[SOLVED] Why does python2-numpy depend on python3 packages?

Looking at the dependencies for python2-numpy it seems to depend on a bunch of stuff that's for version 3 of python:
https://www.archlinux.org/packages/extr … on2-numpy/
In particular python-nose is an optional dependency but python-nose is a python3 package. The python2 version is, as should be, in python2-nose. It could be a bug, but I figured I'd ask if anyone can think of a good reason for this before reporting.
Last edited by vlsd (2013-10-22 19:38:49)

Scimmia wrote:Do you know for sure that the python2-numpy testsuite doesn't require python3?
Well, given that the test suite is part of numpy and is called from inside python2 I would be *very* surprised if it depended on python3. It would require importing python3-nose with a python2 interpreter, something that is, to the extent of my knowledge, not possible.
I also just tested this on my system and numpy in python2 doesn't run the tests if python-nose is installed but python2-nose is not, which makes me 100% confident the dependency is wrong.

Similar Messages

Python2-pyqt depends on python3

The package python2-pyqt depends on pyqt, which itself requires python3. Is it a normal situation, or should I fill in a bug report ?
Tondu

I changed PKGBUILDs of python2-pyqt and python2-sip when noticed that it requires python3, because there are no applications that uses python3 in my system yet. I will share them here, but use it on your own risk (if you want to, of course), because when an application that uses python3 will appear in your system - it will fail:
pkgbase=pyqt
pkgname='python2-pyqt-pure'
pkgver=4.8.4
pkgrel=1
arch=('i686' 'x86_64')
url="http://riverbankcomputing.co.uk/software/pyqt/intro"
license=('GPL')
pkgdesc="PyQt: A set of Python2 bindings for the Qt toolkit"
depends=('python2-sip' 'dbus-python')
makedepends=('qt' 'dbus-python' 'python2-sip' 'phonon'
'python-opengl' 'qt-assistant-compat')
optdepends=('phonon: enable audio and video in PyQt applications'
'python-opengl: enable OpenGL 3D graphics in PyQt applications'
'qscintilla: QScintilla API'
'qt-assistant-compat: add PyQt online help in Qt Assistant')
replaces=('python2-qt')
provides=('python2-qt' 'pyqt' 'python2-pyqt')
conflicts=('python2-pyqt')
source=("http://riverbankcomputing.co.uk/static/Downloads/PyQt4/PyQt-x11-gpl-${pkgver}.tar.gz")
md5sums=('97c5dc1042feb5b3fe20baabad055af1')
build() {
cd "${srcdir}"
cd "${srcdir}/PyQt-x11-gpl-${pkgver}"
python2 configure.py \
--confirm-license \
-v /usr/share/sip \
--qsci-api
# Thanks Gerardo for the rpath fix
find -name 'Makefile' | xargs sed -i 's|-Wl,-rpath,/usr/lib||g;s|-Wl,-rpath,.* ||g'
make
package() {
cd "${srcdir}/PyQt-x11-gpl-${pkgver}"
# INSTALL_ROOT is needed for the QtDesigner module, the other Makefiles use DESTDIR
make DESTDIR="${pkgdir}" INSTALL_ROOT="${pkgdir}" instal
pkgbase='sip'
pkgname='python2-sip-pure'
pkgver=4.12.2
pkgdesc="A tool that makes it easy to create Python2 bindings for C and C++ libraries"
pkgrel=1
arch=('i686' 'x86_64')
depends=('python2')
makedepends=('python2')
provides=('sip' 'python2-sip' 'python-sip')
conflicts=('python2-sip')
url="http://www.riverbankcomputing.com/software/sip/"
license=('custom:"sip"')
source=("http://www.riverbankcomputing.com/static/Downloads/sip4/${pkgbase}-${pkgver}.tar.gz")
md5sums=('9df80f88e0e4022cdd8a8891c6c38048')
build() {
cd "${srcdir}"
cd "${srcdir}/${pkgbase}-${pkgver}"
python2 configure.py CFLAGS="${CFLAGS}" LFLAGS="${LDFLAGS}"
make
package() {
cd "${srcdir}/${pkgbase}-${pkgver}"
make DESTDIR="${pkgdir}" install
install -Dm644 LICENSE "${pkgdir}/usr/share/licenses/${pkgname}/LICENSE"

[SOLVED] Why does jre7-openjdk-headless require a version bump?

I'm attempting to update my system, and I get the following error.
sudo pacman -Syu
:: Starting full system upgrade...
resolving dependencies...
looking for inter-conflicts...
:: jre7-openjdk-headless and jre7-openjdk-headless-infinality are in conflict. Remove jre7-openjdk-headless-infinality? [y/N]
error: unresolvable package conflicts detected
error: failed to prepare transaction (conflicting dependencies)
:: jre7-openjdk-headless and jre7-openjdk-headless-infinality are in conflict
I presume pacman wants to remove jre7-openjdk-headless-infinality because it's 7.u65_2.5.1-3, while jre7-openjdk-headless is 7.u65_2.5.1-8. I presume jre7-openjdk-headless-infinality will be updated soon, and I won't get this error.
However, I was interested in what exactly was requesting this version bump in the first place. I presume that it's a package that jre7-openjdk-headless-infinality is a dependency for. I looked firstly at what needed to be updated on my system, noting jre7-openjdk.
$ pacman -Qu
calibre 1.206.0-1
device-mapper 2.02.108-1
ffmpeg 1:2.3.1-1
jdk7-openjdk 7.u65_2.5.1-3
jre7-openjdk 7.u65_2.5.1-3
lib32-krb5 1.12.1-2
lvm2 2.02.108-1
mesa-demos 8.2.0-1
xf86-input-synaptics 1.8.0-2
I had a look at the reverse dependencies for jre7-openjdk-headless-infinality, noting jre7-openjdk again.
$ pacman -Qii jre7-openjdk-headless-infinality
Provides : java-runtime-headless=7 jre7-openjdk-headless
Depends On : libjpeg-turbo lcms2 nss ca-certificates-java
Required By : jre7-openjdk
Conflicts With : openjdk6 jre7-openjdk-headless jre7-openjdk-headless-fontfix
I interpret that as pacman -Syu saying that jre7-openjdk needs an update, and this also depends on jre7-openjdk-headless-infinality. I double checked this.
$ pacman -Qi jre7-openjdk
Provides : java-runtime=7
Depends On : jre7-openjdk-headless xdg-utils hicolor-icon-theme
So I can see that jre7-openjdk actually depends on jre7-openjdk-headless, which can be provided by jre7-openjdk-headless-infinality (as above). However, I don't understand why a new version of jre7-openjdk requires a bump in version of jre7-openjdk-headless (or jre7-openjdk-headless-infinality). Have I misunderstood something?
Last edited by Salkay (2014-08-24 12:34:29)

EDIT: actually, when testing the different "useSystemAAFontSettings" parameters I somehow missed "on" - that one does indeed work well enough with the arch java packages. I still prefer the "--enable-infinality=yes" version, but you can scrap my whining below - PEBCAK.
Wikimig wrote:What I mean is: I think you don't even need Infinality patch (just like most of people using this package). I am pretty sure you can get to a nice clean look with official Arch Linux packages.
I completely understand that the official package has to remove the infinality flag if it breaks (default) non-infinality setups - but:
You should definitely give fontconfig-infinality a try (very easy to install thanks to bohoomil's bundle). In my experience it massively improves font rendering. However, the current java7-openjdk really does not work with it, there is no way to get a nice clean look. I tried all the AA settings, and they were all terrible.
Best settings with vanilla java7
Same settings with --enable-infinality
It looks even worse in reality, but just zoom in all the way to the left upper corner and compare... these are some pretty sparkling rainbow fonts
Again, unfortunate, but I rather rebuild from ABS with "--enable-infinality=yes" than drop infinality completely.
Last edited by hokasch (2014-09-01 13:25:32)

[SOLVED]Why does .bash_profile seem tobe read in non-login shells too?

~/.bash_profile:
. $HOME/.bashrc
~/.bashrc:
export PATH=$PATH:$HOME/code/shell
(irrelevant lines left out)
(I have no ~/.profile, ~/.bash_login or * ~/.bash_logout. I have also never edited /etc/profile or /etc/bash.bashrc.)
I've read that in a login shell (e.g. TTY1) only the former is read, which in my case sources the latter. Running "export -p" in TTY1 gives a result consistent with this:
declare -x PATH="/usr/local/bin:/usr/bin:/bin:/usr/local/sbin:/usr/sbin:/sbin:/usr/bin/core_perl:/home/lazar/code/shell" (/home/lazar/code/shell is only listed once)
I've also read that in a non-login shell (e.g. xterm) only the latter is read. However, when I run "export -p" in xterm or urxvt I get an unexpected result:
declare -x PATH="/usr/local/bin:/usr/bin:/bin:/usr/local/sbin:/usr/sbin:/sbin:/usr/bin/core_perl:/home/lazar/code/shell:/home/lazar/code/shell" (/home/lazar/code/shell is listed twice!)
This implies that ~.bashrc is read twice, perhaps because .bash_profile is read in non-login shells too, even though it shouldn't be, should it? It doesn't seem to matter really, but I can't help but wonder why...
[EDIT]
I've found the answer. Exports from .bash_profile are done at boot time and *remain* active while X is active. Opening a non-login shell will cause .bashrc to export as well. Hence the duplication. My solution:
~/.bash_profile:
export PATH=$PATH:$HOME/code/shell
. $HOME/.bashrc
~/.bashrc:
(lines other than export commands)
Last edited by Lazar (2011-02-14 07:57:56)

As I wondered, I think it's a problem with passwd. I don't have a dbus entry and must have missed it when I was copying things around between my version and the .pacnew. Downloading an install disc now since I can't seem to change it from the recovery console due to mounting root read-only.
I'd love to know of a way to boot that avoids starting dbus and other things that might break while still having read/write access to root. Will mark solved once I update /etc/passwd and reboot.

[SOLVED] why does cuda-gdb require root permissions?

I am doing GPGPU development on Arch Linux with the cuda-sdk and
cuda-toolkit packages. My attempts to run cuda-gdb as a normal
user on a simple program results in:
$ cuda-gdb ./driver
NVIDIA (R) CUDA Debugger
4.2 release
Portions Copyright (C) 2007-2012 NVIDIA Corporation
GNU gdb (GDB) 7.2
Copyright (C) 2010 Free Software Foundation, Inc.
License GPLv3+: GNU GPL version 3 or later <http://gnu.org/licenses/gpl.html>
This is free software: you are free to change and redistribute it.
There is NO WARRANTY, to the extent permitted by law. Type "show copying"
and "show warranty" for details.
This GDB was configured as "x86_64-unknown-linux-gnu".
For bug reporting instructions, please see:
<http://www.gnu.org/software/gdb/bugs/>...
Reading symbols from /home/nwh/Dropbox/projects/G4CU/driver...done.
(cuda-gdb) run
Starting program: /home/nwh/Dropbox/projects/G4CU/driver
warning: Could not load shared library symbols for linux-vdso.so.1.
Do you need "set solib-search-path" or "set sysroot"?
[Thread debugging using libthread_db enabled]
fatal: The CUDA driver initialization failed. (error code = 1)
If I run cuda-gdb as root, it behaves normally:
# cuda-gdb ./driver
NVIDIA (R) CUDA Debugger
4.2 release
Portions Copyright (C) 2007-2012 NVIDIA Corporation
GNU gdb (GDB) 7.2
Copyright (C) 2010 Free Software Foundation, Inc.
License GPLv3+: GNU GPL version 3 or later <http://gnu.org/licenses/gpl.html>
This is free software: you are free to change and redistribute it.
There is NO WARRANTY, to the extent permitted by law. Type "show copying"
and "show warranty" for details.
This GDB was configured as "x86_64-unknown-linux-gnu".
For bug reporting instructions, please see:
<http://www.gnu.org/software/gdb/bugs/>...
Reading symbols from /home/nwh/Dropbox/work/2012-09-06-cuda_gdb/driver...done.
(cuda-gdb) run
Starting program: /home/nwh/Dropbox/work/2012-09-06-cuda_gdb/driver
warning: Could not load shared library symbols for linux-vdso.so.1.
Do you need "set solib-search-path" or "set sysroot"?
[Thread debugging using libthread_db enabled]
[New Thread 0x7ffff5ba8700 (LWP 11386)]
[Context Create of context 0x6e8a30 on Device 0]
[Launch of CUDA Kernel 0 (thrust::detail::backend::cuda::detail::launch_closure_by_value<thrust::detail::backend::cuda::for_each_n_closure<thrust::device_ptr<unsigned long long>, unsigned int, thrust::detail::device_generate_functor<thrust::detail::fill_functor<unsigned long long> > > ><<<(1,1,1),(704,1,1)>>>) on Device 0]
[Launch of CUDA Kernel 1 (set_vector<<<(1,1,1),(10,1,1)>>>) on Device 0]
vd[0] = 0
vd[1] = 1
vd[2] = 2
vd[3] = 3
vd[4] = 4
vd[5] = 5
vd[6] = 6
vd[7] = 7
vd[8] = 8
vd[9] = 9
[Thread 0x7ffff5ba8700 (LWP 11386) exited]
Program exited normally.
[Termination of CUDA Kernel 1 (set_vector<<<(1,1,1),(10,1,1)>>>) on Device 0]
[Termination of CUDA Kernel 0 (thrust::detail::backend::cuda::detail::launch_closure_by_value<thrust::detail::backend::cuda::for_each_n_closure<thrust::device_ptr<unsigned long long>, unsigned int, thrust::detail::device_generate_functor<thrust::detail::fill_functor<unsigned long long> > > ><<<(1,1,1),(704,1,1)>>>) on Device 0]
The test program driver.cu is:
// needed for nvcc with gcc 4.7 and iostream
#undef _GLIBCXX_ATOMIC_BUILTINS
#undef _GLIBCXX_USE_INT128
#include <iostream>
#include <thrust/device_vector.h>
#include <thrust/host_vector.h>
__global__
void set_vector(int *a)
// get thread id
int id = threadIdx.x + blockIdx.x * blockDim.x;
a[id] = id;
__syncthreads();
int main(void)
// settings
int len = 10; int trd = 10;
// allocate vectors
thrust::device_vector<int> vd(len);
// get the raw pointer
int *a = thrust::raw_pointer_cast(vd.data());
// call the kernel
set_vector<<<1,trd>>>(a);
// print vector
for (int i=0; i<len; i++)
std::cout << "vd[" << i << "] = " << vd[i] << std::endl;
return 0;
driver.c is compiled with the command:
$ nvcc -g -G -gencode arch=compute_20,code=sm_20 driver.cu -o driver
How can I get cuda-gdb to run with out root permissions?
Last edited by nwhsvc (2012-10-21 19:07:08)

Question: What is one mayor difference between root and pretty much any other user account ?
Hint: Access rights....

[SOLVED] Why does my JAVA_HOME point to jre instead of jdk?

I am having a strange phanomenon.
I have jdk7-openjdk installed, but I still have
% echo $JAVA_HOME
/usr/lib/jvm/java-7-openjdk/jre
This is only with my current user. When I try root or a freshly created user, I correctly get
% echo $JAVA_HOME
/usr/lib/jvm/java-7-openjdk
So I guess this is caused by something in my home directory. As 'grep -r JAVA_HOME ~/.*' does not give any hints, I do not know how to investigate this further.
Maybe someone here has any idea how to find out more.
Last edited by cbuehler (2013-12-22 10:20:30)

As far as I understand, the order is not important in this case, because jre.sh only overrides JAVA_HOME if it is not set before.
I also like to avoid making configuration changes when it looks like the default config should work fine.
But brebs' reply inspired me to run 'sh -x /etc/profile', which finally shed light into my issue:
+++ . /home/cbuehler/Siemens/Teamcenter10.1/Visualization/bin/visviewwebinit /home/cbuehler/Siemens/Teamcenter10.1/Visualization
++++ '[' X_ = X_True ']'
++++ '[' -n /home/cbuehler/Siemens/Teamcenter10.1/Visualization ']'
++++ export EAI_INSTALL=/home/cbuehler/Siemens/Teamcenter10.1/Visualization
++++ EAI_INSTALL=/home/cbuehler/Siemens/Teamcenter10.1/Visualization
++++ '[' -d /usr/lib/jvm/java-7-openjdk/jre -o -h /usr/lib/jvm/java-7-openjdk/jre ']'
++++ export JAVA_HOME=/usr/lib/jvm/java-7-openjdk/jre
++++ JAVA_HOME=/usr/lib/jvm/java-7-openjdk/jre
I have Teamcenter Visualization installed in my home directory. And this changes my JAVA_HOME in /home/cbuehler/Siemens/Teamcenter10.1/Visualization/bin/visviewwebinit:
# If JAVA_HOME points to the top level java directory, reset it to point to the JRE directory.
if [ -d $JAVA_HOME/jre -o -h $JAVA_HOME/jre ]; then
export JAVA_HOME=$JAVA_HOME/jre
fi
I commented this part out and now everything works fine.
Thank you very much for your help.

[SOLVED] Why does nano "ctrl+o" `selects all`?

I am frequently using nano, but recently when I try to save the file and hit "ctrl+o" all text gets selected instead of saving the file.
Nevertheless, I can save the file in nano only if I switch my keyboard to Russian layout. Then "ctrl+o" correctly works.
Using English layout I can invoke `save file` by hitting "ctrl+shift+o" but it all worked perfectly with "ctrl+o" before.
I have had this nasty behaviour before but forgot what was the reason and how to fix it.
Does anybody know what's going on?
Last edited by minerals (2015-03-26 17:00:45)

Alad wrote:I didn't know you could select all text with a single shortcut... usually people use ctrl+a with ctrl+/ for that. What terminal and window manager do you use? Have you run the keys through xev (from xorg-xev) ?
Oh, you're right. It is my "lilyterm" that overrides the default nano binding. In lilyterm "ctrl+o" selects all text area. Now, I reassigned the hotkey and nano is back to normal. Thanks!

[SOLVED] Why does android-sdk remain in AUR?

It seems that with 1500+ votes it might be worthy of being put in [community]. I'm just curious.
Last edited by amadar (2013-01-14 00:07:12)

The two most common reasons that popular packages stay in the AUR:
1. Licensing
2. No dev or TU uses them and can't maintain them.
A quick search through the license turns up this, emphasis added by me:
3.3 You may not use the SDK for any purpose not expressly permitted by this License Agreement. Except to the extent required by applicable third party licenses, you may not: (a) copy (except for backup purposes), modify, adapt, redistribute, decompile, reverse engineer, disassemble, or create derivative works of the SDK or any part of the SDK; or (b) load any part of the SDK onto a mobile handset or any other hardware device except a personal computer, combine any part of the SDK with other software, or distribute any software or device incorporating a part of the SDK.
Last edited by Scimmia (2012-12-27 15:58:41)

[SOLVED] Why does `dd if=/dev/zero of=/dev/sdc` cause disk-reads?

I have noticed this on multiple occasions. When wiping a drive, but also when writing other content to it; both atop and iotop show that the disk is reading and writing the same amount of data. This feels like it is slowing down the writing I want it to do. Is there an option in dd I'm missing? Happens on multiple machines, kernels and disks. It feels like the linux-kernel is verifying what was written to the disk; but I cannot find any documentation on this
Last edited by Spider.007 (2014-07-15 07:43:21)

The default dd block size is 512 bytes, which is the traditional size of HDD sectors.
However, most of the time, Linux performs disk IO in 4kB blocks. When you submit a sequence of 512 byte write requests, it reads a 4kB block, modifies it part-by-part in 512 byte chunks, writes it back, reads another 4kB block and so on.
Setting bs to at least 4kB should eliminate the reads. Increasing it to few MB may improve performance slightly further by reducing the number of individual write requests and hence the time spent by the kernel on "understanding" them.
Last edited by mich41 (2014-07-15 09:31:43)

Why does Arch use .tar.gz as package files?

As opposed to a tar.gz file with a .pac extension much like a .deb from Debian. I'm just curious...
Arch is a great Distro. It has taken over my desktop... something that hasn't happened since 1999!

Nevermind... I found the official reason here:
http://wiki.archlinux.org/index.php/FAQ … _confusing
If possible, I would like to add my vote to the "confusing/long" portion of the above.

Why does my MBA freeze at "running package scripts" when loading.en trying to load itunes 10.4

Can download the 92 mb of upgrade, but cannot get it to get beyond 2 mins and close itunes.

Thanks, Tom
you took me back to first principles (forced me out of laziness, i guess) and another half-day of trying again during which i incrementally reduced the upgrade back to the original condition - video, then ram, then processor - finally brought the machine to life again with new install on the new 120gb disc of 10.4.6 now upgraded to 10.4.10
i am now restarting the upgrade with additional increments learned through this affair, such as testing memory one item at a time to check the machine's view of its compatibility; proposing to start with the ram - on the grounds it should be simplest (after using memory checker), then video, finally then processor.
hope to discover the source of the gremlin and will report what i find, just in case others could benefit
again, thanks for taking an interest
best regards
crish
hope to uncover just what was the cause of the failures

Why does captivate hang when it is closed in window 8.1 professional ? How is it solved?

Why does captivate hang when it is closed in window 8.1 professional ? How is it solved?

Hi,
Do you see the issue every time you close captivate or is it only in some scenarios? Does it show any error messages? What captivate version are you using (64 bit or 32 bit)? Can you provide some more information or screenshots which can help us reproduce the issue.
Tamish

Why does not the scroll on tbody in new versions of Firefox and how it can solve the problem?

why does not the scroll on tbody in new versions of Firefox and how it can solve the problem?

scrolling overflow on tbody is no longer supported because that is not allowed in CSS 2.1 specification.
See [https://developer.mozilla.org/en/CSS/overflow notes for overflow]

Why does KOffice depend on gtk?

When I try to install KOffice, I get:
resolving dependencies... done.
looking for inter-conflicts... done.
Targets: libgsf-1.14.3-2 wv2-0.2.3-1 lcms-1.16-1 libwmf-0.2.8.4-4
atk-1.18.0-2 cairo-1.4.6-2 libthai-0.1.8-1 pango-1.16.4-1
gtk2-2.10.13-1 libcroco-0.6.1-1 librsvg-2.16.1-1
imagemagick-6.3.4.4-1 python-2.5.1-1 libwpd-0.8.9-1
libexif-0.6.16-1 koffice-1.6.3-2 libdatrie-0.1.1-1
Total Package Size: 49.97 MB
Proceed with installation? [Y/n]
Why does it depend on gtk2?
Last edited by miggols99 (2007-07-08 17:30:57)

librsvg is dependent on GTK2. This was discussed at KDE-devel:
http://lists.kde.org/?l=koffice-devel&m … 425620&w=2
Last edited by skottish (2007-07-08 17:44:57)

Why does my Alesis ioj2 express only come out through one side? When I sing into the microphone, it only comes out one side. I've set everything to stereo and it still has not solved the problem. Please help ASAP.

Please help with my problem. I need to start recording!!

Chris Huber wrote:
Why does my Alesis ioj2 express only come out through one side?
http://www.bulletsandbones.com/GB/GBFAQ.html#leftspeakeronly
(Let the page FULLY load. The link to your answer is at the top of your screen)

[SOLVED] Why does python2-numpy depend on python3 packages?

Similar Messages

Maybe you are looking for