Running Python

To start using Python, we need access to the python program in a terminal. It is recommended to use Anaconda and its environments whenever possible.

Alternatively, you can directly use a version on lxplus. The version installed on lxplus is 2.7.15, which is old. However, we can get a newer version along with various useful packages (see details on the LCG stacks) by setting up an LCG environment:

$ source /cvmfs/sft.cern.ch/lcg/views/setupViews.sh LCG_94python3 x86_64-slc6-gcc62-opt

If you have a computer running MacOS or some Linux distribution, it will have come will Python pre-installed. Either way, a simple way to get Python on your computer is to install miniforge (miniconda with the conda-forge channel as default). In any case, it is highly recommended to use a virtual environment, be it with conda or other similar packages, and not your system installation.

Python 2 or 3?

You might see material that talks about Python 3 or Python 2. Like a lot of other software, Python is regularly updated and groups batches of updates, including bug fixes and new features, into versions. Python 3 is a new major release and is not backwards compatible with Python 2, which is an exceptional break and not intended to happen again with Python 4. This means that code that works when run with version 2 of Python may not necessarily work when run with version 3. Python 2 was around for a long time and the process of migrating to Python 3 has been slow. However the support for Python 2 will soon end (2 month, in 2020!) and with it, most packages completely drop the support or already have.

Note that, for instance, Alice and LHCb software is not (yet) fully compatible with Python 3. We strongly encourage you to always use Python 3 and only switch to Python 2 when you have to (e.g. when you need to use some software which is not compatible with Python 3). You can install both Python 2 and Python 3 using [conda][miniforge.

Python is a very user-friendly language. If you’re used to having to compile your code, this might seem refreshing:

$python
Python 3.6.3 (default, Dec 19 2017, 22:45:30) 
[GCC 6.2.0] on linux
Type "help", "copyright", "credits" or "license" for more information.
>>> a = 3.14
>>> print(a + 1.1)
4.24

Woah! What just happened?

  1. We started an interactive Python session, also known as a Python shell, by executing the python command;

  2. We typed a line of code, a = 3.14, and hit enter. this let the Python interpreter compile and execute this line of code. Here, the value 3.14 was assigned to the variable a;

  3. We typed another line of code, print(a + 1.1), and hit enter;

  4. The value 4.14 was printed to the terminal.

In this interactive shell, if a value is not assigned to a variable, it will automatically be printed. This is a special feature of the interactive shell and not general Python behavior.

So

>>> a
3.14
>>> a + 1
4.14

directly prints the result.

This interactive session is sometimes called a REPL: a Read Evaluate Print Loop. This is similar to bash, where you type your command, run it, see the results, and can then type the next line. Sometimes there are no results, so you don’t see anything in bash being printed (just like running the true command in bash); this, as seen above, is not always true for an interactive Python shell and a convenience feature.

You can leave the session by running exit(), or by using the Ctrl-d key combination.

Everything that can be done in Python can be done in an interactive session; it’s a great way to experiment. An enhanced version of this session is called IPython.

$ipython
Python 3.6.3 (default, Dec 19 2017, 22:45:30) 
Type "copyright", "credits" or "license" for more information.

IPython 5.4.1 -- An enhanced Interactive Python.
?         -> Introduction and overview of IPython's features.
%quickref -> Quick reference.
help      -> Python's own help system.
object?   -> Details about 'object', use 'object??' for extra details.

In [1]: print(1 + 3)
4

There are a few advantages to using ipython over python:

  • Command history persists across sessions. This also works just like bash: hit the up arrow to go through lines you typed in the past. If you already have part of a command typed out, and then hit the up arrow, IPython will only show you lines that started with the same characters.

  • Autocompletion. If you hit the tab key whilst typing something, IPython will present you with a list of things that match the word you’re in the middle of typing.

In [2]: abc_my_var = 3.14

In [3]: abc_<tab>

In [4]: import math

In [5]: math.s<tab>
              math.sin
              math.sinh
              math.sqrt

  • Easily run shell commands by starting your line with an exclamation mark (again, this is now an IPython feature and has nothing to do with Python!).

In [6]: !cal
    October 2017
Su Mo Tu We Th Fr Sa
 1  2  3  4  5  6  7
 8  9 10 11 12 13 14
15 16 17 18 19 20 21
22 23 24 25 26 27 28
29 30 31

Your best friend in a (I)Python shell is the help method. If you want more information on something, just ask for help!

In [7]: help()

In [8]: help(math)

In general, this command line utilities can be used to play around or do simply a calculation. For coding, other tools are used that write scripts, which are essentially a text file (with the .py extension), that is line-by-line interpreted by Python.

A usefule tool in between these two worlds are Jupyter Notebooks that have cells (code blocks) that are executed.

In the next few chapters, we will start with the basics of Python to get a feeling for what can be done.