Create app.py

app.py

@@ -0,0 +1,185 @@
+import gradio as gr
+from docx import Document
+from datetime import datetime
+import re
+import tempfile
+import pytesseract
+from PIL import Image
+from groq import Groq
+import io
+import os
+
+# Initialize APIs
+client = Groq(api_key=os.environ.get("GROQ_API_KEY"))
+pytesseract.pytesseract.tesseract_cmd = r'/usr/bin/tesseract'  # Update path as needed
+
+# Helper Functions
+def extract_text_from_image(image_path):
+    try:
+        img = Image.open(image_path)
+        text = pytesseract.image_to_string(img)
+        return text.strip() if text else "No text found in image"
+    except Exception as e:
+        return f"OCR Error: {str(e)}"
+
+def groq_processing(prompt):
+    try:
+        completion = client.chat.completions.create(
+            model="mixtral-8x7b-32768",
+            messages=[
+                {
+                    "role": "user",
+                    "content": prompt
+                }
+            ]
+        )
+        return completion.choices[0].message.content
+    except Exception as e:
+        return f"AI Processing Error: {str(e)}"
+
+def analyze_sentiment(text):
+    prompt = f"""Analyze the sentiment of this text with detailed emotions:
+    {text}
+    
+    Format response as: 
+    Primary Emotion: [emotion]
+    Secondary Emotions: [comma-separated list]
+    Confidence Level: [percentage]"""
+    return groq_processing(prompt)
+
+def categorize_content(text):
+    prompt = f"""Categorize this text into specific themes from the following options:
+    Technology, Business, Lifestyle, Education, Politics, Health, Entertainment, Sports, Art, Science
+    
+    Text: {text}
+    
+    Respond with top 3 relevant categories in order of relevance."""
+    return groq_processing(prompt)
+
+# Document Processing Functions
+def create_word_document(posts):
+    doc = Document()
+    doc.add_heading('Social Media Data Extraction Report', 0)
+    
+    for idx, post in enumerate(posts):
+        doc.add_heading(f'Post {idx+1}', level=1)
+        
+        data = [
+            ("Date of Post", post.get('date', 'N/A')),
+            ("Media Type", post.get('media_type', 'N/A')),
+            ("Number of Pictures", post.get('num_pictures', 'N/A')),
+            ("Likes", post.get('likes', 'N/A')),
+            ("Comments", post.get('comments', 'N/A')),
+            ("Caption", post.get('caption', 'N/A')),
+            ("OCR Text", post.get('ocr_text', 'N/A')),
+            ("Language", post.get('language', 'N/A')),
+            ("Sentiment Analysis", post.get('sentiment', 'N/A')),
+            ("Content Categories", post.get('categories', 'N/A')),
+            ("Hashtags", ', '.join(post.get('hashtags', [])) if post.get('hashtags') else 'N/A'),
+            ("Concept Keywords", ', '.join(post.get('concepts', [])) if post.get('concepts') else 'N/A')
+        ]
+        
+        for label, value in data:
+            doc.add_paragraph(f"{label}: {value}")
+        
+        doc.add_page_break()
+    
+    temp_file = tempfile.NamedTemporaryFile(delete=False, suffix=".docx")
+    doc.save(temp_file.name)
+    return temp_file.name
+
+# Processing Functions 
+def process_social_media(profile_link, hashtags, concepts, platform, date_range, num_posts):
+    # Mock data - Replace with actual social media API calls
+    mock_image = Image.new('RGB', (800, 600), color='white')
+    mock_image_path = tempfile.mktemp(suffix='.jpg')
+    mock_image.save(mock_image_path)
+    
+    posts = [{
+        'date': datetime.now().strftime("%Y-%m-%d"),
+        'media_type': 'Image',
+        'num_pictures': 1,
+        'likes': 150,
+        'comments': 20,
+        'caption': 'Example caption with #technology',
+        'ocr_text': extract_text_from_image(mock_image_path),
+        'hashtags': ['#technology'],
+        'concepts': [c.strip() for c in concepts.split(',')],
+        'sentiment': analyze_sentiment('Example caption with #technology'),
+        'categories': categorize_content('Example caption with #technology')
+    }]
+    
+    os.remove(mock_image_path)
+    return create_word_document(posts[:num_posts])
+
+def process_word_file(file, concepts):
+    doc = Document(file.name)
+    posts = []
+    
+    for para in doc.paragraphs:
+        if para.text.startswith('Post'):
+            posts.append({'caption': ''})
+        elif posts:
+            posts[-1]['caption'] += para.text + '\n'
+    
+    processed_posts = []
+    for post in posts:
+        caption = post.get('caption', '')
+        processed_post = {
+            'date': datetime.now().strftime("%Y-%m-%d"),
+            'media_type': 'Text',
+            'caption': caption,
+            'sentiment': analyze_sentiment(caption),
+            'categories': categorize_content(caption),
+            'concepts': [c.strip() for c in concepts.split(',')],
+            'hashtags': re.findall(r'#\w+', caption)
+        }
+        processed_posts.append(processed_post)
+    
+    return create_word_document(processed_posts)
+
+# Gradio Interface
+with gr.Blocks(title="Social Media Analyzer") as app:
+    gr.Markdown("# Social Media Data Extraction Tool")
+    
+    with gr.Tabs():
+        with gr.TabItem("Social Media Extraction"):
+            gr.Markdown("## Analyze Social Media Profiles")
+            with gr.Row():
+                with gr.Column():
+                    profile_link = gr.Textbox(label="Profile URL")
+                    hashtags = gr.Textbox(label="Hashtags (comma separated)")
+                    concepts = gr.Textbox(label="Concept Keywords (comma separated)")
+                    platform = gr.Radio(["Instagram", "Twitter", "Facebook"], label="Platform")
+                    date_range = gr.DatePicker(label="Date Range")
+                    num_posts = gr.Slider(1, 100, value=10, label="Number of Posts")
+                    sm_submit = gr.Button("Analyze Profile", variant="primary")
+                
+                with gr.Column():
+                    sm_output = gr.File(label="Download Analysis Report")
+
+        with gr.TabItem("Document Processing"):
+            gr.Markdown("## Analyze Word Documents")
+            with gr.Row():
+                with gr.Column():
+                    word_file = gr.File(label="Upload Word Document")
+                    wp_concepts = gr.Textbox(label="Concept Keywords (comma separated)")
+                    wp_submit = gr.Button("Analyze Document", variant="primary")
+                
+                with gr.Column():
+                    wp_output = gr.File(label="Download Analysis Report")
+
+    sm_submit.click(
+        fn=process_social_media,
+        inputs=[profile_link, hashtags, concepts, platform, date_range, num_posts],
+        outputs=sm_output
+    )
+    
+    wp_submit.click(
+        fn=process_word_file,
+        inputs=[word_file, wp_concepts],
+        outputs=wp_output
+    )
+
+if __name__ == "__main__":
+    app.launch()