(exercise2)=
# Exercise 2: Modifying a histogram 
**(15 minutes)**

You're probably still plotting your histograms as a solid curve. Most
of the time, your supervisor will want to see histograms with errors.
Figure out how to draw one of your histograms (e.g., **`ebeam`**)
with error bars. (Here's a {ref}`hint <working-with-histograms>`.)

:::{figure-md} ebeam-errors-rdf-fig
:class: align-center

<img src="ebeamHistogram.png" alt="ebeam histogram">

What I get when I plot `ebeam` with errors bars turned on.
:::

That was pretty easy once you looked at the hint. We'll take it up a notch:
Let's be good scientists and set/improve the axis labels; set the
x-axis label to "Beam energy [GeV]" and set the y-axis label to
"Number of events".

This is tricky to do "from scratch" without reading the ROOT documentation in detail as described
in {ref}`detective-work`, eventually reaching a 
description of the [TGAxis class](https://root.cern.ch/doc/master/classTGaxis.html). 
Fortunately, you've got a faster way, if you {ref}`remember an earlier
lesson <saving-c1>`.

:::{admonition} Need more help?
:class: hint
Here's another hint if the above hint is too opaque for you: While
`RDataFrame` is a useful tool, sometimes the easiest way to answer
a question is to go back to the basics.

Another thing to remember that if you can {ref}`use Draw on a histogram <rdf-make-histogram>`, there are
other ROOT histogram methods you can use on it.

Also remember that histogram names are arbitrary. Maybe someone else calls a histogram
something like **`hist__1__1`**, but you can call it **`histebeam`** or whatever you want.

One last tip for Python programmers: The `->` used in C++ becomes a simple `.` for you.
:::

:::::{admonition} Are you using the command line?
:class: note
If you're using the command line instead of the {ref}`notebook`, at this
point in the tutorial it's going to be more convenient to start editing
an external file and executing it from within your command
environment.

   - {command}`root`

     You may remember this from {ref}`saving-c1`: Create a file containing
     your C++ code, preferably with the extension `.C`; e.g.,
     `exercise2.C`. You'll want to set up your code as a C++ function;
     e.g.:

     :::{code-block} c++
     void exercise2()
     {
        // This is not the real code for Exercise 2!
        cout << "Hello, world!" << endl;
     }
     :::

     You can then execute this file from within {command}`root`  with:

         .x exercise2.C

   - {command}`ipython`

     If you're using [interactive
     Python](https://ipython.readthedocs.io/en/stable/) outside of a
     notebook, then you'll want to use [the `%run`
     command](https://ipython.readthedocs.io/en/stable/interactive/magics.html).
     Create a file containing your Python code, preferably with the
     extension `.py`; e.g., `exercise2.py`. You can then execute this file
     from within {command}`ipython` with:

         %run exercise2.py

:::{tip}
If you're exiting your command-line environment to edit a file, quitting your
editor, then starting up your environment again, there's a simpler way:
just start up another {ref}`Terminal window <terminal>`. Run your command-line
environment in one window and edit files in another.[^trite]

[^trite]: I apologize if this tip seems trite to you. But you might be surprised
    how many students have come from a {abbr}`GUI (Graphical User Interface)`
    environment that did not encourage them to open multiple windows at the 
    same time. 
:::

:::::