Update app.py

app.py

@@ -3,85 +3,72 @@ from transformers import pipeline
 from newspaper import Article
 import nltk
 from nltk.tokenize import sent_tokenize
-import re
 
-# Download punkt tokenizer for sentence splitting
-nltk.download('punkt')
+nltk.download("punkt")
 
-# Load grammar correction and toxicity detection models
+# Load models
 grammar_corrector = pipeline("text2text-generation", model="vennify/t5-base-grammar-correction")
 toxicity_classifier = pipeline("text-classification", model="unitary/toxic-bert")
 
+# Extract text from blog or URL
 def extract_text(input_type, text_input, url_input):
-    if input_type == "URL" and url_input:
-        try:
-            article = Article(url_input)
-            article.download()
-            article.parse()
-            return article.text
-        except Exception as e:
-            return f"Error extracting from URL: {str(e)}"
-    return text_input
-
-def check_grammar(text):
+    if input_type == "Text":
+        return text_input
     try:
-        corrected = grammar_corrector(text, max_length=512)[0]['generated_text']
-        return corrected
+        article = Article(url_input)
+        article.download()
+        article.parse()
+        return article.text
     except Exception as e:
-        return f"Error in grammar correction: {str(e)}"
-
-def detect_sensitive_content(text):
-    sentences = sent_tokenize(text)
-    sensitive = []
-    for i, sentence in enumerate(sentences):
-        result = toxicity_classifier(sentence)
-        if result[0]['label'] == 'toxic' and result[0]['score'] > 0.7:
-            sensitive.append({
-                "sentence": sentence,
-                "score": result[0]['score'],
-                "index": i
-            })
-    return sensitive
-
-def highlight_sensitive(text, sensitive_issues):
-    highlighted = text
-    for issue in sensitive_issues:
-        sentence = issue['sentence']
-        highlighted = highlighted.replace(sentence, f"<span style='background-color:red'>{sentence}</span>")
-    return highlighted
+        return f"Error fetching URL: {str(e)}"
 
+# Highlight grammar and toxic issues
 def review_blog(input_type, text_input, url_input):
     text = extract_text(input_type, text_input, url_input)
     if text.startswith("Error"):
-        return text, text
+        return text, "", []
 
-    corrected_text = check_grammar(text)
-    sensitive_issues = detect_sensitive_content(corrected_text)
-    highlighted = highlight_sensitive(corrected_text, sensitive_issues)
+    # Grammar correction
+    grammar_output = grammar_corrector(text, max_length=512)[0]["generated_text"]
+
+    # Toxic content detection
+    sentences = sent_tokenize(text)
+    toxic_sentences = []
+    for sent in sentences:
+        result = toxicity_classifier(sent)[0]
+        if result["label"] == "toxic" and result["score"] > 0.7:
+            toxic_sentences.append(sent)
+
+    # Highlight toxic sentences
+    highlighted = text
+    for sent in toxic_sentences:
+        highlighted = highlighted.replace(sent, f"<span style='background-color:red'>{sent}</span>")
 
-    return highlighted, corrected_text
+    return highlighted, grammar_output, toxic_sentences
 
 # Gradio UI
 with gr.Blocks(theme=gr.themes.Soft()) as demo:
-    gr.Markdown("# Blog Grammar & Toxicity Reviewer")
+    gr.Markdown("## 📝 Blog Review AI")
+    gr.Markdown("Checks for grammar & sensitive content (toxicity) in blog text or URL.")
 
-    input_type = gr.Radio(["Text", "URL"], label="Input Type", value="Text")
-    text_input = gr.Textbox(label="Blog Text", lines=10, visible=True)
-    url_input = gr.Textbox(label="Blog URL", visible=False)
+    input_type = gr.Radio(["Text", "URL"], value="Text", label="Input Type")
+    text_input = gr.Textbox(label="Enter blog text", lines=10, visible=True)
+    url_input = gr.Textbox(label="Enter blog URL", visible=False)
 
-    def toggle_inputs(choice):
-        return {
-            text_input: gr.update(visible=choice == "Text"),
-            url_input: gr.update(visible=choice == "URL")
-        }
+    def toggle_input(t):
+        return {text_input: gr.update(visible=t == "Text"), url_input: gr.update(visible=t == "URL")}
 
-    input_type.change(fn=toggle_inputs, inputs=input_type, outputs=[text_input, url_input])
+    input_type.change(toggle_input, input_type, [text_input, url_input])
 
-    review_button = gr.Button("Review Blog")
-    highlighted_output = gr.HTML(label="Highlighted Output (Red = Sensitive)")
-    corrected_output = gr.Textbox(label="Corrected Text", lines=10)
+    review_btn = gr.Button("Review")
+    highlight_output = gr.HTML(label="Toxic Highlighted Text")
+    corrected_text = gr.Textbox(label="Grammar Corrected Text", lines=10)
+    toxic_list = gr.Textbox(label="Toxic Sentences Detected", lines=5)
 
-    review_button.click(fn=review_blog, inputs=[input_type, text_input, url_input], outputs=[highlighted_output, corrected_output])
+    review_btn.click(
+        review_blog,
+        inputs=[input_type, text_input, url_input],
+        outputs=[highlight_output, corrected_text, toxic_list]
+    )
 
-if __name__ == "__main__":
-    demo.launch()
+demo.launch()