src/streamlit_app.py

import streamlit as st
import requests
import pandas as pd
from PIL import Image
import io
from io import BytesIO

def check_image_accessibility(url):
    try:
        response = requests.get(url, timeout=5)
        if response.status_code == 200:
            try:
                image = Image.open(BytesIO(response.content))
                # Verify if image can be displayed by checking its mode and size
                if image.mode in ['RGB', 'RGBA', 'L'] and image.size[0] > 0 and image.size[1] > 0:
                    return "Accessible", image
                else:
                    return "Not Accessible", None
            except Exception:
                return "Not Accessible", None
        else:
            return "Not Accessible", None
    except Exception:
        return "Not Accessible", None

def analyze_image(image_file):
    # TinEye API endpoint
    URL = "https://tineye.com/api/v1/result_json/?sort=score&order=desc"
    
    # Read image file
    #image_data = image_file.read()
    image_data = image_file.getvalue()
    # Prepare files for API request
    files = {'image': ("test.png", image_data)}
    
    # Make API request
    response = requests.post(URL, files=files)
    output = response.json()
    
    # Extract data from response
    urls = []
    backlinks = []
    crawl_dates = []
    image_urls = []
    image_statuses = []
    images = []
    
    for match in output['matches']:
        for backlink in match['backlinks']:
            urls.append(backlink['url'])
            backlinks.append(backlink['backlink'])
            crawl_dates.append(backlink['crawl_date'])
            image_urls.append(match['image_url'])
            
            # Check image accessibility
            status, image = check_image_accessibility(match['image_url'])
            image_statuses.append(status)
            images.append(image)
    
    # Create DataFrame
    df = pd.DataFrame({
        'URL': urls,
        'Backlink': backlinks,
        'Crawl Date': crawl_dates,
        'Image URL': image_urls,
        'Image Status': image_statuses
    })
    
    # Sort by crawl date
    df = df.sort_values(by='Crawl Date', ascending=True)
    
    return df, output['first_crawl_date'], images

def main():
    st.title("Image Analysis Tool")
    
    st.header("Image Analysis")
    uploaded_file = st.file_uploader("Choose an image file", type=['png', 'jpg', 'jpeg'])
    
    if uploaded_file is not None:
        # Display the uploaded image
        image = Image.open(uploaded_file)
        st.image(image, caption='Uploaded Image', use_column_width=True)
        
        if st.button("Analyze Image"):
            with st.spinner("Analyzing image..."):
                try:
                    df, first_crawl_date, images = analyze_image(uploaded_file)
                    
                    # Display first crawl date
                    st.subheader("First Found Date")
                    st.write(f"This image was first found on: {first_crawl_date}")
                    
                    # Display matches with images and details
                    st.subheader("Matched Images")
                    
                    for idx, row in df.iterrows():
                        # Create a container for each match
                        with st.container():
                            # Display the matched image if available and accessible
                            if row['Image Status'] == "Accessible" and images[idx] is not None:
                                try:
                                    st.image(images[idx], use_column_width=True)
                                except Exception:
                                    # If image display fails, update status
                                    df.at[idx, 'Image Status'] = "Not Accessible"
                                    st.warning("Image not accessible")
                            else:
                                st.warning("Image not accessible")
                            
                            # Display details in a clean format
                            st.markdown("---")
                            st.markdown(f"**Match {idx + 1} Details:**")
                            st.markdown(f"**URL:** {row['URL']}")
                            st.markdown(f"**Backlink:** {row['Backlink']}")
                            st.markdown(f"**Crawl Date:** {row['Crawl Date']}")
                            st.markdown(f"**Image Status:** {row['Image Status']}")
                            st.markdown("---")
                    
                    # Add a separator before the DataFrame
                    st.markdown("---")
                    st.markdown("### Complete Results Table")
                    
                    # Display the DataFrame with clickable links
                    st.dataframe(
                        df,
                        use_container_width=True,
                        hide_index=True,
                        column_config={
                            "URL": st.column_config.LinkColumn("URL"),
                            "Backlink": st.column_config.LinkColumn("Backlink"),
                            "Crawl Date": st.column_config.DateColumn("Crawl Date"),
                            "Image URL": st.column_config.LinkColumn("Image URL"),
                            "Image Status": st.column_config.TextColumn("Image Status")
                        }
                    )
                    
                    # Download button for results
                    csv = df.to_csv(index=False)
                    st.download_button(
                        label="Download Results as CSV",
                        data=csv,
                        file_name="image_analysis_results.csv",
                        mime="text/csv"
                    )
                except Exception as e:
                    st.error(f"An error occurred: {str(e)}")

if __name__ == "__main__":
    main()