Update src/streamlit_app.py

src/streamlit_app.py

@@ -260,10 +260,8 @@ def get_similar_content(content, n_recommendations=5):
             "accept": "application/json"
         }
         
-        # Debug: Print the title and show_id we're searching for
         title = content.get('title', '')
         show_id = content.get('show_id', '')
-        st.write(f"🔍 Searching for similar content to: {title} (ID: {show_id})")
         
         # URLs for similarity tables using table IDs
         similarity_table_urls = [
@@ -272,57 +270,30 @@ def get_similar_content(content, n_recommendations=5):
             "https://mtoft20-potm.hf.space/api/v1/db/data/noco/p9pozkcw81t9aee/m2driodimid10k6"   # Part 3 similarities
         ]
         
-        similar_items = []
+        # Try finding by both show_id and title
+        query = f'where=(show_id,eq,{show_id})~and(title,eq,{title})'
+        params = {
+            "where": query
+        }
         
-        # Check both tables for similarities
         for table_url in similarity_table_urls:
-            # Debug: Get a sample row to see table structure
-            sample_params = {
-                "limit": 1
-            }
-            sample_response = requests.get(table_url, headers=headers, params=sample_params)
-            if sample_response.status_code == 200:
-                sample_data = sample_response.json()
-                if sample_data.get('list'):
-                    st.write(f"📋 Table structure for {table_url.split('/')[-1]}:")
-                    st.write("Columns:", list(sample_data['list'][0].keys()))
-                    st.write("Sample row:", sample_data['list'][0])
-            
-            # Try finding by both show_id and title
-            escaped_title = title.replace('"', '\\"')  # Escape any quotes in the title
-            query = f'where=(show_id,eq,{show_id})~and(title,eq,"{escaped_title}")'  # Add quotes around title
-            params = {
-                "where": query
-            }
-            
-            # Debug: Print the request details
-            st.write(f"📡 Querying table: {table_url.split('/')[-1]}")
-            st.write(f"🔍 Query params: {params}")
-            
-            # Try first query format
             try:
                 response = requests.get(table_url, headers=headers, params=params)
-                st.write(f"📥 Response status: {response.status_code}")
-                st.write("📥 Request URL:", response.url)
                 
                 if response.status_code == 200:
                     data = response.json()
                     if data.get('list'):
-                        st.write(f"✅ Found exact match")
                         for entry in data['list']:
-                            st.write(f"Entry title: {entry.get('title')}, show_id: {entry.get('show_id')}")
-                            
                             # Parse similar items
                             try:
                                 similar_items = json.loads(entry['similar_items'])
-                                st.write(f"✅ Successfully parsed {len(similar_items)} similar items")
                                 
                                 # Get full content details for each similar item
                                 similar_content = []
-                                for item in similar_items:
+                                for item in similar_items[:n_recommendations]:  # Only process top N items
                                     # Query main content table for full details using show_id
                                     show_id = item.get('show_id', '')
-                                    query = f'where=(show_id,eq,"{show_id}")'  # Add quotes around show_id
+                                    query = f'where=(show_id,eq,{show_id})'
                                     content_params = {
                                         "where": query
                                     }
@@ -334,58 +305,15 @@ def get_similar_content(content, n_recommendations=5):
                                             content_dict = content_data['list'][0]
                                             content_dict['similarity'] = f"{item['similarity']:.2%}"
                                             similar_content.append(content_dict)
-                                            st.write(f"✅ Found details for: {item['title']}")
-                                        else:
-                                            st.write(f"❌ No content details found for: {item['title']}")
                                 
-                                return similar_content
+                                return similar_content[:n_recommendations]  # Return top N items
                             except Exception as parse_error:
-                                st.write(f"❌ Error parsing similar items: {str(parse_error)}")
-                                st.write("Raw data:", entry)
-                    else:
-                        st.write("❌ No exact match found")
+                                continue
             except Exception as e:
-                st.write(f"❌ Request error: {str(e)}")
-        
-        st.write(f"📝 Total similar items found: {len(similar_items)}")
-        
-        if similar_items:
-            # Sort by similarity score and limit to requested number
-            similar_items.sort(key=lambda x: x['similarity'], reverse=True)
-            similar_items = similar_items[:n_recommendations]
-            
-            # Get full content details for each similar item
-            similar_content = []
-            for item in similar_items:
-                # Query main content table for full details using show_id if available
-                show_id = item.get('show_id', '')
-                if show_id:
-                    query = f'(show_id,eq,"{show_id}")'
-                else:
-                    # Fallback to title if show_id not available
-                    query = f'(title,eq,"{item["title"]}")'
-                    
-                content_params = {
-                    "where": query
-                }
-                content_response = requests.get(NOCODB_URL, headers=headers, params=content_params)
-                
-                if content_response.status_code == 200:
-                    content_data = content_response.json()
-                    if content_data and len(content_data.get('list', [])) > 0:
-                        content_dict = content_data['list'][0]
-                        content_dict['similarity'] = f"{item['similarity']:.2%}"
-                        similar_content.append(content_dict)
-                        st.write(f"✅ Found details for: {item['title']}")
-                    else:
-                        st.write(f"❌ No content details found for: {item['title']}")
-            
-            return similar_content
+                continue
         
         return []
     except Exception as e:
-        st.error(f"Error fetching similar content: {str(e)}")
-        st.write("Full error details:", e)
         return []
 
 # =============================================================================
@@ -643,33 +571,31 @@ def main():
                     similar_button = st.button(f"🔍 Find Similar Content", key=f"similar_{i}")
                     if similar_button:
                         with st.spinner("Finding similar content..."):
-                            similar_content = get_similar_content(content)
+                            similar_content = get_similar_content(content, n_recommendations=5)
                             
                             if similar_content:
                                 # Create tabs for different aspects of recommendations
                                 sim_tab1, sim_tab2 = st.tabs(["📺 Similar Titles", "🔍 Why These Recommendations"])
                                 
                                 with sim_tab1:
-                                    for sim_content in similar_content:
+                                    for sim_content in similar_content[:5]:  # Show top 5 similar items
                                         with st.container():
                                             col1, col2 = st.columns([3, 1])
                                             with col1:
                                                 st.write(f"**{sim_content.get('title')}** ({sim_content.get('type')}, {sim_content.get('release_year')})")
                                                 st.write(f"*Available on:* {sim_content.get('streaming_service')}")
                                                 st.write(f"*Genres:* {sim_content.get('listed_in')}")
+                                                st.write(f"*Cast:* {sim_content.get('cast')}")
+                                                st.write(f"*Director:* {sim_content.get('director')}")
+                                                st.write(f"*Description:* {sim_content.get('description')}")
                                             with col2:
                                                 st.write(f"**Match:** {sim_content.get('similarity', 'N/A')}")
-                                            
-                                            with st.expander("See more details"):
-                                                st.write(f"**Cast:** {sim_content.get('cast', 'N/A')}")
-                                                st.write(f"**Director:** {sim_content.get('director', 'N/A')}")
-                                                st.write(f"**Description:** {sim_content.get('description', 'N/A')}")
                                             st.write("---")
                                 
                                 with sim_tab2:
                                     st.write("**Why these recommendations?**")
                                     st.write("""
-                                    These recommendations are pre-computed using advanced content analysis:
+                                    These recommendations are based on multiple factors:
                                     - Genre and theme matching
                                     - Plot similarity analysis
                                     - Cast and director relationships