Update app.py

app.py

@@ -6,183 +6,75 @@ import google.generativeai as genai
 import os
 import re
 import tempfile
-import asyncio
-import time
 
-# Download NLTK tokenizer
+# Download NLTK tokenizer data
 download('punkt')
 
-# Configure Gemini API
+# Set your Gemini API key
 api_key = os.environ.get("GEMINI_API_KEY")
 if not api_key:
-    raise ValueError("GEMINI_API_KEY not found in environment variables.")
+    raise ValueError("GEMINI_API_KEY not found in environment variables")
 
 genai.configure(api_key=api_key)
+model = genai.GenerativeModel("gemini-pro")
 
-# Load Gemini model
-try:
-    model = genai.GenerativeModel('gemini-1.5-flash')
-except Exception as e:
-    print(f"Error initializing model: {e}")
-    for m in genai.list_models():
-        print(m.name)
-    raise
-
-# Prompt template
-PROMPT = """
-You are an AI content reviewer. Analyze the provided text for the following:
-1. Grammar Issues: Identify and suggest corrections for grammatical errors.
-2. Legal Policy Violations: Flag content that may violate common legal policies.
-3. Crude/Abusive Language: Detect crude, offensive, or abusive language.
-4. Sensitive Topics: Identify content related to sensitive topics like racism or discrimination.
-
-Return the results in this markdown format:
-# Blog Review Report
-
-## Grammar Corrections
-1. [Heading of issue]
-   - CONTENT: [Text]
-   - SUGGESTION: [Fix]
-   - ISSUE: [Explanation]
-
-## Legal Policy Violations
-- CONTENT: [...]
-  SUGGESTION: [...]
-  ISSUE: [...]
-
-## Crude/Abusive Language
-- [Text or “None detected”]
-
-## Sensitive Topics
-- [Text or “None detected”]
-"""
-
-# Fetch blog content from URL
-async def fetch_url_content(url):
+# Helper functions
+def extract_text_from_url(url):
     try:
         response = requests.get(url, timeout=10)
-        response.raise_for_status()
-        soup = BeautifulSoup(response.text, 'html.parser')
-        content = ' '.join([p.get_text(strip=True) for p in soup.find_all(['p', 'article', 'div'])])
-        return content or "No readable content found on the page."
-    except Exception as e:
-        return f"Error fetching URL: {str(e)}"
-
-# Review function
-async def review_blog(text_input, url_input, progress=gr.Progress()):
-    if text_input and not url_input:
-        input_text = text_input
-    elif url_input and not text_input:
-        input_text = await fetch_url_content(url_input)
-        if input_text.startswith("Error"):
-            return "Review Blog", input_text, gr.update(visible=False)
-    else:
-        return "Review Blog", "Error: Provide either text or URL, not both.", gr.update(visible=False)
+        soup = BeautifulSoup(response.text, "html.parser")
 
-    sentences = sent_tokenize(input_text)
-    analysis_text = "\n".join(sentences)
+        for script in soup(["script", "style", "header", "footer", "nav"]):
+            script.decompose()
 
-    progress(0.2, desc="Analyzing blog content...")
-    try:
-        response = await asyncio.to_thread(model.generate_content, PROMPT + "\n\nText to analyze:\n" + analysis_text)
-        report = response.text.strip()
-        report = re.sub(r'^markdown\n|$', '', report, flags=re.MULTILINE)
-    except Exception as e:
-        return "Review Blog", f"Error with Gemini API: {str(e)}", gr.update(visible=False)
+        paragraphs = soup.find_all("p")
+        blog_text = "\n".join(p.get_text() for p in paragraphs if len(p.get_text()) > 40)
 
-    try:
-        with tempfile.NamedTemporaryFile(mode='w', suffix='.md', delete=False, encoding='utf-8') as temp_file:
-            temp_file.write(report)
-            file_path = temp_file.name
-        progress(1.0, desc="Report Ready!")
-        return "Review Blog", report, gr.update(visible=True, value=file_path)
+        return blog_text.strip()
     except Exception as e:
-        return "Review Blog", f"Error saving report: {str(e)}", gr.update(visible=False)
-
-# ------------------ CSS ------------------ #
-custom_css = """
-@import url('https://fonts.googleapis.com/css2?family=Inter:wght@400;600;700&display=swap');
-
-.gradio-container {
-    font-family: 'Inter', sans-serif !important;
-    background: linear-gradient(120deg, #e0f7fa 0%, #f9f9f9 100%);
-    padding: 40px;
-    border-radius: 20px;
-    box-shadow: 0 10px 40px rgba(0, 0, 0, 0.07);
-}
-
-h1, h3 {
-    font-weight: 700;
-    color: #2b6cb0;
-}
-
-input, textarea {
-    border: 2px solid #cbd5e0;
-    border-radius: 12px;
-    padding: 15px;
-    background-color: #ffffff;
-    font-size: 16px;
-    color: #2d3748;
-    transition: border 0.3s ease;
-}
-
-input:focus, textarea:focus {
-    border-color: #3182ce;
-    outline: none;
-}
-
-.review-btn {
-    background: linear-gradient(45deg, #3182ce, #63b3ed);
-    padding: 14px 28px;
-    border-radius: 12px;
-    font-size: 16px;
-    color: #fff;
-    font-weight: 600;
-    border: none;
-    cursor: pointer;
-    transition: all 0.3s ease;
-}
-
-.review-btn:hover {
-    transform: scale(1.04);
-    background: linear-gradient(45deg, #63b3ed, #3182ce);
-}
-
-.markdown-container {
-    background: #ffffff;
-    padding: 25px;
-    border-radius: 16px;
-    box-shadow: 0 8px 24px rgba(0, 0, 0, 0.08);
-    margin-top: 25px;
-    font-size: 15px;
-    line-height: 1.6;
-}
-
-.download-btn {
-    background-color: #38a169;
-    color: white;
-    padding: 12px 20px;
-    border-radius: 10px;
-    font-weight: 600;
-    transition: background 0.3s ease;
-}
-
-.download-btn:hover {
-    background-color: #2f855a;
-}
+        return f"Error extracting content: {str(e)}"
+
+def review_blog(blog_text):
+    prompt = f"""
+You are a professional AI blog reviewer and sensitive content analyst.
+
+Carefully analyze the following blog content for:
+1. Grammar, spelling, and clarity improvements
+2. Crude, harsh, or unprofessional language
+3. Sensitive, biased, offensive, or policy-violating content
+
+Provide detailed suggestions and highlight what should be improved.
+
+BLOG:
+\"\"\"
+{blog_text}
+\"\"\"
 """
+    response = model.generate_content(prompt)
+    return response.text
+
+def process_input(text_input, url_input):
+    if url_input:
+        blog_content = extract_text_from_url(url_input)
+    else:
+        blog_content = text_input
+
+    if not blog_content or blog_content.strip() == "":
+        return "Please provide blog content or a valid URL.", None
+
+    analysis = review_blog(blog_content)
+
+    # Create a temporary file with the markdown review
+    with tempfile.NamedTemporaryFile(mode="w", delete=False, suffix=".md") as tmp_file:
+        tmp_file.write(analysis)
+        file_path = tmp_file.name
+
+    return analysis, file_path
 
-# ------------------ UI ------------------ #
-with gr.Blocks(css=custom_css, theme=gr.themes.Soft()) as demo:
-    with gr.Row():
-        gr.HTML("""
-        <div style="text-align: center; margin-bottom: 10px;">
-            <h1 style="color:#2b6cb0; font-size: 2.5rem;">✨ AI Blog Reviewer</h1>
-            <p style="font-size: 1.1rem; color: #4a5568;">
-                Analyze your blog for grammar, legal violations, abusive content, and more.
-            </p>
-        </div>
-        """)
+# Gradio UI
+with gr.Blocks(theme=gr.themes.Soft()) as demo:
+    gr.Markdown("# 🧠 Blog Checker AI")
+    gr.Markdown("Review blogs for grammar, clarity, and policy issues using Gemini AI.")
 
     with gr.Row(equal_height=True):
         with gr.Column(scale=3):
@@ -191,35 +83,29 @@ with gr.Blocks(css=custom_css, theme=gr.themes.Soft()) as demo:
                     text_input = gr.Textbox(
                         label="Paste Your Blog Content",
                         placeholder="Write or paste your blog text here...",
-                        lines=12,
-                        elem_classes=["input-text"]
+                        lines=12
                     )
                 with gr.TabItem("🔗 URL Input"):
                     url_input = gr.Textbox(
                         label="Enter Blog URL",
-                        placeholder="https://example.com/blog",
-                        lines=1,
-                        elem_classes=["input-url"]
+                        placeholder="https://example.com/blog"
                     )
         with gr.Column(scale=2):
             gr.Markdown("""
 ### 🚀 How It Works
 1. **Input**: Paste your blog text or provide a blog URL  
-2. **Analysis**: Gemini AI checks for grammar issues, policy violations, and more  
-3. **Report**: Instantly download a clean, structured Markdown review  
-""")
-            status_button = gr.Button("🧠 Review Blog", elem_classes=["review-btn"])
+2. **Analysis**: Gemini AI checks for grammar, legal violations, and sensitive content  
+3. **Report**: Instantly download a structured Markdown review
+            """)
+            status_button = gr.Button("🧠 Review Blog")
 
-    gr.Markdown("### 📄 Review Report")
-    with gr.Row():
-        with gr.Column():
-            report_output = gr.Markdown(elem_classes=["markdown-container"])
-            download_btn = gr.File(label="Download Report", visible=False, elem_classes=["download-btn"])
+    output_text = gr.Textbox(label="Review Summary", lines=20)
+    download_button = gr.File(label="Download Markdown Review")
 
     status_button.click(
-        fn=review_blog,
+        fn=process_input,
         inputs=[text_input, url_input],
-        outputs=[status_button, report_output, download_btn]
+        outputs=[output_text, download_button]
     )
 
 demo.launch()