#!/usr/bin/env python3
# Copyright (c) 2023 Intel Corporation
# SPDX-License-Identifier: 0BSD
"""
Customized Navigation Chart
===========================

A customized navigation chart.

In this example, we show how to customize a navigation chart by increasing the
number of axis ticks and by annotating one of the datapoints. Adjusting the
number of axis ticks can improve a reader's ability to discern between two
similar values, while annotations can be useful to draw attention to certain
points and/or provide some additional context.

Instead of trying to expose all possible customization options as arguments to
:py:func:`p3analysis.plot.navchart`, the function returns a
:py:class:`p3analysis.plot.NavChart` object that provides direct access to library
internals. When using the :py:mod:`matplotlib` backend it is possible to
access the :py:class:`matplotlib.axes.Axes` that were used and subsequently
call any number of :py:mod:`matplotlib` functions. In our example, we can
use :py:meth:`matplotlib.axes.Axes.set_xticks` and
:py:meth:`matplotlib.axes.Axes.set_yticks` to control the ticks, and can use
:py:meth:`matplotlib.axes.Axes.annotate` for annotations.

.. NOTE::
   :py:mod:`matplotlib` is currently the only backend supported by the P3
   Analysis Library, but this is subject to change.

.. TIP::
   If you have any trouble customizing a navigation chart, or the
   :py:class:`~p3analysis.plot.backend.NavChart` object does not provide access to the
   internals you are looking for, then please `open an issue
   <https://github.com/P3HPC/p3-analysis-library/issues/new/choose>`_.
"""

# Initialize synthetic data
# (not shown, but available in script download)
# sphinx_gallery_start_ignore
from collections import defaultdict

import matplotlib.pyplot as plt
import pandas as pd

import p3analysis

pp_data = defaultdict(list)
cd_data = defaultdict(list)
for data in [pp_data, cd_data]:
    data["problem"] += ["Synthetic"] * 5
    data["application"] += [
        "Unportable",
        "Ideal",
        "Per-Platform Source",
        "Portability Framework",
        "Specialized",
    ]
pp_data["app pp"] += [0, 1, 1, 0.5, 0.7]
cd_data["divergence"] += [1, 0, 1, 0, 0.3]
# sphinx_gallery_end_ignore

# Read performance portability and code divergence data into pandas DataFrame
pp = pd.DataFrame(pp_data)
cd = pd.DataFrame(cd_data)

# Generate a navigation chart with custom style options
legend = p3analysis.plot.Legend(loc="center left", bbox_to_anchor=(1.0, 0.5))
astyle = p3analysis.plot.ApplicationStyle(markers=["x", "*", "s", "o", "P"])
navchart = p3analysis.plot.navchart(pp, cd, size=(5, 5), legend=legend, style=astyle)

# Further customize the navigation chart using matplotlib
# In this example, we add a label and adjust the ticks
ax = navchart.get_axes()
ax.annotate(
    "Balances performance and code re-use.",
    xy=(0.7, 0.7),
    xytext=(0.2, 0.55),
    arrowprops=dict(facecolor='black', shrink=0.05),
)
ax.set_xticks([x * 0.1 for x in range(0, 11)])
ax.set_yticks([y * 0.1 for y in range(0, 11)])

navchart.save("customized-navchart.png")