.. _tutorial_fortran:

==============================================
Load a 32-bit FORTRAN library in 64-bit Python
==============================================

This example shows how to access a 32-bit FORTRAN library from 64-bit Python.
:class:`~msl.examples.loadlib.fortran32.Fortran32` is the 32-bit server and
:class:`~msl.examples.loadlib.fortran64.Fortran64` is the 64-bit client.
The source code of the FORTRAN program is available :ref:`here <fortran-lib>`.

.. note::
   If you have issues running the example please make sure that you have the
   :ref:`prerequisites <loadlib-prerequisites>` installed for your operating system.

.. important::
   By default :mod:`ctypes` expects that a :class:`ctypes.c_int` data type is
   returned from the library call. If the returned value from the library is
   not a :class:`ctypes.c_int` then you must redefine the ctypes
   :ref:`restype <python:ctypes-return-types>` value to be the appropriate data
   type. The :class:`~msl.examples.loadlib.fortran32.Fortran32` class shows various
   examples of redefining the :ref:`restype <python:ctypes-return-types>` value.

Create a :class:`~msl.examples.loadlib.fortran64.Fortran64` client to communicate
with the 32-bit :ref:`fortran_lib32 <fortran-lib>` library

.. invisible-code-block: pycon

   >>> SKIP_IF_MACOS()

.. code-block:: pycon

   >>> from msl.examples.loadlib import Fortran64
   >>> f = Fortran64()

Add two ``int8`` values, see
:meth:`~msl.examples.loadlib.fortran64.Fortran64.sum_8bit`

.. code-block:: pycon

   >>> f.sum_8bit(-50, 110)
   60

Add two ``int16`` values, see
:meth:`~msl.examples.loadlib.fortran64.Fortran64.sum_16bit`

.. code-block:: pycon

   >>> f.sum_16bit(2**15-1, -1)
   32766

Add two ``int32`` values, see
:meth:`~msl.examples.loadlib.fortran64.Fortran64.sum_32bit`

.. code-block:: pycon

   >>> f.sum_32bit(123456788, 1)
   123456789

Add two ``int64`` values, see
:meth:`~msl.examples.loadlib.fortran64.Fortran64.sum_64bit`

.. code-block:: pycon

   >>> f.sum_64bit(-2**63, 1)
   -9223372036854775807...

Multiply two ``float32`` values, see
:meth:`~msl.examples.loadlib.fortran64.Fortran64.multiply_float32`

.. code-block:: pycon

   >>> f.multiply_float32(1e30, 2e3)
   1.99999998899...e+33

Multiply two ``float64`` values, see
:meth:`~msl.examples.loadlib.fortran64.Fortran64.multiply_float64`

.. code-block:: pycon

   >>> f.multiply_float64(1e30, 2e3)
   2.00000000000...e+33

Check if a value is positive, see
:meth:`~msl.examples.loadlib.fortran64.Fortran64.is_positive`

.. code-block:: pycon

   >>> f.is_positive(1e-100)
   True
   >>> f.is_positive(-1e-100)
   False

Add or subtract two integers, see
:meth:`~msl.examples.loadlib.fortran64.Fortran64.add_or_subtract`

.. code-block:: pycon

   >>> f.add_or_subtract(1000, 2000, True)
   3000
   >>> f.add_or_subtract(1000, 2000, False)
   -1000

Calculate the n'th factorial, see
:meth:`~msl.examples.loadlib.fortran64.Fortran64.factorial`

.. code-block:: pycon

   >>> f.factorial(0)
   1.0
   >>> f.factorial(127)
   3.012660018457658e+213

Calculate the standard deviation of an list of values, see
:meth:`~msl.examples.loadlib.fortran64.Fortran64.standard_deviation`

.. code-block:: pycon

   >>> f.standard_deviation([float(val) for val in range(1,10)])
   2.73861278752583...

Compute the Bessel function of the first kind of order 0, see
:meth:`~msl.examples.loadlib.fortran64.Fortran64.besselJ0`

.. code-block:: pycon

   >>> f.besselJ0(8.6)
   0.0146229912787412...

Reverse a string, see
:meth:`~msl.examples.loadlib.fortran64.Fortran64.reverse_string`

.. code-block:: pycon

   >>> f.reverse_string('hello world!')
   '!dlrow olleh'

Add two 1D arrays, see
:meth:`~msl.examples.loadlib.fortran64.Fortran64.add_1d_arrays`

.. code-block:: pycon

   >>> a = [float(val) for val in range(1, 10)]
   >>> b = [0.5*val for val in range(1, 10)]
   >>> a
   [1.0, 2.0, 3.0, 4.0, 5.0, 6.0, 7.0, 8.0, 9.0]
   >>> b
   [0.5, 1.0, 1.5, 2.0, 2.5, 3.0, 3.5, 4.0, 4.5]
   >>> f.add_1d_arrays(a, b)
   [1.5, 3.0, 4.5, 6.0, 7.5, 9.0, 10.5, 12.0, 13.5]

Multiply two matrices, see
:meth:`~msl.examples.loadlib.fortran64.Fortran64.matrix_multiply`

.. code-block:: pycon

   >>> m1 = [[1, 2, 3], [4, 5, 6]]
   >>> m2 = [[1, 2], [3, 4], [5, 6]]
   >>> f.matrix_multiply(m1, m2)
   [[22.0, 28.0], [49.0, 64.0]]

Shutdown the 32-bit server when you are done communicating with the 32-bit library

.. code-block:: pycon

   >>> stdout, stderr = f.shutdown_server32()