Update app.py
Browse files
app.py
CHANGED
|
@@ -42,75 +42,139 @@ def process_nlp(system_message):
|
|
| 42 |
colorized_text = colorize_text(system_message['content'])
|
| 43 |
return colorized_text
|
| 44 |
|
| 45 |
-
#
|
| 46 |
COLORS = {
|
| 47 |
-
"NOUN": "#
|
| 48 |
-
"VERB": "#
|
| 49 |
-
"ADJ": "#
|
| 50 |
-
"ADV": "#
|
| 51 |
-
"digit": "#
|
| 52 |
-
"punct": "#
|
| 53 |
-
"quote": "#
|
| 54 |
}
|
| 55 |
|
| 56 |
-
#
|
| 57 |
DYSLEXIA_COLORS = {
|
| 58 |
-
"NOUN": "#
|
| 59 |
-
"VERB": "#
|
| 60 |
-
"ADJ": "#
|
| 61 |
-
"ADV": "#
|
| 62 |
-
"digit": "#
|
| 63 |
-
"punct": "#
|
| 64 |
-
"quote": "#
|
| 65 |
}
|
| 66 |
|
| 67 |
-
#
|
| 68 |
-
|
| 69 |
|
| 70 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 71 |
if colors is None:
|
| 72 |
colors = COLORS
|
| 73 |
-
|
| 74 |
colorized_text = ""
|
| 75 |
lines = text.split("\n")
|
| 76 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 77 |
for line in lines:
|
|
|
|
| 78 |
doc = nlp(line)
|
| 79 |
-
|
| 80 |
for token in doc:
|
|
|
|
| 81 |
if token.ent_type_:
|
| 82 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 83 |
else:
|
|
|
|
| 84 |
color = colors.get(token.pos_, None)
|
| 85 |
if color is not None:
|
| 86 |
colorized_text += (
|
| 87 |
f'<span style="color: {color}; '
|
| 88 |
-
f'background-color:
|
| 89 |
-
f
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 90 |
)
|
| 91 |
elif token.is_digit:
|
| 92 |
colorized_text += (
|
| 93 |
f'<span style="color: {colors["digit"]}; '
|
| 94 |
-
f'background-color:
|
| 95 |
-
f
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 96 |
)
|
| 97 |
elif token.is_punct:
|
| 98 |
colorized_text += (
|
| 99 |
f'<span style="color: {colors["punct"]}; '
|
| 100 |
-
f'background-color:
|
| 101 |
-
f
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 102 |
)
|
| 103 |
elif token.is_quote:
|
| 104 |
colorized_text += (
|
| 105 |
f'<span style="color: {colors["quote"]}; '
|
| 106 |
-
f'background-color:
|
| 107 |
-
f
|
|
|
|
|
|
|
|
|
|
|
|
|
| 108 |
)
|
| 109 |
else:
|
| 110 |
-
colorized_text +=
|
| 111 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 112 |
colorized_text += "<br>"
|
| 113 |
-
|
| 114 |
return colorized_text
|
| 115 |
|
| 116 |
|
|
|
|
| 42 |
colorized_text = colorize_text(system_message['content'])
|
| 43 |
return colorized_text
|
| 44 |
|
| 45 |
+
# define color combinations for different parts of speech
|
| 46 |
COLORS = {
|
| 47 |
+
"NOUN": "#5e5e5e", # Dark gray
|
| 48 |
+
"VERB": "#ff6936", # Orange
|
| 49 |
+
"ADJ": "#4363d8", # Blue
|
| 50 |
+
"ADV": "#228b22", # Green
|
| 51 |
+
"digit": "#9a45d6", # Purple
|
| 52 |
+
"punct": "#ffcc00", # Yellow
|
| 53 |
+
"quote": "#b300b3" # Magenta
|
| 54 |
}
|
| 55 |
|
| 56 |
+
# define color combinations for individuals with dyslexia
|
| 57 |
DYSLEXIA_COLORS = {
|
| 58 |
+
"NOUN": "#5e5e5e",
|
| 59 |
+
"VERB": "#ff6936",
|
| 60 |
+
"ADJ": "#4363d8",
|
| 61 |
+
"ADV": "#228b22",
|
| 62 |
+
"digit": "#9a45d6",
|
| 63 |
+
"punct": "#ffcc00",
|
| 64 |
+
"quote": "#b300b3"
|
| 65 |
}
|
| 66 |
|
| 67 |
+
# define a muted background color
|
| 68 |
+
BACKGROUND_COLOR = "#f5f5f5" # Light gray
|
| 69 |
|
| 70 |
+
# define font and size
|
| 71 |
+
FONT = "Arial"
|
| 72 |
+
FONT_SIZE = "14px"
|
| 73 |
+
|
| 74 |
+
# load the English language model
|
| 75 |
+
nlp = spacy.load('en_core_web_sm')
|
| 76 |
+
|
| 77 |
+
def colorize_text(text, colors=None, background_color=None):
|
| 78 |
if colors is None:
|
| 79 |
colors = COLORS
|
|
|
|
| 80 |
colorized_text = ""
|
| 81 |
lines = text.split("\n")
|
| 82 |
+
|
| 83 |
+
# set background color
|
| 84 |
+
if background_color is None:
|
| 85 |
+
background_color = BACKGROUND_COLOR
|
| 86 |
+
|
| 87 |
+
# iterate over the lines in the text
|
| 88 |
for line in lines:
|
| 89 |
+
# parse the line with the language model
|
| 90 |
doc = nlp(line)
|
| 91 |
+
# iterate over the tokens in the line
|
| 92 |
for token in doc:
|
| 93 |
+
# check if the token is an entity
|
| 94 |
if token.ent_type_:
|
| 95 |
+
# use dyslexia colors for entity if available
|
| 96 |
+
if colors == COLORS:
|
| 97 |
+
color = DYSLEXIA_COLORS.get(token.pos_, None)
|
| 98 |
+
else:
|
| 99 |
+
color = colors.get(token.pos_, None)
|
| 100 |
+
# check if a color is available for the token
|
| 101 |
+
if color is not None:
|
| 102 |
+
colorized_text += (
|
| 103 |
+
f'<span style="color: {color}; '
|
| 104 |
+
f'background-color: {background_color}; '
|
| 105 |
+
f'font-family: {FONT}; '
|
| 106 |
+
f'font-size: {FONT_SIZE}; '
|
| 107 |
+
f'font-weight: bold; '
|
| 108 |
+
f'text-decoration: none; '
|
| 109 |
+
f'padding-right: 0.5em;">' # Add space between tokens
|
| 110 |
+
f"{token.text}</span>"
|
| 111 |
+
)
|
| 112 |
+
else:
|
| 113 |
+
colorized_text += (
|
| 114 |
+
f'<span style="font-family: {FONT}; '
|
| 115 |
+
f'font-size: {FONT_SIZE}; '
|
| 116 |
+
f'font-weight: bold; '
|
| 117 |
+
f'text-decoration: none; '
|
| 118 |
+
f'padding-right: 0.5em;">' # Add space between tokens
|
| 119 |
+
f"{token.text}</span>"
|
| 120 |
+
)
|
| 121 |
else:
|
| 122 |
+
# check if a color is available for the token
|
| 123 |
color = colors.get(token.pos_, None)
|
| 124 |
if color is not None:
|
| 125 |
colorized_text += (
|
| 126 |
f'<span style="color: {color}; '
|
| 127 |
+
f'background-color: {background_color}; '
|
| 128 |
+
f'font-family: {FONT}; '
|
| 129 |
+
f'font-size: {FONT_SIZE}; '
|
| 130 |
+
f'font-weight: bold; '
|
| 131 |
+
f'text-decoration: none; '
|
| 132 |
+
f'padding-right: 0.5em;">' # Add space between tokens
|
| 133 |
+
f"{token.text}</span>"
|
| 134 |
)
|
| 135 |
elif token.is_digit:
|
| 136 |
colorized_text += (
|
| 137 |
f'<span style="color: {colors["digit"]}; '
|
| 138 |
+
f'background-color: {background_color}; '
|
| 139 |
+
f'font-family: {FONT}; '
|
| 140 |
+
f'font-size: {FONT_SIZE}; '
|
| 141 |
+
f'font-weight: bold; '
|
| 142 |
+
f'text-decoration: none; '
|
| 143 |
+
f'padding-right: 0.5em;">' # Add space between tokens
|
| 144 |
+
f"{token.text}</span>"
|
| 145 |
)
|
| 146 |
elif token.is_punct:
|
| 147 |
colorized_text += (
|
| 148 |
f'<span style="color: {colors["punct"]}; '
|
| 149 |
+
f'background-color: {background_color}; '
|
| 150 |
+
f'font-family: {FONT}; '
|
| 151 |
+
f'font-size: {FONT_SIZE}; '
|
| 152 |
+
f'font-weight: bold; '
|
| 153 |
+
f'text-decoration: none; '
|
| 154 |
+
f'padding-right: 0.5em;">' # Add space between tokens
|
| 155 |
+
f"{token.text}</span>"
|
| 156 |
)
|
| 157 |
elif token.is_quote:
|
| 158 |
colorized_text += (
|
| 159 |
f'<span style="color: {colors["quote"]}; '
|
| 160 |
+
f'background-color: {background_color}; '
|
| 161 |
+
f'font-family: {FONT}; '
|
| 162 |
+
f'font-size: {FONT_SIZE}; '
|
| 163 |
+
f'text-decoration: none; '
|
| 164 |
+
f'padding-right: 0.5em;">' # Add space between tokens
|
| 165 |
+
f"{token.text}</span>"
|
| 166 |
)
|
| 167 |
else:
|
| 168 |
+
colorized_text += (
|
| 169 |
+
f'<span style="font-family: {FONT}; '
|
| 170 |
+
f'font-size: {FONT_SIZE}; '
|
| 171 |
+
f'font-weight: bold; '
|
| 172 |
+
f'text-decoration: none; '
|
| 173 |
+
f'padding-right: 0.5em;">' # Add space between tokens
|
| 174 |
+
f"{token.text}</span>"
|
| 175 |
+
)
|
| 176 |
colorized_text += "<br>"
|
| 177 |
+
|
| 178 |
return colorized_text
|
| 179 |
|
| 180 |
|