Agroalimentary (2 points)
=========================

From the Catalan Government web we have downloaded the file
:download:`agroalimentary.csv` related to Catalan agroalimentary
companies, that you already know. We have extended it with a randomly
generated column named ``Facturacio`` showing the company billing in
thousands of euros (:py:class:`int`). Download and study this file. We
have already created a :py:ref:`DataFrame <basics.dataframe>` with
this data.

The following example shows the names of all the columns of this
:py:ref:`DataFrame <basics.dataframe>`, then a subdataframe example
with columns corresponding to the company name (``Establiment``), the
region (``Comarca``), the company type (``Titularitat``) and the
billing (``Facturacio``), and, below, the same subdataframe with
colums ``Establiment`` and ``Sector``:

.. literalinclude:: test-agroalim1.txt
   :language: python
   :lines: 4-31
	  
Save the two following functions to file ``agroalim.py``.

#. Write function :py:func:`agroalim1` that takes a :py:ref:`DataFrame
   <basics.dataframe>`, ``dfagro``, as that described and a string
   (:py:class:`str`) corresponding to a region, ``'Comarca'``, and
   returns a dictionary (:py:class:`dict`) in which keys are company
   types, ``'Titularitat'`` (:py:class:`str`), and values are the
   number of companies of this type, corresponding to the given region
   (1 point). Examples:
   
   .. literalinclude:: test-agroalim1.txt
      :language: python
      :lines: 36-41

   .. note:: You have more tests in file :download:`test-agroalim1.txt`
	   
   .. hint:: Use selection, methods ``groupby`` and ``size`` and function ``dict``.
   
#. Write function :py:func:`agroalim2` that takes a :py:ref:`DataFrame
   <basics.dataframe>`, ``dfagro``, as that described, and two values
   (:py:class:`int`) corresponding to two billings, :math:`b1` and
   :math:`b2`, such that :math:`b1<b2`. This function considers the
   subset of companies with a billing :math:`b` such that :math:`b1
   \leq b \leq b2` and returns a tuple with two values: the name of
   the sector , ``'Sector'``, (:py:class:`str`) for which the billing
   sum of the companies of this subset corresponding to this sector is
   minimum, and the corresponding minimum billing sum
   (:py:class:`int`) (1 point). Examples:
   
   .. literalinclude:: test-agroalim2.txt
      :language: python
      :lines: 6-9

   .. note:: You have more tests in file :download:`test-agroalim2.txt`

   .. hint:: Use selection and methods ``groupby``, ``sum``, ``idxmin`` and ``min``.


.. note:: To run your functions you don't actually need to read the given data file. But you need to download and save it into your working folder because the given testfiles read from it to create the :py:ref:`DataFrame <basics.dataframe>`.