wordless/wl_figs/wl_figs.py
# ----------------------------------------------------------------------
# Wordless: Figures - Figures
# Copyright (C) 2018-2024 Ye Lei (叶磊)
#
# This program is free software: you can redistribute it and/or modify
# it under the terms of the GNU General Public License as published by
# the Free Software Foundation, either version 3 of the License, or
# (at your option) any later version.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License
# along with this program. If not, see <http://www.gnu.org/licenses/>.
# ----------------------------------------------------------------------
import re
import PIL.Image
from PyQt5.QtCore import QCoreApplication
from PyQt5.QtWidgets import QDesktopWidget
import matplotlib
import matplotlib.pyplot
import networkx
import numpy
import wordcloud
from wordless.wl_utils import wl_paths, wl_misc
_tr = QCoreApplication.translate
# Restore Matplotlib's default settings
def restore_matplotlib_rcparams():
matplotlib.pyplot.rcParams['savefig.facecolor'] = 'auto'
def get_data_ranks(data_files_items, fig_settings):
if fig_settings['rank_min_no_limit']:
rank_min = 1
else:
rank_min = fig_settings['rank_min']
if fig_settings['rank_max_no_limit']:
rank_max = None
else:
rank_max = fig_settings['rank_max']
return data_files_items[rank_min - 1 : rank_max]
def generate_line_chart(
main,
data_files_items, fig_settings,
file_names_selected, label_x
): # pylint: disable=unused-argument
restore_matplotlib_rcparams()
data_files_items = get_data_ranks(data_files_items, fig_settings)
items = [item for item, vals in data_files_items]
vals = numpy.array([vals for item, vals in data_files_items])
# Frequency data
if fig_settings['use_data'] == _tr('wl_figs', 'Frequency') or re.search(_tr('wl_figs', r'^[LR][1-9][0-9]*$'), fig_settings['use_data']):
if fig_settings['use_cumulative']:
vals = numpy.cumsum(vals, axis = 0)
if fig_settings['use_pct']:
total_freqs = numpy.array([vals for item, vals in data_files_items]).sum(axis = 0)
for i, (file_name, total_freq) in enumerate(zip(file_names_selected, total_freqs)):
matplotlib.pyplot.plot(vals[:, i] / total_freq * 100, label = file_name)
else:
for i, file_name in enumerate(file_names_selected):
matplotlib.pyplot.plot(vals[:, i], label = file_name)
if fig_settings['use_cumulative']:
if fig_settings['use_pct']:
matplotlib.pyplot.ylabel(_tr('wl_figs', 'Cumulative Percentage Frequency'))
else:
matplotlib.pyplot.ylabel(_tr('wl_figs', 'Cumulative Frequency'))
else:
if fig_settings['use_pct']:
matplotlib.pyplot.ylabel(_tr('wl_figs', 'Percentage Frequency'))
else:
matplotlib.pyplot.ylabel(_tr('wl_figs', 'Frequency'))
# Non-frequency data
else:
for i, file_name in enumerate(file_names_selected):
matplotlib.pyplot.plot(vals[:, i], label = file_name)
matplotlib.pyplot.ylabel(fig_settings['use_data'])
matplotlib.pyplot.xlabel(label_x)
matplotlib.pyplot.xticks(
range(len(items)),
labels = items,
rotation = 90
)
matplotlib.pyplot.grid(True, color = 'silver')
matplotlib.pyplot.legend()
def generate_word_cloud(main, data_file_items, fig_settings):
restore_matplotlib_rcparams()
data_file_items = get_data_ranks(data_file_items, fig_settings)
items = [item for item, val in data_file_items]
# Convert to numpy.float64 to fix zeros
vals = numpy.array([val for item, val in data_file_items], dtype = numpy.float64)
val_min = numpy.min(vals[vals != -numpy.inf]) if vals[vals != -numpy.inf].size > 0 else -10
val_max = numpy.max(vals[vals != numpy.inf]) if vals[vals != numpy.inf].size > 0 else 10
# Fix +/-inf
vals = numpy.where(vals != numpy.inf, vals, val_max * 10)
vals = numpy.where(vals != -numpy.inf, vals, val_min * 10)
# WordCloud always displays data descendingly
if fig_settings['use_data'] == _tr('wl_figs', 'p-value'):
vals = 1 - vals
# Fix negative numbers
if vals[vals < 0].size > 0:
vals += (-numpy.min(vals)) + 1e-15
# Fix zeros
if vals[vals == 0].size > 0:
vals += 1e-15
settings = main.settings_custom['figs']['word_clouds']
desktop_widget = QDesktopWidget()
if settings['font_settings']['font'] == 'Droid Sans Mono':
font_path = wordcloud.wordcloud.FONT_PATH
elif settings['font_settings']['font'] == 'GNU Unifont':
font_path = wl_paths.get_path_data('unifont-15.1.02.otf')
elif settings['font_settings']['font'] == _tr('wl_figs', 'Custom'):
font_path = settings['font_settings']['font_path']
if settings['font_settings']['relative_scaling'] == -0.01:
relative_scaling = 'auto'
else:
relative_scaling = settings['font_settings']['relative_scaling']
if settings['font_settings']['font_color'] == _tr('wl_figs', 'Monochrome'):
color_func = lambda *args, **kwargs: settings['font_settings']['font_color_monochrome'] # pylint: disable=unnecessary-lambda-assignment
elif settings['font_settings']['font_color'] == _tr('wl_figs', 'Colormap'):
color_func = None
if settings['bg_settings']['bg_color_transparent']:
# Modify Matplotlib's global settings
matplotlib.pyplot.rcParams['savefig.facecolor'] = (0, 0, 0, 0)
bg_color = None
mode = 'RGBA'
else:
bg_color = settings['bg_settings']['bg_color']
mode = 'RGB'
if settings['mask_settings']['mask_settings']:
mask = numpy.array(PIL.Image.open(settings['mask_settings']['mask_path']))
else:
mask = None
word_cloud = wordcloud.WordCloud(
width = desktop_widget.width(),
height = desktop_widget.height(),
font_path = font_path,
min_font_size = settings['font_settings']['font_size_min'],
max_font_size = settings['font_settings']['font_size_max'],
relative_scaling = relative_scaling,
color_func = color_func,
colormap = settings['font_settings']['font_color_colormap'],
background_color = bg_color,
mode = mode,
mask = mask,
# Known issue: Error when background is transparent and contour width > 0
# Reference: https://github.com/amueller/word_cloud/issues/501
contour_width = settings['mask_settings']['contour_width'],
contour_color = settings['mask_settings']['contour_color'],
# The ratio of times to try horizontal fitting as opposed to vertical, ranging from 0 to 1 inclusive
prefer_horizontal = settings['advanced_settings']['prefer_hor'] / 100,
repeat = settings['advanced_settings']['allow_repeated_words']
)
word_cloud.generate_from_frequencies(dict(zip(items, vals)))
matplotlib.pyplot.imshow(word_cloud, interpolation = 'bilinear')
matplotlib.pyplot.axis('off')
def generate_network_graph(main, data_file_items, fig_settings):
restore_matplotlib_rcparams()
data_file_items = dict(get_data_ranks(data_file_items, fig_settings))
settings = main.settings_custom['figs']['network_graphs']
graph = networkx.MultiDiGraph()
graph.add_edges_from(data_file_items)
pos = settings['advanced_settings']['layout'](graph)
networkx.draw_networkx_nodes(
graph,
pos = pos,
node_shape = settings['node_settings']['node_shape'],
node_size = settings['node_settings']['node_size'],
node_color = settings['node_settings']['node_color'],
alpha = settings['node_settings']['node_opacity']
)
networkx.draw_networkx_labels(
graph,
pos = {node: (x, y + 0.05) for node, (x, y) in pos.items()},
font_family = settings['node_label_settings']['label_font'],
font_size = settings['node_label_settings']['label_font_size'],
font_weight = settings['node_label_settings']['label_font_weight'],
font_color = settings['node_label_settings']['label_font_color'],
alpha = settings['node_label_settings']['label_opacity']
)
if fig_settings['use_data'] == _tr('wl_figs', 'p-value'):
precision = main.settings_custom['tables']['precision_settings']['precision_p_vals']
reverse = True
else:
precision = main.settings_custom['tables']['precision_settings']['precision_decimals']
reverse = False
data_file_items = {
item: round(val, precision)
for item, val in data_file_items.items()
}
networkx.draw_networkx_edges(
graph,
pos = pos,
edgelist = data_file_items,
width = wl_misc.normalize_nums(
data_file_items.values(),
normalized_min = settings['edge_settings']['edge_width_min'],
normalized_max = settings['edge_settings']['edge_width_max'],
reverse = reverse
),
connectionstyle = settings['edge_settings']['connection_style'],
style = settings['edge_settings']['edge_style'],
edge_color = settings['edge_settings']['edge_color'],
alpha = settings['edge_settings']['edge_opacity'],
arrowstyle = settings['edge_settings']['arrow_style'],
arrowsize = settings['edge_settings']['arrow_size'],
# Used to determine edge positions
node_shape = settings['node_settings']['node_shape'],
node_size = settings['node_settings']['node_size']
)
networkx.draw_networkx_edge_labels(
graph,
pos = pos,
edge_labels = data_file_items,
label_pos = settings['edge_label_settings']['label_position'],
rotate = settings['edge_label_settings']['rotate_labels'],
font_family = settings['edge_label_settings']['label_font'],
font_size = settings['edge_label_settings']['label_font_size'],
font_weight = settings['edge_label_settings']['label_font_weight'],
font_color = settings['edge_label_settings']['label_font_color'],
alpha = settings['edge_label_settings']['label_opacity']
)
def show_fig():
matplotlib.pyplot.get_current_fig_manager().window.showMaximized()