1+pybuild.debdiff:
debian/README.Debian |6 -
debian/changelog | 20 +++
debian/compat|2
debian/control | 45 +++
debian/patches/fix-python3.5.diff|2
debian/patches/install-layout.diff | 20 ---
debian/python-pkg-resources.install |1
debian/python-setuptools.install |4
debian/python3-pkg-resources.install |1
debian/python3-setuptools.install|4
debian/rules | 207 ---
11 files changed, 82 insertions(+), 230 deletions(-)
python-setuptools_17.0-1+pybuild+pypy.debdiff:
debian/changelog |6 ++
debian/control| 23 +++
debian/pypy-pkg-resources.install |1 +
debian/pypy-setuptools.install|4
debian/rules | 19 ++-
5 files changed, 48 insertions(+), 5 deletions(-)
Two patches that port setuptools to pybulid, and then (trivially) add
pypy binary packages. There is a large amount of unrelated cleanup in
the pybuild patch, because the alternative would be to re-implement what
I think is unnecessary cruft.
Take your pick :)
SR
--
Stefano Rivera
http://tumbleweed.org.za/
+1 415 683 3272
diff -Nru python-setuptools-17.0/debian/changelog
python-setuptools-17.0/debian/changelog
--- python-setuptools-17.0/debian/changelog 2015-05-29 06:14:38.0
-0700
+++ python-setuptools-17.0/debian/changelog 2015-07-05 20:23:16.0
-0700
@@ -1,3 +1,10 @@
+python-setuptools (17.0-1+pypy) UNRELEASED; urgency=medium
+
+ * Add binary packages for pypy.
+- Build-Depend on dh-python and pypy.
+
+ -- Stefano Rivera Sun, 05 Jul 2015 18:45:21 -0700
+
python-setuptools (17.0-1) unstable; urgency=medium
* New upstream version.
diff -Nru python-setuptools-17.0/debian/control
python-setuptools-17.0/debian/control
--- python-setuptools-17.0/debian/control 2015-05-19 06:37:18.0
-0700
+++ python-setuptools-17.0/debian/control 2015-07-05 18:46:52.0
-0700
@@ -5,7 +5,9 @@
Build-Depends-Indep: python-all-dev (>= 2.6.6-2~), python3-all-dev (>=
3.1.2-8~),
python3.5-dev,
python3-sphinx,
- python3-wheel
+ python3-wheel,
+ dh-python,
+ pypy
Build-Depends: debhelper (>= 5.0.37.1), python3
Standards-Version: 3.9.6
Homepage: https://pypi.python.org/pypi/setuptools
@@ -68,6 +70,28 @@
Description: Python3 Distutils Enhancements
Extensions to the python-distutils for large or complex distributions.
+Package: pypy-pkg-resources
+Suggests: pypy-setuptools
+Architecture: all
+Depends: ${misc:Depends}, ${pypy:Depends}
+Description: Package Discovery and Resource Access using pkg_resources
+ The pkg_resources module provides an API for Python libraries to
+ access their resource files, and for extensible applications and
+ frameworks to automatically discover plugins. It also provides
+ runtime support for using C extensions that are inside zipfile-format
+ eggs, support for merging packages that have separately-distributed
+ modules or subpackages, and APIs for managing Python's current
+ "working set" of active packages.
+
+Package: pypy-setuptools
+Architecture: all
+Depends:
+ pypy-pkg-resources (= ${binary:Version}),
+ ${misc:Depends},
+ ${pypy:Depends}
+Des
Bug#849643: magic-wormhole: unnecessary Build-Depends on python3-cffi
Package: magic-wormhole Version: 0.8.1-2 Severity: normal Tags: patch I'm pretty sure that you added a Build-Dependency on python3-cffi in 47343f28a9f95ec7720e81bc13b2b52eaa4f8e43 to work around a bug in the python3-nacl package's dependencies (#801786). That has now been fixed, so I'd recommend dropping the unused dependency. Patch attached, thanks :) SR diff --git a/debian/control b/debian/control index e6c014b..f509200 100644 --- a/debian/control +++ b/debian/control @@ -8,7 +8,6 @@ Build-Depends: dh-python, python3, python3-autobahn (>= 0.14.1), - python3-cffi, python3-click, python3-hkdf, python3-humanize,
Bug#849648: mitmproxy: Unnecessary Build-Depends on python-cffi (and broken Vcs-Git field)
Package: mitmproxy Version: 0.18.1-1 Severity: normal Tags: patch Upstream dropped their last cffi module (certffi) in 2723a0e5739412953f60c37d0dab81d684ba5f26 That means that the Build-Depends on python-cffi is unnecessary. Additionally, when attempting to submit a patch, I noticed that the Vcs-Git field is incorrect. Patches attached. SR From 93ec45a04325aaf86169bfe3f95e5707f3fcc181 Mon Sep 17 00:00:00 2001 From: Stefano Rivera Date: Thu, 29 Dec 2016 15:55:38 +0100 Subject: [PATCH 1/3] Drop python-cffi Build-Depends, unnecessary --- debian/control | 1 - 1 file changed, 1 deletion(-) diff --git a/debian/control b/debian/control index a7d18e2d..569eba3b 100644 --- a/debian/control +++ b/debian/control @@ -13,7 +13,6 @@ Build-Depends: debhelper (>= 9), dh-python, libssl-dev, python-dev, python-setuptools, python-urwid (>= 1.3.1), python-backports.ssl-match-hostname (>= 3.5.0.1), - python-cffi, python-configargparse (>= 0.10), python-flask (>= 0.10.1), python-html2text (>= 2016.1.8), -- 2.11.0 From e6de4b41b8521f3a2f8da524bd5448ba1ce4db50 Mon Sep 17 00:00:00 2001 From: Stefano Rivera Date: Thu, 29 Dec 2016 15:57:22 +0100 Subject: [PATCH 2/3] Fix Vcs-Git field --- debian/control | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/debian/control b/debian/control index 569eba3b..04c0ac54 100644 --- a/debian/control +++ b/debian/control @@ -38,7 +38,7 @@ Build-Depends: debhelper (>= 9), dh-python, libssl-dev, python-dev, python-jsbeautifier (>= 1.6.3), python-tz Standards-Version: 3.9.8 -Vcs-Git: https://anonscm.debian.org/collab-maint/mitmproxy.git +Vcs-Git: https://anonscm.debian.org/git/collab-maint/mitmproxy.git Vcs-Browser: https://anonscm.debian.org/git/collab-maint/mitmproxy.git Homepage: https://mitmproxy.org -- 2.11.0
Bug#849744: src:python-pygit2: Unnecessary Depends on python{3,}-cffi
Package: src:python-pygit2
Severity: normal
Tags: patch
python-pygit2 uses the out-of-line mode for driving cffi, which means
that only the cffi backend is required at runtime, not support for
building the cffi modules.
This means that the Depends on python{3,}-cffi can be dropped.
SR
From 971246798914223a745214ce65a02564ea96db56 Mon Sep 17 00:00:00 2001
From: Stefano Rivera
Date: Fri, 30 Dec 2016 12:00:02 +0100
Subject: [PATCH] Drop unnecessary Depends on python-cffi
---
debian/control | 6 ++
1 file changed, 2 insertions(+), 4 deletions(-)
diff --git a/debian/control b/debian/control
index eacd5a9..4e0a3db 100644
--- a/debian/control
+++ b/debian/control
@@ -26,8 +26,7 @@ Homepage: https://github.com/libgit2/pygit2
Package: python-pygit2
Architecture: any
-Depends: python-cffi (>= 0.9.2),
- ${misc:Depends},
+Depends: ${misc:Depends},
${python:Depends},
${shlibs:Depends},
Suggests: python-pygit2-doc,
@@ -40,8 +39,7 @@ Description: bindings for libgit2 - Python 2.x
Package: python3-pygit2
Architecture: any
-Depends: python3-cffi (>= 0.9.2),
- ${misc:Depends},
+Depends: ${misc:Depends},
${python3:Depends},
${shlibs:Depends},
Suggests: python-pygit2-doc,
--
2.11.0
Bug#849745: src:python-jack-client: Unnecessary (and possibly incorrect) cffi dependencies
Package: src:python-jack-client
Version: 0.4.2-1
Severity: normal
Tags: patch
Hi,
This package uses one of the in-line cffi modes [0]. This means it needs
to depend on python-cffi, not just the backend. The binaries currently
in the archive do this, but they wouldn't if they'd been built in the
presence of python-cffi (because it has a pydist override that generates
a dependency on the backend package).
Also, they build-depend on the backend which is a bit weird, and not
doing anything useful.
[0]:
https://cffi.readthedocs.io/en/latest/overview.html#simple-example-abi-level-in-line
So, here's a patch series to tidy this all up. And fix the broken clean rule.
SR
From 3a7092cac7ee34b651140cd0e5f7238bb71cdcc0 Mon Sep 17 00:00:00 2001
From: Stefano Rivera
Date: Fri, 30 Dec 2016 12:30:28 +0100
Subject: [PATCH 1/3] Drop python{3,}-cffi-backend Build-Deps - there is no
test suite
---
debian/control | 4 +---
1 file changed, 1 insertion(+), 3 deletions(-)
diff --git a/debian/control b/debian/control
index 271bcf2..40ca5cf 100644
--- a/debian/control
+++ b/debian/control
@@ -7,9 +7,7 @@ Build-Depends: debhelper (>=9),
python-all (>= 2.6.6-3~),
python-setuptools,
python3-all,
- python-cffi-backend (>= 1.6.0),
- python3-setuptools,
- python3-cffi-backend (>= 1.6.0)
+ python3-setuptools
Standards-Version: 3.9.8
Homepage: http://jackclient-python.rtfd.org
X-Python-Version: >= 2.7
--
2.11.0
From 7b2c1ca6a3e2aee95b1834bc2a569141b34ebfac Mon Sep 17 00:00:00 2001
From: Stefano Rivera
Date: Fri, 30 Dec 2016 12:31:53 +0100
Subject: [PATCH 2/3] This package uses inline cffi mode
So explicitly depend on python{3,}-cffi. The backend alone is not
sufficient.
---
JACK_Client.egg-info/SOURCES.txt | 1 +
debian/control | 4 ++--
2 files changed, 3 insertions(+), 2 deletions(-)
diff --git a/JACK_Client.egg-info/SOURCES.txt b/JACK_Client.egg-info/SOURCES.txt
index de9ad60..6387d3b 100644
--- a/JACK_Client.egg-info/SOURCES.txt
+++ b/JACK_Client.egg-info/SOURCES.txt
@@ -4,6 +4,7 @@ MANIFEST.in
NEWS.rst
README.rst
jack.py
+setup.cfg
setup.py
JACK_Client.egg-info/PKG-INFO
JACK_Client.egg-info/SOURCES.txt
diff --git a/debian/control b/debian/control
index 40ca5cf..e47d46d 100644
--- a/debian/control
+++ b/debian/control
@@ -18,8 +18,8 @@ Vcs-Browser: https://anonscm.debian.org/cgit/collab-maint/python-jack-client.git
Package: python-jack-client
Architecture: all
Depends: ${python:Depends},
- python-cffi-backend (>= 1.6.0),
${misc:Depends},
+ python-cffi,
libjack-jackd2-0 (>= 1.9.10+20150825) | libjack-0.125
Suggests: python-numpy
Description: JACK Audio Connection Kit (JACK) Client for Python
@@ -33,7 +33,7 @@ Package: python3-jack-client
Architecture: all
Depends: ${python3:Depends},
${misc:Depends},
- python3-cffi-backend (>= 1.6.0),
+ python3-cffi,
libjack-jackd2-0 (>= 1.9.10+20150825) | libjack-0.125
Suggests: python3-numpy
Description: JACK Audio Connection Kit (JACK) Client for Python 3
--
2.11.0
From 3fcab5fdd5655940471c52e83d2c45d06f9f8421 Mon Sep 17 00:00:00 2001
From: Stefano Rivera
Date: Fri, 30 Dec 2016 12:35:08 +0100
Subject: [PATCH 3/3] Clean up egg-info
---
debian/clean | 1 +
1 file changed, 1 insertion(+)
create mode 100644 debian/clean
diff --git a/debian/clean b/debian/clean
new file mode 100644
index 000..45149aa
--- /dev/null
+++ b/debian/clean
@@ -0,0 +1 @@
+*.egg-info/*
--
2.11.0
Bug#849746: src:python-xattr: Unnecessary python{3,}-cffi Depends
Package: src:python-xattr
Version: 0.9.1-1
Severity: normal
Tags: patch
Now that the upstream is using an out-of-line cffi mode [0], it's no
longer necessary to Depend on python{3,}-cffi, only the backend is
needed.
So, we can revert the patch that fixed #814650.
[0]:
https://cffi.readthedocs.io/en/latest/overview.html#purely-for-performance-api-level-out-of-line
Patch attached.
SR
From 55f4ad7835ff36071bd6b16ea5f53d0d350cd919 Mon Sep 17 00:00:00 2001
From: Stefano Rivera
Date: Fri, 30 Dec 2016 12:44:58 +0100
Subject: [PATCH] Drop unnecessary python{3,}-cffi Depends
---
debian/control | 2 --
1 file changed, 2 deletions(-)
diff --git a/debian/control b/debian/control
index b461cae..bf8334a 100644
--- a/debian/control
+++ b/debian/control
@@ -41,7 +41,6 @@ Depends:
${misc:Depends},
${python:Depends},
${shlibs:Depends},
- python-cffi,
Conflicts: python-pyxattr
Provides: ${python:Provides}
, python-pyxattr
@@ -59,7 +58,6 @@ Depends:
${misc:Depends},
${python3:Depends},
${shlibs:Depends},
- python3-cffi,
Conflicts: python3-pyxattr
Provides: ${python3:Provides}
, python3-pyxattr
--
2.11.0
Bug#698413: usb-modeswitch-data: Huawei e3276 no longer switched
Package: usb-modeswitch-data Version: 20121109-1 Severity: normal With 20120815-2, the Huawei e3276 is successfully switched and provides a /dev/ttyUSB0. With 20121109-1, it doesn't. $ lsusb | grep Huawei Bus 001 Device 054: ID 12d1:14fe Huawei Technologies Co., Ltd. SR -- System Information: Debian Release: 7.0 APT prefers testing APT policy: (900, 'testing'), (800, 'unstable'), (1, 'experimental') Architecture: amd64 (x86_64) Foreign Architectures: i386 Kernel: Linux 3.7-trunk-amd64 (SMP w/4 CPU cores) Locale: LANG=en_ZA.UTF-8, LC_CTYPE=en_ZA.UTF-8 (charmap=UTF-8) Shell: /bin/sh linked to /bin/dash Versions of packages usb-modeswitch-data depends on: ii dpkg 1.16.9 Versions of packages usb-modeswitch-data recommends: ii udev175-7 ii usb-modeswitch 1.2.3+repack0-1 usb-modeswitch-data 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#698413: usb-modeswitch-data: Huawei e3276 no longer switched
Hi Didier (2013.01.18_11:01:29_+0200) > That's very weird as 20120815-2 and 20121109-1 have the exact same > configuration file for 12d1:14fe : Yeah, I found it fairly confusing too. > So: are you sure these are the only packages you changed ? Did you also > upgrade usb-modeswitch itself? Yes, tried all the combinations, it seems to be the new data package that causes the trouble, but it's hard to figure out why. > Does this still happen if you launch "# udevadm trigger" between > installations > of usb-modeswitch-data? Yeah, doesn't help. How do I go about debugging usb_modeswitch_dispatcher? SR -- Stefano Rivera http://tumbleweed.org.za/ H: +27 21 461 1230 C: +27 72 419 8559 -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Bug#698413: usb-modeswitch-data: Huawei e3276 no longer switched
Hi Josua (2013.01.18_11:37:34_+0200) > Stefano, can you retrieve a mode-switching debug log? > See /etc/usb_modeswitch.conf for the setting and some hints. Aha. With 20121109-1 I get /var/log/usb_modeswitch_1-1.2 containing: | USB_ModeSwitch log from Sat Jan 19 11:25:40 SAST 2013 | | Raw args from udev: /1-1.2 | | Using global config file: /etc/usb_modeswitch.conf | Could not determine top device dir from udev values! Exiting With 20120815-2 I get /var/log/usb_modeswitch_1-1.2:1.0 that's far longer and happier. It begins: | USB_ModeSwitch log from Sat Jan 19 11:27:07 SAST 2013 | | Raw args from udev: /1-1.2:1.0 | | Using global config file: /etc/usb_modeswitch.conf | Using top device dir /sys/bus/usb/devices/1-1.2 | | USB values from sysfs: | manufacturerHUAWEI Technology | product HUAWEI Mobile | serial | ... SR -- Stefano Rivera http://tumbleweed.org.za/ H: +27 21 461 1230 C: +27 72 419 8559 -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Bug#698413: usb-modeswitch-data: Huawei e3276 no longer switched
Hi Josua (2013.01.19_13:15:14_+0200) > You need usb-modeswitch 1.2.5 for the new udev rule flavour of data 20121109. > > I think I failed to stress that requirement - sorry! > The head information in the rules file is obviously incorrect. Should read: > > # Use with USB_ModeSwitch >= 1.2.5 ! And the Depends/Breaks should be adjusted appropriately. Hrm, yes, now I find that that usb-modeswitch 1.2.5+repack0-1 works with the new data package. I'm sure I couldn't get it to work on Friday (which was what lead to me filing this bug). Oh well, all's well that ends well... SR -- Stefano Rivera http://tumbleweed.org.za/ H: +27 21 461 1230 C: +27 72 419 8559 -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Bug#698413: usb-modeswitch-data: Huawei e3276 no longer switched
Control: reopen -1 > I'm sure I couldn't get it to work on Friday (which was what lead to > me filing this bug). Aha. Reproduced it. The problem seems to be with the USB hub on my desk at work. When I plug the modem into it, I get /var/log/usb_modeswitch_1-1.1.1.4: | USB_ModeSwitch log from Thu Jan 24 17:40:13 SAST 2013 | | Raw args from udev: /1-1.1.1.4 | | Using global config file: /etc/usb_modeswitch.conf | Called by new rules file - remember to check class of first interface ... | Using top device dir /sys/bus/usb/devices/1-1.1 |Warning: USB attribute "manufacturer" not found |Warning: USB attribute "product" not found | Check class of first interface ... | Device is not in install mode. Exiting And this in the kernel log: [426027.648903] usb 1-1.1.1.4: new high-speed USB device number 113 using ehci_hcd [426027.762013] usb 1-1.1.1.4: New USB device found, idVendor=12d1, idProduct=14fe [426027.762017] usb 1-1.1.1.4: New USB device strings: Mfr=2, Product=1, SerialNumber=0 [426027.762019] usb 1-1.1.1.4: Product: HUAWEI Mobile [426027.762021] usb 1-1.1.1.4: Manufacturer: HUAWEI Technology [426027.802104] scsi118 : usb-storage 1-1.1.1.4:1.0 [426027.802267] scsi119 : usb-storage 1-1.1.1.4:1.1 [426028.800119] scsi 118:0:0:0: CD-ROMHUAWEI Mass Storage 2.31 PQ: 0 ANSI: 2 [426028.800132] scsi 119:0:0:0: Direct-Access HUAWEI TF CARD Storage 2.31 PQ: 0 ANSI: 2 [426028.800875] sd 119:0:0:0: Attached scsi generic sg4 type 0 [426028.803620] sr1: scsi-1 drive [426028.803780] sr 118:0:0:0: Attached scsi CD-ROM sr1 [426028.803907] sr 118:0:0:0: Attached scsi generic sg5 type 5 [426028.804112] sd 119:0:0:0: [sdd] Attached SCSI removable disk [426028.975051] ISO 9660 Extensions: Microsoft Joliet Level 1 [426028.977946] ISOFS: changing to secondary root SR -- Stefano Rivera http://tumbleweed.org.za/ H: +27 21 465 6908 C: +27 72 419 8559 UCT: x3127 -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Bug#700540: python-launchpadlib: corrupts the cache occasionally
Control: clone -1 -2 Control: reassign -2 lazr.restfulclient There were some upstream lazr.restfulclient commits to improve concurrency. Paul reports that this improved things (but did not resolve the issue entirely). SR -- Stefano Rivera http://tumbleweed.org.za/ H: +27 21 465 6908 C: +27 72 419 8559 UCT: x3127 -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Bug#736260: pu: package lazr.restfulclient/0.12.0-2
Package: release.debian.org Severity: normal User: release.debian@packages.debian.org Usertags: pu Hi. Paul Wise reported launchpadlib cache corruption in the Debian wiki in #736259. It looks like the concurrency improvements in lazr.restfulclient 0.13.1 helped with the issue, and he's now running a version with the patch attached, and reports an improvement. This seems a useful commit to backport for everyone. SR diff -Nru lazr.restfulclient-0.12.0/debian/changelog lazr.restfulclient-0.12.0/debian/changelog --- lazr.restfulclient-0.12.0/debian/changelog 2012-01-23 16:59:18.0 +0200 +++ lazr.restfulclient-0.12.0/debian/changelog 2014-01-21 18:59:08.0 +0200 @@ -1,3 +1,11 @@ +lazr.restfulclient (0.12.0-2+deb7u1) stable; urgency=low + + * debian/patches/concurrency_fixes_rev_122.patch: backport rev 122: Creates +AtomicFileCache as a parent class for MultipleRepresentationCache to +handle concurrent use issues. (Closes: #736259) + + -- Stefano Rivera Sun, 05 Jan 2014 09:02:52 +0200 + lazr.restfulclient (0.12.0-2) unstable; urgency=low * New maintainer. diff -Nru lazr.restfulclient-0.12.0/debian/patches/concurrency_fixes_rev_122.patch lazr.restfulclient-0.12.0/debian/patches/concurrency_fixes_rev_122.patch --- lazr.restfulclient-0.12.0/debian/patches/concurrency_fixes_rev_122.patch 1970-01-01 02:00:00.0 +0200 +++ lazr.restfulclient-0.12.0/debian/patches/concurrency_fixes_rev_122.patch 2014-01-21 18:59:46.0 +0200 @@ -0,0 +1,309 @@ +From: Jonathan Lange +Subject: Creates AtomicFileCache as a parent class for MultipleRepresentationCache to handle concurrent use issues. +Bug-Ubuntu: http://bugs.launchpad.net/bugs/459418 +Bug-Debian: http://bugs.debian.org/736259 + +--- a/src/lazr/restfulclient/_browser.py b/src/lazr/restfulclient/_browser.py +@@ -1,4 +1,4 @@ +-# Copyright 2008 Canonical Ltd. ++# Copyright 2008,2012 Canonical Ltd. + + # This file is part of lazr.restfulclient. + # +@@ -29,19 +29,21 @@ + 'RestfulHttp', + ] + +- + import atexit +-import gzip ++import errno ++import os + import shutil ++import sys + import tempfile + # Import sleep directly into the module so we can monkey-patch it + # during a test. + from time import sleep + from httplib2 import ( +-FailedToDecompressContent, FileCache, Http, urlnorm) ++Http, ++urlnorm, ++) + import simplejson + from cStringIO import StringIO +-import zlib + + from urllib import urlencode + from wadllib.application import Application +@@ -49,6 +51,7 @@ + from errors import error_for, HTTPError + from _json import DatetimeJSONEncoder + ++ + # A drop-in replacement for httplib2's safename. + from httplib2 import _md5, re_url_scheme, re_slash + def safename(filename): +@@ -136,7 +139,99 @@ + return None + + +-class MultipleRepresentationCache(FileCache): ++class AtomicFileCache(object): ++"""A FileCache that can be shared by multiple processes. ++ ++Based on a patch found at ++<http://code.google.com/p/httplib2/issues/detail?id=125>. ++""" ++ ++TEMPFILE_PREFIX = ".temp" ++ ++def __init__(self, cache, safe=safename): ++"""Construct an ``AtomicFileCache``. ++ ++:param cache: The directory to use as a cache. ++:param safe: A function that takes a key and returns a name that's ++safe to use as a filename. The key must never return a string ++that begins with ``TEMPFILE_PREFIX``. By default uses ++``safename``. ++""" ++self._cache_dir = os.path.normpath(cache) ++self._get_safe_name = safe ++try: ++ os.makedirs(self._cache_dir) ++except OSError, e: ++if e.errno != errno.EEXIST: ++raise ++ ++def _get_key_path(self, key): ++"""Return the path on disk where ``key`` is stored.""" ++safe_key = self._get_safe_name(key) ++if safe_key.startswith(self.TEMPFILE_PREFIX): ++# If the cache key starts with the tempfile prefix, then it's ++# possible that it will clash with a temporary file that we ++# create. ++raise ValueError( ++"Cache key cannot start with '%s'" % self.TEMPFILE_PREFIX) ++return os.path.join(self._cache_dir, safe_key) ++ ++def get(self, key): ++"""Get the value of ``key`` if set. ++ ++This behaves slightly differently to ``FileCache`` in that if ++``set()`` fails to store a key, this ``get()`` will behave as if that ++key were never set whereas ``FileCache`` returns the empty string. ++ ++:param key: The key to retrieve. Must be either bytes or unicode ++text. ++:return: The value of ``key`` if set, None otherwise. ++""" +
