MartinsM - Personal blog

How to install COBRApy in Lubuntu 15.10

2016-02-06T16:00:00+02:00

sudo su
aptitude install python-pip
aptitude install python-dev

Install libSBML

The next step failed for me, but one can try:

pip install python-libsbml

This method worked for me:

sh -c "echo 'deb http://download.opensuse.org/repositories/home:/fbergman:/libsbml/xUbuntu_15.10/ /' >> /etc/apt/sources.list.d/python-libsbml.list"
apt-get update
apt-get install python-libsbml

Install COBRApy dependencies

apt-get install python-glpk
apt-get install python-numpy
apt-get install python-scipy
apt-get install glpk-utils
apt-get install libglpk-dev

Install COBRApy

pip install cobra
apt-get install ipython

Test COBRApy

In python console run:

from cobra.test import test_all
test_all()

Ran 83 tests in 66.987s
OK (skipped=1)

As it is noted in the title these instructions work on Ubuntu. Unfortunately I couldn't get COBRApy to pass tests on Fedora 23 because it can't locate glpk. Any help in comments is appreciated.

Python Pandas reindex and new index

2016-01-19T00:55:00+02:00

The answer comes from stackoverflow.com:

So if you have a DataFrame with index [0, 1, 2], then doing a reindex([2, 1, 0]) will return the rows in reverse order. Doing something like reindex([8, 9, 10]) does not make a new index for the rows; rather, it will return a DataFrame with NaN values, since there are no rows with indices 8, 9, or 10.

To generate new index for existing rows try this:

df.index = range(len(df))

How to manage printers in Fedora

2016-01-15T09:00:00+02:00

In browser open http://localhost:631

Fedora 23 long boot time

2016-01-11T18:00:00+02:00

First rows of systemd-analyze blame reveal that mlocate-updatedb.service takes more than one minute during boot process:

1min 34.881s mlocate-updatedb.service
1min 13.159s plymouth-quit-wait.service
     13.293s firewalld.service
      8.945s dev-sda5.device
      8.904s systemd-udev-settle.service

I found the solution on stackexchange.com. The first answer by slm suggests to simply remove mlocate. Quoting the answer:

First find out where the program is located on disk.
```
$ type updatedb
updatedb is /usr/bin/updatedb
```

Next find out what package provides updatedb.

$ rpm -qf /usr/bin/updatedb
mlocate-0.26-3.fc19.x86_64

See if anything requires mlocate.

$ rpm -q --whatrequires mlocate
no package requires mlocate

Nothing requires it so you can remove the package.
```
$ dnf remove mlocate
```

I followed the instructions and now my Fedora 23 boots faster:

21.899s plymouth-quit-wait.service
10.138s systemd-udev-settle.service
 7.866s dev-sda5.device
 5.571s firewalld.service
 4.935s systemd-journald.service

How to install LaTeX and XeLaTeX in Fedora 23

2016-01-10T18:00:00+02:00

XeLaTeX requires a number packages. First become root, then run this command:

dnf -y install texlive
    texlive-latex
    texlive-xetex
    texlive-collection-latex
    texlive-collection-latexrecommended
    texlive-xetex-def
    texlive-collection-xetex

Total download size: 182 M

Installed size: 307 M

Only if needed:

dnf -y install texlive-collection-latexextra

Use gi.require_version('Gtk', '3.0') before import

2015-12-21T09:30:00+02:00

On recent Fedora importing Gtk the following way

from gi.repository import Gtk

gives the following error:

./hclean.py:27: PyGIWarning: Gtk was imported without specifying a version first.
Use gi.require_version('Gtk', '3.0') before import to ensure that the right version gets loaded.
from gi.repository import Gtk

The error is fixed by requiring the Gtk version:

import gi
gi.require_version("Gtk", "3.0")
from gi.repository import Gtk

Now the program starts without errors.

Solution found on programcreek.com.

How to disable SElinux on Fedora 23

2015-12-15T10:00:00+02:00

Use the getenforce or sestatus commands to check the status of SELinux.

On systems with SELinux disabled, the SELINUX=disabled option is configured in /etc/selinux/config

As root open the file /etc/selinux/config

su -
gedit /etc/selinux/config

Change the line SELINUX=enforcing to SELINUX=disabled

Reboot your system. After reboot, confirm that the getenforce command returns Disabled.

Sources: docs.fedoraproject.org.

Things to do after fresh Fedora 23 installation

2015-12-02T13:51:00+02:00

Become root and run update

su -
dnf update

RPM Fusion repository

http://rpmfusion.org/Configuration

Install the missing software

dnf install gnome-tweak-tool gimp inkscape thunderbird bluefish filezilla vlc python-openoffice epiphany

dnf install VirtualBox
dnf install dconf-editor

Software for development

dnf groupinstall "Development Tools"

Scientific software

dnf install numpy scipy python-matplotlib python-pandas python-ipython

Upgrade SciPy

dnf install python-pip blas-devel lapack-devel gcc-c++ rpm-build
dnf groupinstall "Development Tools"
pip install --upgrade scipy

Install scikit-learn

pip install -U scikit-learn

How to upgrade SciPy on Fedora 23

2015-10-19T12:58:00+03:00

As it was with Fedora 22, the Fedora 23 still ships with SciPy version 0.14.1. To upgrade SciPy please run the following commands as root:

dnf install python-pip
dnf install blas-devel
dnf install lapack-devel
dnf install gcc-c++
dnf groupinstall "Development Tools"

If you now try to upgrade scipy you will get the following error

g++: error: /usr/lib/rpm/redhat/redhat-hardened-cc1: No such file or directory

On Fedora 23 one has also to install rpm-build to avoid errors during compilation of scipy:

dnf search rpm-build

Finally uninstall scipy 0.14.1, download 0.16.0, compile it and install. All with the following command:

pip install --upgrade scipy

How to play encrypted DVDs on Fedora 22

2015-09-08T19:27:00+03:00

Solution found here: http://forums.fedoraforum.org/showthread.php?t=304745

First become root:

su -

Then execute these 7 commands:

dnf install dh-autoreconf
git clone git://git.videolan.org/libdvdcss
cd libdvdcss/
autoreconf -i
./configure --prefix=/usr
make
make install

LibreOffice: Readonly content cannot be changed

2013-09-19T10:41:00+03:00

By default Libre/Open Office will not allow you to delete form fields in Microsoft .doc or .docx files. When you try to get rid of form fields, you will get an error message like one below.

People have been annoyed (including me) about this anti feature of protecting MS Word documents since the early versions of OpenOffice.org. Many internet forums and discussions advise to check filesystem permissions or to enable document editing using "Edit File" button.

However, the problem is not with editing the text. It is possible to change formatting, add & delete texts unless you try to mess with form elements. Input fields can not be moved, resized or deleted. Try to select and delete a checkbox or input field and you will get the infamous "Readonly content cannot be changed" error box. Also, enabling Form Controls doesn't help.

The solution

The solution is called "Cursor in Protected areas" and is hidden in "Formatting Aids" options. Go to Main menu -> Tools -> Options.

Under LibreOffice.org Writer, choose "Formatting Aids" and uncheck the "enable" box for "Cursor in Protected areas". Once you do that, you will be able to delete form fields.

Solution found in: Nilesh Prajapati's blog. Read also how to fix Mendeley plugin for LibreOffice.

Using Android phone without Google account

2013-09-16T11:31:00+03:00

If you have ever wondered how usable Android phone is without having Google account, most probably you have already discovered the ugly truth - Android phone is a trap. There is no visible chance to have a smartphone (with apps and so on) without having Gmail account. Majority of Alternative Markets, like GetJar, are available only from Google Play. Sadly, most developers publish their Android apps only in Play store. You can't download or buy their apps from developers websites. You can't even download APK files to your computer using fake gmail account. And when you give up and enter your Gmail credentials in your Android phone there is no turning back - you can't just disconnect your phone from Google.

In this article I share my experience and guidelines on using Android phone avoiding Google Play store.

Step one (reinstall) Root your phone and get rid of the original firmware! I would recommend to install the fully free Android distribution Replicant from replicant.us. However the list of devices they support is quite short. My HTC Desire is not among supported devices so I have not tried it. For the rest there is Cyanogenmod which provides firmware from many Android phones, including my HTC Desire.

Step two (don't spoil it) On Cyanogenmod (and any new Android phone) during the first bootup, you will be asked to enter Gmail credentials. NEVER EVER do it. Do not enter false email, just skip the process. Earlier releases of Cyanogenmod shipped without Google-essential apps, and if one wanted them there was a method how to install them. The latest Cyanogenmod ships with built in GMail, Market, Talk, Google search and Car Home. It is impossible to uninstall these apps. The only thing you can do is to never configure your gmail account in your phone.

Step three (install apps) Using computer or the built-in web browser go to f-droid.org. Download and install F-Droid. It is a market-like app containing only free, open source Android apps. Here is the list of apps that I have found to be useful:

Web browser Zirco (because you can't trust the built-in browser);
Firewall DroidWall. There is also the newer AFWall+ which is based on DroidWall but it was not working on my phone;
Maps & Navigation app OsmAnd. You can get the latest version from their own server, but it is also on F-Droid;
RSS reader Sparse rss;
GPS tracking app Aripuca GPS Tracker. This is the only GPS tracking app that works properly on my phone and is available outside of Google PLay. I have also tried GPSLogger and OSMTracker, but no luck with them (keeps crashing and rebooting my phone after several minutes of working);
Tor (anonymity) client Orbot.
File synchronization ownCloud. On Google Play you have to pay for this app. In F-Droid it is for free;

What is not working I still have not found a solution to synchronize phone address book and calendar with my ownCloud server.

htaccess allow IP addresses

2013-09-10T15:52:00+03:00

If your website is under construction, you can simply edit .htaccess file to disallow visits to your website. Here is the code :

&lt;Limit GET POST PUT&gt;
order deny,allow
deny from all
allow from 12.245.67.190
&lt;/Limit&gt;

allow from 12.245.67.190 command allows access to your server only from 12.245.67.190 IP address.

You can allow more IP’s to access your website : <Limit GET POST PUT> order deny,allow deny from all allow from 12.245.67.190 allow from 12.245.67.192 </Limit> This practice is good in case you are modifying your website (performing a new joomla installation) and don’t want others to see the process.

Original source: http://htaccessfile.com/htaccess-allow-ip-addresses

Running Mendeley Desktop in Fedora 19

2013-09-05T11:26:00+03:00

By default Mendeley Desktop refuses to work. For me the solution was to install QtWebKit and QtSvg packages.

yum -y install qtwebkit
yum -y install qt5-qtsvg

How to add manual entries to Gnome Shell Launcher

2013-06-01T14:45:00+03:00

With gnome-shell, you'll need to create a .desktop file in '~/.local/share/applications/' which will then allow you to add the application to the gnome-shell dash.

My .desktop file for Sublime Text looks like this:

#!/usr/bin/env xdg-open
[Desktop Entry]
Version=1.0
Type=Application
Name=Sublime Text
GenericName=Text Editor
Comment=Sophisticated text editor for code, markup and prose
#Exec=/opt/sublime_text/sublime_text %F
Exec=/home/username/workspace/sublime_text_3/sublime_text %F
Terminal=false
MimeType=text/plain;
Icon=/home/username/workspace/sublime_text_3/Icon/48x48/sublime-text.png
Categories=TextEditor;Development;
StartupNotify=true
Actions=Window;Document;
[Desktop Action Window]
Name=New Window
Exec=/home/username/workspace/sublime_text_3/sublime_text -n
[Desktop Action Document]
Name=New File
Exec=/home/username/workspace/sublime_text_3/sublime_text --command new_file

Original source here.

How to download a complete website

2013-05-28T09:43:00+03:00

If you ever need to download an entire Web site, wget can do the job:

wget \
     --recursive \
     --no-clobber \
     --page-requisites \
     --adjust-extension \
     --convert-links \
     --restrict-file-names=windows \
     --domains example.com \
     --no-parent http://example.com.lv/thefolder/

The options are: --recursivedownload the entire Web site.

--domains example.com don't follow links outside example.com.

--no-parent don't follow links outside the directory thefolder/.

--page-requisites get all the elements that compose the page (images, CSS and so on).

--html-extension save files with the .html extension.

--convert-links convert links so that they work locally, off-line.

--restrict-file-names=windows modify filenames so that they will work in Windows as well.

--no-clobber don't overwrite any existing files (used in case the download is interrupted and resumed).

Also note that in wget 1.12, –html-extension is renamed to –adjust-extension

The orignal resource: http://www.linuxjournal.com/...

Mendeley's LibreOffice Plugin

2013-05-14T22:29:00+03:00

Got this error:

(com.sun.start.lang.DisposedException) {{{Message="Binary URP bridge disposed during call", Context=(com.sun.star.uno.XInterface) @7f9191447838}}}

I solved this by installing Python support for LibreOffice.

yum install python-openoffice

gtk import Segmentation fault

2013-04-30T17:19:00+03:00

When I tried to run my program (ModeRator) I get this error:

/usr/lib64/python2.7/site-packages/gobject/constants.py:24:
Warning: g_boxed_type_register_static: assertion `g_type_from_name (name) == 0' failed
  import gobject._gobject
/usr/lib64/python2.7/site-packages/gtk-2.0/gtk/__init__.py:40:
Warning: specified class size for type `PyGtkGenericCellRenderer' is smaller than the parent type's `GtkCellRenderer' class size
  from gtk import _gtk
/usr/lib64/python2.7/site-packages/gtk-2.0/gtk/__init__.py:40:
Warning: g_type_get_qdata: assertion `node != NULL' failed
  from gtk import _gtk
Segmentation fault (core dumped)

I 'fixed' the problem by disabling the import of matplotlib

It turns out that it is possible to import only one of them:

from gi.repository import GObject, Gtk

import matplotlib.pyplot as plt

Comment out one of these lines end the error is gone! Better explanation of problems like this is found here: http://blog.rabbitvcs.org/archives/312.

Gtk color names

2013-04-16T21:17:00+03:00

To use color names (not codes or objects) in Gtk GUI applications, one have use X11 color names.

There is no dedicated list of Gtk/Gdk/Gnome color names. GTK is using X11 color names.

Enable Dell Latitude WiFi on Debian

2013-04-08T11:28:00+03:00

I partly followed instructions on this page: http://wiki.debian.org/wl.

In terminal (as root):

aptitude install broadcom-sta-dkms
modprobe -r b44 b43 b43legacy ssb brcmsmac
modprobe wl

Run this command to verify:

iwconfig

How to create an SBML file using libSBML in Python

2012-09-25T16:31:00+03:00

This tutorial is intended as a quick introduction on how to create an SBML file using libSBML. For a full developer's guide, please see libSBML Python API has no title attribute.

from libsbml import *
document = SBMLDocument()
model = document.createModel(2, 4) # SBML level and version

First you have to create compartments

c1 = model.createCompartment()
c1.setName('Forest')
c1.setId('comp_FOREST')

Set the volume of compartments

c1.setVolume(1000)

Continue with other compartments

c2 = model.createCompartment()
c2.setName('Meadow')
c2.setId('comp_Meadow')
c2.setVolume(1000)

When you have finished with compartments, proceed with species.

s1 = model.createSpecies()
s1.setName('Foxes')
s1.setId('spec_FOXES')
s1.setCompartment('comp_FOREST')
s1.setInitialAmount(10)

s2 = model.createSpecies()
s2.setName('Rabbits')
s2.setId('spec_RABBITS')
s2.setCompartment('comp_FOREST')
s2.setInitialAmount(50)

s3 = model.createSpecies()
s3.setName('Snakes')
s3.setId('spec_Snakes')
s3.setCompartment('comp_Meadow')
s3.setInitialAmount(100)

When you are done with compartments and species, create reactions.

r1 = model.createReaction()
r1.setName('Foxes eat rabbits')
r1.setId('R1')

Create reactants and products

reac1 = r1.createReactant()
reac1.setSpecies('spec_RABBITS')
prod1 = r1.createProduct()
prod1.setSpecies('spec_FOXES')

if necessary, set Stoichiometry

reac1.setStoichiometry(2) # 2 rabbits
prod1.setStoichiometry(1) # 1 fox

Notice that we create seemingly independent objects for comprartments, reaction, reactant and product. We set some values to the properties of these objects and never actually add these elements to the model. Thats because they are automatically added. You can try this inside a loop and it will still work.

Now when you have done whith your model you have to add it to the document. document.setModel(model)

If you wish you can add optional information.

model.setName('My test model')
model.setId('MyModelID')

Save to file.

writeSBMLToFile(document,'name_of_your_model.xml')

That's it! Copy&paste this code and try it yourself! Please note that this tutorial is really basic. For more comprehensive examples, please see the examples/ folder in the libSBML source tree.

How to install libSBML in Ubuntu 12.04 with Python bindings

2012-01-13T12:41:00+02:00

Install compiler

apt-get install build-essential

On a fresh ubuntu 12.04 setup you will need to install some additional packages

apt-get install python2.7-dev
apt-get install libxml2-dev
apt-get install swig

Compile and install libSBML

Go to the directory where you have exctracted libsbml source and run these 3 commands

./configure --with-python
make
make install