Re: [Kwant] Want to know more about bulid_discretized

[Bas Nijholt]

Oops, never mind my remark about the documentation being incorrect!
It was just because discretize_symbolic returns a tuple which you fed as a
first argument to build_discretized.

On Fri, Mar 29, 2019 at 3:28 PM Bas Nijholt  wrote:

> It seems like the documentation here
> <https://gitlab.kwant-project.org/kwant/kwant/blob/fdd4674/kwant/continuum/discretizer.py#L263>
> is wrong because it expects a dictionary.
>
>
> I see that you do
>
> template = kwant.continuum.discretize_symbolic(hamiltonian)
> t = kwant.continuum.build_discretized(template, coords = ('x','y','z'))
>
> However, discretize_symbolic returns two variables. Instead do:
>
> template, coods = kwant.continuum.discretize_symbolic(hamiltonian)
> t = kwant.continuum.build_discretized(template, coords)
>
> The list of coods should be passed to discretize_symbolic. However, if
> you omit it, it will be inferred from your Hamiltonian’s k_{x, y, z}
> terms.
>
> Best, Bas
>
> On Fri, Mar 29, 2019 at 3:11 PM Naveen Yadav 
> wrote:
>
>> Hi Bas,
>>
>> Thanks for your reply.
>>
>> I have used the same syntax. But it is showing an Attribute Error: 'tuple
>> object has no attribute 'items'
>>
>> from __future__ import division
>> import kwant
>> import kwant.continuum
>> import scipy.sparse.linalg
>> import scipy.linalg
>> import numpy as np
>> import sympy
>> from math import sin, cos, sqrt, pi
>> import matplotlib.pyplot as plt
>> hamiltonian = ("""k_x**2 + k_y**2 + k_z**2""")
>> hamiltonian = kwant.continuum.sympify(hamiltonian)
>> template = kwant.continuum.discretize_symbolic(hamiltonian)
>> t = kwant.continuum.build_discretized(template, coords = ('x', 'y', 'z'))
>> print(t)
>>
>> AttributeErrorTraceback (most recent call 
>> last) in  11 hamiltonian = 
>> kwant.continuum.sympify(hamiltonian) 12 template = 
>> kwant.continuum.discretize_symbolic(hamiltonian)---> 13 t = 
>> kwant.continuum.build_discretized(template, coords = ('x','y','z')) 14 
>> print(t) 15 hamiltonian
>> /usr/local/lib/python3.6/dist-packages/kwant/continuum/discretizer.py in 
>> build_discretized(tb_hamiltonian, coords, grid_spacing, locals)281 
>> 282 with reraise_warnings():--> 283 for k, v in 
>> tb_hamiltonian.items():284 tb_hamiltonian[k] = sympify(v, 
>> locals)285
>> AttributeError: 'tuple' object has no attribute 'items'
>>
>> What should I do.
>>
>> Naveen
>>
>>
>> On Fri, Mar 29, 2019 at 5:37 PM Bas Nijholt  wrote:
>>
>>> Hi Naveen,
>>>
>>> In the documentation
>>> <https://kwant-project.org/doc/dev/reference/generated/kwant.continuum.build_discretized>
>>> you can see that coords require a sequence of strings, so for example:
>>>
>>> coords = ('x',)  # for 1D
>>> coords = ('x', 'y', 'z')  # for 3D
>>>
>>> Best, Bas
>>>
>>> On Fri, Mar 29, 2019 at 12:59 PM Naveen Yadav 
>>> wrote:
>>>
>>>> I want to know more about the builid_discretized(tb_hamiltonian,
>>>> coords).
>>>> I did not undersand what should be the syntax to write the coords.Could
>>>> you please help me by sending an example code for writing coords?
>>>>
>>>> Thanks in Advance.
>>>>
>>>>
>>>>
>>>>
>>>>
>>>>
>>>>
>>>>
>>>>
>>>>
>>>>
>>>>
>>>>
>>>> Naveen
>>>> Department of Physics & Astrophysics
>>>> University of Delhi
>>>> 110007
>>>>
>>>
>>
>> --
>>
>>
>> With Best Regards
>> NAVEEN YADAV
>> Ph.D Research Scholar
>> Deptt. Of Physics & Astrophysics
>> University Of Delhi.
>>
>

Re: [Kwant] Want to know more about bulid_discretized

[Bas Nijholt]

It seems like the documentation here
<https://gitlab.kwant-project.org/kwant/kwant/blob/fdd4674/kwant/continuum/discretizer.py#L263>
is wrong because it expects a dictionary.


I see that you do

template = kwant.continuum.discretize_symbolic(hamiltonian)
t = kwant.continuum.build_discretized(template, coords = ('x','y','z'))

However, discretize_symbolic returns two variables. Instead do:

template, coods = kwant.continuum.discretize_symbolic(hamiltonian)
t = kwant.continuum.build_discretized(template, coords)

The list of coods should be passed to discretize_symbolic. However, if you
omit it, it will be inferred from your Hamiltonian’s k_{x, y, z} terms.

Best, Bas

On Fri, Mar 29, 2019 at 3:11 PM Naveen Yadav 
wrote:

> Hi Bas,
>
> Thanks for your reply.
>
> I have used the same syntax. But it is showing an Attribute Error: 'tuple
> object has no attribute 'items'
>
> from __future__ import division
> import kwant
> import kwant.continuum
> import scipy.sparse.linalg
> import scipy.linalg
> import numpy as np
> import sympy
> from math import sin, cos, sqrt, pi
> import matplotlib.pyplot as plt
> hamiltonian = ("""k_x**2 + k_y**2 + k_z**2""")
> hamiltonian = kwant.continuum.sympify(hamiltonian)
> template = kwant.continuum.discretize_symbolic(hamiltonian)
> t = kwant.continuum.build_discretized(template, coords = ('x', 'y', 'z'))
> print(t)
>
> AttributeErrorTraceback (most recent call 
> last) in  11 hamiltonian = 
> kwant.continuum.sympify(hamiltonian) 12 template = 
> kwant.continuum.discretize_symbolic(hamiltonian)---> 13 t = 
> kwant.continuum.build_discretized(template, coords = ('x','y','z')) 14 
> print(t) 15 hamiltonian
> /usr/local/lib/python3.6/dist-packages/kwant/continuum/discretizer.py in 
> build_discretized(tb_hamiltonian, coords, grid_spacing, locals)281 
> 282 with reraise_warnings():--> 283 for k, v in 
> tb_hamiltonian.items():284 tb_hamiltonian[k] = sympify(v, 
> locals)285
> AttributeError: 'tuple' object has no attribute 'items'
>
> What should I do.
>
> Naveen
>
>
> On Fri, Mar 29, 2019 at 5:37 PM Bas Nijholt  wrote:
>
>> Hi Naveen,
>>
>> In the documentation
>> <https://kwant-project.org/doc/dev/reference/generated/kwant.continuum.build_discretized>
>> you can see that coords require a sequence of strings, so for example:
>>
>> coords = ('x',)  # for 1D
>> coords = ('x', 'y', 'z')  # for 3D
>>
>> Best, Bas
>>
>> On Fri, Mar 29, 2019 at 12:59 PM Naveen Yadav 
>> wrote:
>>
>>> I want to know more about the builid_discretized(tb_hamiltonian, coords).
>>> I did not undersand what should be the syntax to write the coords.Could
>>> you please help me by sending an example code for writing coords?
>>>
>>> Thanks in Advance.
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>> Naveen
>>> Department of Physics & Astrophysics
>>> University of Delhi
>>> 110007
>>>
>>
>
> --
>
>
> With Best Regards
> NAVEEN YADAV
> Ph.D Research Scholar
> Deptt. Of Physics & Astrophysics
> University Of Delhi.
>

Re: [Kwant] Want to know more about bulid_discretized

[Bas Nijholt]

Hi Naveen,

In the documentation

you can see that coords require a sequence of strings, so for example:

coords = ('x',)  # for 1D
coords = ('x', 'y', 'z')  # for 3D

Best, Bas

On Fri, Mar 29, 2019 at 12:59 PM Naveen Yadav 
wrote:

> I want to know more about the builid_discretized(tb_hamiltonian, coords).
> I did not undersand what should be the syntax to write the coords.Could
> you please help me by sending an example code for writing coords?
>
> Thanks in Advance.
>
>
>
>
>
>
>
>
>
>
>
>
>
> Naveen
> Department of Physics & Astrophysics
> University of Delhi
> 110007
>

Re: [Kwant] Installation error on Mac

[Bas Nijholt]

Hi Girish,

Currently the recommended way to install kwant on macOS is with conda.

See the instructions on https://kwant-project.org/install#

Best, Bas

> On 15 May 2018, at 09:20, Girish Sharma  wrote:
>
> Dear All,
>
> I am trying to install Kwant on my Mac High Sierra 10.13.4. I am facing the 
> following error after brew tap homebrew/science command:
>
> brew tap homebrew/science
> Updating Homebrew...
> ==> Auto-updated Homebrew!
> Updated 1 tap (homebrew/core).
> ==> Updated Formulae
> aptly   gtksourceview@4 nanorocksdb
> armadillo   jhiccup node-build  solr
> dashkubernetes-helm osquery spigot
> dscannerlibu2f-host pam-u2f spring-roo
> fabric  maxwell postgrest
> fn  mercurial   pound
> grpcmpc proguard
>
> Error: homebrew/science was deprecated. This tap is now empty as all its 
> formulae were migrated.
>
> Could any one of you please help me with this.
>
> Thank you.
>
> Best regards,
> Girish

Re: [Kwant] parallel mumps

[Bas Nijholt]

Hi Benoit,

I took the liberty of copying your private mail here since I think your
efforts will be better spent by working on the already existing repo which
is trying to accomplish what you are trying to do.

Hi Bas,

I saw your python bindings for mumps on github. They look like the ones for
kwant. I wanted to know if there’s something specific in the code that says
that it’s for the sequential version of mumps.

I’m asking you because I’m trying to modify the python bindings for mumps
that can be found in PyMumps (for real numbers), and it’s not a question
related to kwant.

There’s this comm_fortran parameter that could be the key…

There is already an attempt to separate MUMPS from kwant, the repo you
found was only the start (mostly copy-pasting), but Joe improved it a lot
(on a fork) and added the missing data types, and it can be found here
<https://gitlab.kwant-project.org/kwant/mumpy>.

However, I am afraid the repository is still private but I see no reason to
not make it public? Christoph, Joe, or Anton can probably comment on this.

Best, Bas
​

On Fri, Dec 29, 2017 at 5:50 PM, Benoit Gaury <benoitga...@gmail.com> wrote:

> Hi Bas,
>
> I'm just curious about this, because the MUMPS dependency that kwant wants
> to install on Debian is the parallel version of the library. However, as
> far as I know, kwant only uses the sequential version.
>
> Benoit
>
> On Fri, Dec 29, 2017 at 5:43 PM, Bas Nijholt <basnijh...@gmail.com> wrote:
>
>> Hi Benoit,
>>
>> I’ve tried to get kwant to work with MUMPS compiled with MPI support but
>> I couldn’t get it to work.
>>
>> I am sorry, but I don’t remember the reasons exactly, however, I can
>> explain what I did at least.
>>
>> I am one of the maintainers of the MUMPS conda build
>> <https://github.com/conda-forge/mumps-feedstock> where we switched to
>> the parallel version of MUMPS. I tried to get it to work with kwant and
>> you can see my (somewhat cryptic) attempts here
>> <https://github.com/conda-forge/kwant-feedstock/issues/34> and in the
>> issues and PRs linked there. In the end (as I mentioned) I couldn’t get it
>> to work and neither with the help of Joe (one of the Kwant authors.)
>>
>> Hope this helps, best,
>>
>> Bas
>> ​
>>
>> On Fri, Dec 29, 2017 at 5:33 PM, Benoit Gaury <benoitga...@gmail.com>
>> wrote:
>>
>>> Hi,
>>>
>>> I wanted to know what kwant does if the parallel version of the MUMPS
>>> library is installed instead of the sequential version. Will it be ignored?
>>>
>>> Benoit
>>>
>>
>>
>

Re: [Kwant] parallel mumps

[Bas Nijholt]

Hi Benoit,

I’ve tried to get kwant to work with MUMPS compiled with MPI support but I
couldn’t get it to work.

I am sorry, but I don’t remember the reasons exactly, however, I can
explain what I did at least.

I am one of the maintainers of the MUMPS conda build
 where we switched to the
parallel version of MUMPS. I tried to get it to work with kwant and you can
see my (somewhat cryptic) attempts here
 and in the
issues and PRs linked there. In the end (as I mentioned) I couldn’t get it
to work and neither with the help of Joe (one of the Kwant authors.)

Hope this helps, best,

Bas
​

On Fri, Dec 29, 2017 at 5:33 PM, Benoit Gaury  wrote:

> Hi,
>
> I wanted to know what kwant does if the parallel version of the MUMPS
> library is installed instead of the sequential version. Will it be ignored?
>
> Benoit
>

Re: [Kwant] seg fault at test run on CentOS

[Bas Nijholt]

Hi Faith,

I would suggest to try to install kwant via conda.

With: conda install -c conda-forge kwant


Best, Bas

On 27 Feb 2017, 11:55 +0100, Fatih Dogan , wrote:
> Hello,
>
> I have installed the latest version of kwant on a CentOS system that I do not 
> have root access. So I installed it on my home directory. When I run the test 
> file, i get the following seg fault.
>
> python3 -c 'import kwant; kwant.test(verbose=2)'
> /lib/python3.6/site-packages/kwant-unknown-py3.6-linux-x86_64.egg/kwant/solvers/default.py:18:
>  RuntimeWarning: MUMPS is not available, SciPy built-in solver will be used 
> as a fallback. Performance can be very poor in this case.
>   "Performance can be very poor in this case.", RuntimeWarning)
> kwant.graph.tests.test_core.test_empty ... ok
> kwant.graph.tests.test_core.test_num_nodes ... ok
> 
> kwant.tests.test_lattice.test_neighbors ... ok
> kwant.tests.test_lattice.test_shape ... ok
> kwant.tests.test_lattice.test_wire ... ok
> kwant.tests.test_lattice.test_translational_symmetry ... Segmentation fault
>
> Where I skipped a lot of lines, they were ok for most, skipped when related 
> to MUMPS. I am not that familiar with python or kwant to know what this error 
> means. Can someone help?
>
> Thanks
>
> Fatih

Re: [Kwant] Installation instructions for Kwant on WSL

[Bas Nijholt]

Sorry, I wasn’t clear in my last message, I accidentally copied the 
instructions for installing conda and kwant on OSX.

What I meant to say is, that you can just install conda for Linux and use it on 
WSL (at least from Googling a bit I see that other packages work).

On 23 Jan 2017, 22:19 +0100, Bas Nijholt <basnijh...@gmail.com>, wrote:
> As far as I understand, you could just use conda:
>
> Conda
> Conda is a cross-platform package and environment manager that installs, 
> runs, and updates packages and their dependencies. It allows you to easily 
> set up and switch between environments on your local computer. Conda is 
> included in all versions of Anaconda and Miniconda.
> Anaconda is Python distribution that includes 150 installed of the most 
> popular Python packages for science. Miniconda is an lighter alternative to 
> Anaconda and comes just with Python and conda.
>
> 1. > Open a terminal and download Miniconda (or Anaconda) by running
>
> wget https://repo.continuum.io/miniconda/Miniconda3-latest-MacOSX-x86_64.sh
> 2. > Install Miniconda with
>
> bash Miniconda3-latest-MacOSX-x86_64.sh
> and follow its instructions. Make sure that conda is in your PATH, which you 
> can do by adding export PATH="$HOME/miniconda3/bin:$PATH" to your .bashrc or 
> .bash_profile.
> 3. > Add the conda-forge channel and install Kwant and its dependencies with
>
> conda config --add channels conda-forge
> conda install kwant
>
>
>
> On 23 Jan 2017, 21:34 +0100, Christoph Groth <christoph.gr...@cea.fr>, wrote:
> > Hi Andrey,
> >
> > Thanks for sharing your instructions. Do you see a way to
> > automatize this somehow, e.g. to provide Kwant packages for WSL?
> > Is it possible to install our regular Ubuntu Kwant packages [1] in
> > WSL?
> >
> > Christoph
> >
> > [1] https://kwant-project.org/install#ubuntu-and-derivatives

Re: [Kwant] "No module named kwant" when using iPython

[Bas Nijholt]

Hi Cheryne,

I think you are not in the correct environment in Spyder, one in which you
didn't install kwant.

How did you install your Python and kwant?

One easy thing to get everything running is to use Anaconda and then
install kwant with: conda install -c basnijholt kwant=1.2.2

Best, Bas

On Mon, Aug 29, 2016 at 2:21 PM, Cheryne  wrote:

> Hi Kwant-Team!
>
> I'm having trouble to "import kwant" via iPython interfaces (Spyder). I
> can run the python script via "python scriptname.py" but when I try
> "spyder" and then "scriptname.py" I get the error "no module named kwant".
> Do you know what I have to do so that iPython recognizes the module?
>
> Thank you,
> Cheryne
>

Re: [Kwant] Conda packages for Kwant

[Bas Nijholt]

Hi Christoph,

They include MUMPS, Scotch, and Metis.

I do plan to maintain these, since for the coming 4 years I’ll definitely be 
working with Kwant and probably on my Mac.

It’s definitely useful to use conda, installing a complete Python environment 
plus kwant is now done in ~2 minutes.

I’ll give compiling for Windows a try some day, however probably I won’t make 
time for this in the coming weeks.

Bas Nijholt

On Fri, 19 Feb 2016 at 10:56 Christoph Groth

<
mailto:Christoph Groth <christoph.gr...@cea.fr>
> wrote:

Hi Bas,

Thanks for packaging Kwant for conda! Do these packages include

MUMPS?

Do you think you could continue to maintain these packages for a

while? If yes, and if no grave problems show up after a few days

or so, you would be very welcome to update
http://kwant-project.org/install.
(You can either clone the Kwant

website repository or we can give you directly write access to

it.)

I think conda will be especially interesting for people using OS X

and Windows. It would be really nice if we could provide some

easier way to install Kwant on Windows. Do you know how much work

that would be?

Christoph

[Kwant] Conda packages for Kwant

[Bas Nijholt]

Hi all,

I've compiled conda packages for kwant (and it's dependencies) for Python
2.7, 3.4, and 3.5 for both Linux and OS X.

They can be found on anaconda.org/basnijholt/kwant.

The advantage of having these packages is that they are already compiled.
Conda is a packaging tool and installer that aims to do more than what pip
does; handle library dependencies outside of the Python packages as well as
the Python packages themselves.

In this way you won't have to install any other dependencies, so when you
have a fresh Ubuntu install, all you need to do is install miniconda
<http://conda.pydata.org/miniconda.html> or Anaconda
<https://www.continuum.io/downloads> and run:

conda install -c https://conda.anaconda.org/basnijholt kwant


That's it! Let me know if you run into any trouble.


Best, Bas Nijholt

Re: [Kwant] Fresh install of kwant on OS X failed

[Bas Nijholt]

/temp.macosx-10.10-x86_64-2.7/kwant/_system.o

kwant/_system.c:303:10: fatal error: 'numpy/arrayobject.h' file not found

#include numpy/arrayobject.h

         ^

1 error generated.

**

The compilation of Kwant has failed.  Please examine the error message

above and consult the installation instructions in README.rst.

You might have to customize build.conf.

**

Build configuration was:

User-configured LAPACK and BLAS

User-configured MUMPS

**

error: command 'clang' failed with exit status 1

/usr/local/Library/Homebrew/debrew.rb:11:in `raise'

BuildError: Failed executing: python setup.py install 
--prefix=/usr/local/Cellar/kwant/1.0.3

1. raise

2. ignore

3. backtrace

4. irb

5. shell

Choose an action: 1

== Formula

Tap: basnijholt/homebrew-kwant

Path: /usr/local/Library/Taps/basnijholt/homebrew-kwant/kwant.rb

== Configuration

HOMEBREW_VERSION: 0.9.5

ORIGIN: https://github.com/Homebrew/homebrew

HEAD: cea7b94a0b7c96ed5b90af366634e96ba27620a8

Last commit: 11 hours ago

HOMEBREW_PREFIX: /usr/local

HOMEBREW_CELLAR: /usr/local/Cellar

CPU: quad-core 64-bit haswell

OS X: 10.10.3-x86_64

Xcode: 6.3.2

CLT: 6.3.2.0.1.1431401888

Clang: 6.1 build 602

X11: N/A

System Ruby: 2.0.0-p481

Perl: /usr/bin/perl

Python: /usr/local/bin/python = 
/usr/local/Cellar/python/2.7.10/Frameworks/Python.framework/Versions/2.7/bin/python2.7

Ruby: /usr/bin/ruby

Java: 1.8.0_45

== ENV

HOMEBREW_CC: clang

HOMEBREW_CXX: clang++

MAKEFLAGS: -j4

CMAKE_PREFIX_PATH: 
/usr/local/opt/scotch5:/usr/local/opt/metis4:/usr/local/opt/mumps_seq:/usr/local

CMAKE_INCLUDE_PATH: 
/usr/include/libxml2:/System/Library/Frameworks/OpenGL.framework/Versions/Current/Headers

CMAKE_LIBRARY_PATH: 
/System/Library/Frameworks/OpenGL.framework/Versions/Current/Libraries

CMAKE_FRAMEWORK_PATH: /usr/local/opt/python/Frameworks

PKG_CONFIG_PATH: 
/usr/local/opt/python/lib/pkgconfig:/usr/local/opt/open-mpi/lib/pkgconfig

PKG_CONFIG_LIBDIR: /usr/lib/pkgconfig:/usr/local/Library/ENV/pkgconfig/10.10

ACLOCAL_PATH: /usr/local/share/aclocal

PATH: 
/usr/local/opt/python/bin:/usr/local/Library/ENV/4.3:/usr/local/opt/scotch5/bin:/usr/local/opt/metis4/bin:/usr/local/opt/gcc/bin:/usr/local/opt/python/bin:/usr/local/opt/open-mpi/bin:/usr/bin:/bin:/usr/sbin:/sbin:/usr/local/bin




Error: basnijholt/kwant/kwant 1.0.3 did not build

Logs:

     /Users/bnijholt/Library/Logs/Homebrew/kwant/01.python

     /Users/bnijholt/Library/Logs/Homebrew/kwant/01.python.cc

On Tue, Jun 23, 2015 at 8:32 AM, Michael Wimmer
wim...@lorentz.leidenuniv.nl wrote:

 Can you try to run
 brew install -vd kwant
 to get a more detailed error?
 Best,
 Michael
 Am 23 jun. 2015 um 02:00 schrieb Bas Nijholt basnijh...@gmail.com:
 
 I’ve completely reinstalled my OS and am now encountering issues that I’ve 
 heard other talk about, I’ve followed the instructions on the website word 
 by word, except that I forked Michael’s repo and changed to kwant version 
 1.0.3. This has worked before for me.
 
 
 
 This is the error message:
 brew install kwant
 == Installing kwant from basnijholt/homebrew-kwant
 == Using Homebrew-provided fortran compiler.
 This may be changed by setting the FC environment variable.
 == Building with an alternative Fortran compiler
 This is unsupported.
 Warning: No Fortran optimization information was provided.  You may want to 
 consider
 setting FCFLAGS and FFLAGS or pass the `--default-fortran-flags` option to
 `brew install` if your compiler is compatible with GCC.
 
 If you like the default optimization level of your compiler, ignore this
 warning.
 == Downloading pypi.python.org/packages/source/k/kwant/kwant-1.0.3.tar.gz
 Already downloaded: /Library/Caches/Homebrew/kwant-1.0.3.tar.gz
 == Patching
 patching file build.conf
 == python setup.py install --prefix=/usr/local/Cellar/kwant/1.0.3
 Build configuration was:
 User-configured LAPACK and BLAS
 User-configured MUMPS
 **
 error: command 'clang' failed with exit status 1
 
 READ THIS: https://git.io/brew-troubleshooting
 If reporting this issue please do so at (not Homebrew/homebrew):
   https://github.com/basnijholt/homebrew-kwant/issues
 
 
 
 Best Bas
 

[Kwant] Fresh install of kwant on OS X failed

[Bas Nijholt]

I’ve completely reinstalled my OS and am now encountering issues that I’ve 
heard other talk about, I’ve followed the instructions on the website word by 
word, except that I forked Michael’s repo and changed to kwant version 1.0.3. 
This has worked before for me.






This is the error message:
brew install kwant
== Installing kwant from basnijholt/homebrew-kwant
== Using Homebrew-provided fortran compiler.
This may be changed by setting the FC environment variable.
== Building with an alternative Fortran compiler
This is unsupported.
Warning: No Fortran optimization information was provided.  You may want to 
consider
setting FCFLAGS and FFLAGS or pass the `--default-fortran-flags` option to
`brew install` if your compiler is compatible with GCC.


If you like the default optimization level of your compiler, ignore this
warning.
== Downloading pypi.python.org/packages/source/k/kwant/kwant-1.0.3.tar.gz
Already downloaded: /Library/Caches/Homebrew/kwant-1.0.3.tar.gz
== Patching
patching file build.conf
== python setup.py install --prefix=/usr/local/Cellar/kwant/1.0.3
Build configuration was:
User-configured LAPACK and BLAS
User-configured MUMPS
**
error: command 'clang' failed with exit status 1


READ THIS: https://git.io/brew-troubleshooting
If reporting this issue please do so at (not Homebrew/homebrew):
  https://github.com/basnijholt/homebrew-kwant/issues






Best Bas

[Kwant] kwant on OS X

[Bas Nijholt]

Small update on installing kwant on OS X:

brew install gfortran python  brew install gcc python


and why is there: brew tap samueljohn/python instead of brew tap 
homebrew/python?


Best, Bas

[Kwant] Unable to update to Tinyarray 1.0.5

[Bas Nijholt]

Hi, 


I’ve installed kwant through Homebrew, Michaels provided method.


Now I’m trying to update Tinyarray through pip and get the following message:


Bass-MacBook-Pro:Desktop basnijholt$ pip install tinyarray
Requirement already satisfied (use --upgrade to upgrade): tinyarray in 
/usr/local/lib/python2.7/site-packages
Bass-MacBook-Pro:Desktop basnijholt$ pip install tinyarray --upgrade
Collecting tinyarray from 
https://pypi.python.org/packages/source/t/tinyarray/tinyarray-1.0.5.tar.gz#md5=0ef161e8b5ae711fe3e1babb8702105e
  Using cached tinyarray-1.0.5.tar.gz
Installing collected packages: tinyarray
  Found existing installation: tinyarray 1.0.2
    Can't uninstall 'tinyarray'. No files were found to uninstall.
  Running setup.py install for tinyarray
    building 'tinyarray' extension
    clang -fno-strict-aliasing -fno-common -dynamic -I/usr/local/include 
-I/usr/local/opt/sqlite/include -isysroot 
/Applications/Xcode.app/Contents/Developer/Platforms/MacOSX.platform/Developer/SDKs/MacOSX10.10.sdk
 
-I/Applications/Xcode.app/Contents/Developer/Platforms/MacOSX.platform/Developer/SDKs/MacOSX10.10.sdk/System/Library/Frameworks/Tk.framework/Versions/8.5/Headers
 -DNDEBUG -g -fwrapv -O3 -Wall -Wstrict-prototypes 
-I/usr/local/Cellar/python/2.7.9/Frameworks/Python.framework/Versions/2.7/include/python2.7
 -c src/arithmetic.cc -o build/temp.macosx-10.10-x86_64-2.7/src/arithmetic.o
    clang -fno-strict-aliasing -fno-common -dynamic -I/usr/local/include 
-I/usr/local/opt/sqlite/include -isysroot 
/Applications/Xcode.app/Contents/Developer/Platforms/MacOSX.platform/Developer/SDKs/MacOSX10.10.sdk
 
-I/Applications/Xcode.app/Contents/Developer/Platforms/MacOSX.platform/Developer/SDKs/MacOSX10.10.sdk/System/Library/Frameworks/Tk.framework/Versions/8.5/Headers
 -DNDEBUG -g -fwrapv -O3 -Wall -Wstrict-prototypes 
-I/usr/local/Cellar/python/2.7.9/Frameworks/Python.framework/Versions/2.7/include/python2.7
 -c src/array.cc -o build/temp.macosx-10.10-x86_64-2.7/src/array.o
    clang -fno-strict-aliasing -fno-common -dynamic -I/usr/local/include 
-I/usr/local/opt/sqlite/include -isysroot 
/Applications/Xcode.app/Contents/Developer/Platforms/MacOSX.platform/Developer/SDKs/MacOSX10.10.sdk
 
-I/Applications/Xcode.app/Contents/Developer/Platforms/MacOSX.platform/Developer/SDKs/MacOSX10.10.sdk/System/Library/Frameworks/Tk.framework/Versions/8.5/Headers
 -DNDEBUG -g -fwrapv -O3 -Wall -Wstrict-prototypes 
-I/usr/local/Cellar/python/2.7.9/Frameworks/Python.framework/Versions/2.7/include/python2.7
 -c src/functions.cc -o build/temp.macosx-10.10-x86_64-2.7/src/functions.o
    clang++ -bundle -undefined dynamic_lookup -L/usr/local/lib 
-L/usr/local/opt/sqlite/lib -isysroot 
/Applications/Xcode.app/Contents/Developer/Platforms/MacOSX.platform/Developer/SDKs/MacOSX10.10.sdk
 build/temp.macosx-10.10-x86_64-2.7/src/arithmetic.o 
build/temp.macosx-10.10-x86_64-2.7/src/array.o 
build/temp.macosx-10.10-x86_64-2.7/src/functions.o -o 
build/lib.macosx-10.10-x86_64-2.7/tinyarray.so
Successfully installed tinyarray-1.0.2
Bass-MacBook-Pro:Desktop basnijholt$ 






Regards, 


Bas