def classify_word_types(self, to_include):
"""
Classify and highlight specific word types.
"""
self.raw = self.text.get('1.0', tk.END)
# Get current text input.
wc = wd.WordSet(self.raw, self.md_core)
# Label word types.
flatten = [x for y in to_include for x in y]
# Get rid of tupples.
td = log.CodeBlockTimer(
'tfirst') # This loop ~0.01-0.05s per word! Need to speed up:
td.start()
for w, t in list(set(wc.pos)):
if t in flatten:
try:
colour = wc.word_colours[t]
self.highlight_words(w, wc, name=t, color=colour)
except KeyError:
pass
else:
pass
td.finish()
def similarity_of_highlighted_texts(self):
"""
Get the similarities of two sections of highlighted text.
"""
wc = wd.WordSet(self.raw, self.md_core)
self.text.config(cursor='@icons//highlighter_tip.cur')
# Change cursor for highlights.
if self.text.tag_ranges(tk.SEL):
h1 = self.tag_highlighted_text(colour='blue', tag_name='h1')
else:
self.wait_variable(self.text_selected)
h1 = self.tag_highlighted_text(colour='blue', tag_name='h1')
self.wait_variable(self.text_selected)
# Wait for next highlight.
h2 = self.tag_highlighted_text(colour='red', tag_name='h2')
self.text.config(cursor='arrow')
# Return cursor to arrow.
sim, color = wc.spacy_sim(h1, h2)
self.text_selected = tk.StringVar()
# Reset the text selection variable.
return sim
def test_wordcolour(self):
"""Test highlighting words in a given colour."""
log.log_setup()
# Start logging.
text = 'I was about to go shopping, but it was raining. So I decided to stay at home instead.'
app = gu.MainWindow(None, text)
app.title('Prose analysis')
app.grid_config()
ws = wc.WordSet(text)
app.highlight_words('was', ws)
app.mainloop()
def sentiment_analysis(self):
"""
Highlight positive and negative sentiment for all words or
all highlighted words.
"""
if self.highlighted_text_list:
# Highlight only selected words.
wc = wd.WordSet(self.raw, self.md_core)
for s, v in zip(self.highlighted_text_list.keys(),
self.highlighted_text_list.values()):
pos, neg, obj, color = wc.sentiment(s)
if color != '#000000':
self.colourise_text(s, 'snow', color, s, v[0])
# Don't highlight.
else:
# Highlight all words.
wc = wd.WordSet(self.raw, self.md_core)
sentiment_all_vals = wc.sentiment_all()
def similarity_to_all_highlighted(self):
"""
Get similarity of the highlighted word to all other
highlighted words.
"""
wc = wd.WordSet(self.raw, self.md_core)
for k, v in zip(self.highlighted_text_list.keys(),
self.highlighted_text_list.values()):
# Check to see which sentence is being highlighted.
s = ''.join(['' if i.isdigit() else i for i in k])
# Remove numeric elements for words appearing twice.
sim, color = wc.spacy_sim(self.text_selected.get(), s)
self.colourise_text(s, 'snow', color, s, v[0])