Update app.py
Browse files
app.py
CHANGED
@@ -36,40 +36,159 @@ API_KEY = os.environ["API_KEY"]
|
|
36 |
|
37 |
|
38 |
nlp = spacy.load("en_core_web_sm")
|
|
|
39 |
def process_nlp(system_message):
|
40 |
# Colorize the system message text
|
41 |
colorized_text = colorize_text(system_message['content'])
|
42 |
return colorized_text
|
43 |
|
44 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
45 |
colorized_text = ""
|
46 |
lines = text.split("\n")
|
47 |
-
|
|
|
|
|
|
|
|
|
|
|
48 |
for line in lines:
|
|
|
49 |
doc = nlp(line)
|
|
|
50 |
for token in doc:
|
|
|
51 |
if token.ent_type_:
|
52 |
-
|
53 |
-
|
54 |
-
|
55 |
-
|
56 |
-
|
57 |
-
|
58 |
-
|
59 |
-
|
60 |
-
|
61 |
-
|
62 |
-
|
63 |
-
|
64 |
-
|
65 |
-
|
66 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
67 |
else:
|
68 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
69 |
colorized_text += "<br>"
|
70 |
-
|
71 |
return colorized_text
|
72 |
|
|
|
73 |
def colorize_and_update(system_message, submit_update):
|
74 |
colorized_system_message = colorize_text(system_message['content'])
|
75 |
submit_update(None, colorized_system_message) # Pass the colorized_system_message as the second output
|
|
|
36 |
|
37 |
|
38 |
nlp = spacy.load("en_core_web_sm")
|
39 |
+
|
40 |
def process_nlp(system_message):
|
41 |
# Colorize the system message text
|
42 |
colorized_text = colorize_text(system_message['content'])
|
43 |
return colorized_text
|
44 |
|
45 |
+
from colour import Color
|
46 |
+
|
47 |
+
# define color combinations for different parts of speech
|
48 |
+
COLORS = {
|
49 |
+
"NOUN": "#000000", # Black
|
50 |
+
"VERB": "#ff6936", # Orange
|
51 |
+
"ADJ": "#4363d8", # Blue
|
52 |
+
"ADV": "#228b22", # Green
|
53 |
+
"digit": "#9a45d6", # Purple
|
54 |
+
"punct": "#ffcc00", # Yellow
|
55 |
+
"quote": "#b300b3" # Magenta
|
56 |
+
}
|
57 |
+
|
58 |
+
# define color combinations for individuals with dyslexia and color vision deficiencies
|
59 |
+
DYSLEXIA_COLORS = {
|
60 |
+
"NOUN": "#000000",
|
61 |
+
"VERB": "#ff6936",
|
62 |
+
"ADJ": "#4363d8",
|
63 |
+
"ADV": "#228b22",
|
64 |
+
"digit": "#9a45d6",
|
65 |
+
"punct": "#ffcc00",
|
66 |
+
"quote": "#b300b3",
|
67 |
+
}
|
68 |
+
RED_GREEN_COLORS = {
|
69 |
+
"NOUN": "#000000",
|
70 |
+
"VERB": "#fe642e", # Lighter orange
|
71 |
+
"ADJ": "#2e86c1", # Lighter blue
|
72 |
+
"ADV": "#82e0aa", # Lighter green
|
73 |
+
"digit": "#aa6c39", # Brown
|
74 |
+
"punct": "#f0b27a", # Lighter yellow
|
75 |
+
"quote": "#9932cc" # Darker magenta
|
76 |
+
}
|
77 |
+
|
78 |
+
# define a muted background color
|
79 |
+
BACKGROUND_COLOR = "#ffffff" # White
|
80 |
+
|
81 |
+
# define font and size
|
82 |
+
FONT = "OpenDyslexic"
|
83 |
+
FONT_SIZE = "18px"
|
84 |
+
|
85 |
+
def colorize_text(text, colors=DYSLEXIA_COLORS, background_color=None, font=FONT, font_size=FONT_SIZE):
|
86 |
+
if colors is None:
|
87 |
+
colors = COLORS
|
88 |
colorized_text = ""
|
89 |
lines = text.split("\n")
|
90 |
+
|
91 |
+
# set background color
|
92 |
+
if background_color is None:
|
93 |
+
background_color = BACKGROUND_COLOR
|
94 |
+
|
95 |
+
# iterate over the lines in the text
|
96 |
for line in lines:
|
97 |
+
# parse the line with the language model
|
98 |
doc = nlp(line)
|
99 |
+
# iterate over the tokens in the line
|
100 |
for token in doc:
|
101 |
+
# check if the token is an entity
|
102 |
if token.ent_type_:
|
103 |
+
# use dyslexia colors for entity if available
|
104 |
+
if colors == COLORS:
|
105 |
+
color = DYSLEXIA_COLORS.get(token.pos_, None)
|
106 |
+
else:
|
107 |
+
color = colors.get(token.pos_, None)
|
108 |
+
# check if a color is available for the token
|
109 |
+
if color is not None:
|
110 |
+
colorized_text += (
|
111 |
+
f'<span style="color: {color}; '
|
112 |
+
f'background-color: {background_color}; '
|
113 |
+
f'font-family: {font}; '
|
114 |
+
f'font-size: {font_size}; '
|
115 |
+
f'font-weight: bold; '
|
116 |
+
f'text-decoration: none; '
|
117 |
+
f'padding-right: 0.5em;">'
|
118 |
+
f"{token.text}</span>"
|
119 |
+
)
|
120 |
+
else:
|
121 |
+
colorized_text += (
|
122 |
+
f'<span style="font-family: {font}; '
|
123 |
+
f'font-size: {font_size}; '
|
124 |
+
f'font-weight: bold; '
|
125 |
+
f'text-decoration: none; '
|
126 |
+
f'padding-right: 0.5em;">'
|
127 |
+
f"{token.text}</span>"
|
128 |
+
)
|
129 |
else:
|
130 |
+
# check if a color is available for the token
|
131 |
+
color = colors.get(token.pos_, None)
|
132 |
+
if color is not None:
|
133 |
+
colorized_text += (
|
134 |
+
f'<span style="color: {color}; '
|
135 |
+
f'background-color: {background_color}; '
|
136 |
+
f'font-family: {font}; '
|
137 |
+
f'font-size: {font_size}; '
|
138 |
+
f'font-weight: bold; '
|
139 |
+
f'text-decoration: none; '
|
140 |
+
f'padding-right: 0.5em;">'
|
141 |
+
f"{token.text}</span>"
|
142 |
+
)
|
143 |
+
elif token.is_digit:
|
144 |
+
colorized_text += (
|
145 |
+
f'<span style="color: {colors["digit"]}; '
|
146 |
+
f'background-color: {background_color}; '
|
147 |
+
f'font-family: {font}; '
|
148 |
+
f'font-size: {font_size}; '
|
149 |
+
f'font-weight: bold; '
|
150 |
+
f'text-decoration: none; '
|
151 |
+
f'padding-right: 0.5em;">'
|
152 |
+
f"{token.text}</span>"
|
153 |
+
)
|
154 |
+
elif token.is_punct:
|
155 |
+
colorized_text += (
|
156 |
+
f'<span style="color: {colors["punct"]}; '
|
157 |
+
f'background-color: {background_color}; '
|
158 |
+
f'font-family: {font}; '
|
159 |
+
f'font-size: {font_size}; '
|
160 |
+
f'font-weight: bold; '
|
161 |
+
f'text-decoration: none; '
|
162 |
+
f'padding-right: 0.5em;">'
|
163 |
+
f"{token.text}</span>"
|
164 |
+
)
|
165 |
+
elif token.is_quote:
|
166 |
+
colorized_text += (
|
167 |
+
f'<span style="color: {colors["quote"]}; '
|
168 |
+
f'background-color: {background_color}; '
|
169 |
+
f'font-family: {font}; '
|
170 |
+
f'font-size: {font_size}; '
|
171 |
+
f'text-decoration: none; '
|
172 |
+
f'padding-right: 0.5em;">'
|
173 |
+
f"{token.text}</span>"
|
174 |
+
)
|
175 |
+
else:
|
176 |
+
# use larger font size for specific parts of speech, such as nouns and verbs
|
177 |
+
font_size = FONT_SIZE
|
178 |
+
if token.pos_ in ["NOUN", "VERB"]:
|
179 |
+
font_size = "22px"
|
180 |
+
colorized_text += (
|
181 |
+
f'<span style="font-family: {font}; '
|
182 |
+
f'font-size: {font_size}; '
|
183 |
+
f'font-weight: bold; '
|
184 |
+
f'text-decoration: none; '
|
185 |
+
f'padding-right: 0.5em;">'
|
186 |
+
f"{token.text}</span>"
|
187 |
+
)
|
188 |
colorized_text += "<br>"
|
|
|
189 |
return colorized_text
|
190 |
|
191 |
+
|
192 |
def colorize_and_update(system_message, submit_update):
|
193 |
colorized_system_message = colorize_text(system_message['content'])
|
194 |
submit_update(None, colorized_system_message) # Pass the colorized_system_message as the second output
|