Update app.py

app.py

@@ -173,22 +173,31 @@ def on_select(instruction1, instruction2, instruction3, evt: gr.SelectData):
 # Grammar metrics
 import re
 
-def preprocess_text(text):
-    # Remove extra spaces and special characters except punctuation
-    text = re.sub(r'\s+', ' ', text).strip()
-    return text
-
-def count_sentences(text):
-    # Split sentences using punctuation followed by a space or end of string
-    sentences = re.split(r'(?<=[.!?]) +', text)
-    return len(sentences), sentences
+def pre_process_text(text):
+    sentences_list = re.split(r'(?<!\w\.\w.)(?<![A-Z][a-z]\.)(?<=\.|\?)\s', text)
+    # Split the elements of the list by newline characters
+    split_sentences = []
+    for sentence in sentences_list:
+        split_sentences.extend(re.split(r'\n+', sentence))
+    # Remove empty elements
+    cleaned_sentences = [sentence for sentence in split_sentences if sentence.strip()]
+    sentences_number = len(cleaned_sentences)
+    return sentences, sentences_number
+
+# Function to clean the sentences list and return words only
+def extract_words(sentences):
+    words = []
+    for sentence in sentences:
+        # Extract words using regex, ignoring special characters
+        words.extend(re.findall(r'\b\w+\b', sentence))
+    return words
 
 def count_syllables(word):
     return len(re.findall(r'[aeiouyAEIOUY]', word))
 
 def flesch_kincaid_grade_level(text):
-    sentences_count, sentences = count_sentences(text)
-    words = len(text.split())
+    sentences, sentences_count = pre_process_text(text)
+    words = extract_words(sentences)
     syllables = sum([count_syllables(word) for word in text.split()])
 
     if sentences_count == 0 or words == 0:
@@ -196,8 +205,8 @@ def flesch_kincaid_grade_level(text):
     return 0.39 * (words / sentences_count) + 11.8 * (syllables / words) - 15.59
 
 def flesch_reading_ease(text):
-    sentences_count, sentences = count_sentences(text)
-    words = len(text.split())
+    sentences_count, sentences = pre_process_text(text)
+    words = extract_words(sentences)
     syllables = sum([count_syllables(word) for word in text.split()])
 
     if sentences_count == 0 or words == 0:
@@ -205,8 +214,8 @@ def flesch_reading_ease(text):
     return 206.835 - 1.015 * (words / sentences_count) - 84.6 * (syllables / words)
 
 def gunning_fog_index(text):
-    sentences_count, sentences = count_sentences(text)
-    words = len(text.split())
+    sentences_count, sentences = pre_process_text(text)
+    words = extract_words(sentences)
     complex_words = len([word for word in text.split() if count_syllables(word) >= 3])
 
     if sentences_count == 0 or words == 0: