Update app.py

app.py

@@ -1,4 +1,4 @@
-
+# Import libraries and references:
 import anthropic
 import base64
 import glob
@@ -15,7 +15,6 @@ import time
 import traceback
 import uuid
 import zipfile
-
 from PIL import Image
 from azure.cosmos import CosmosClient, exceptions
 from datetime import datetime
@@ -24,13 +23,13 @@ from github import Github
 from gradio_client import Client
 from urllib.parse import quote
 
+
 # 🎭 App Configuration - Because every app needs a good costume!
-Site_Name = '🐙Git🌌Cosmos💫 - Azure Cosmos DB and Github Agent'
-title = "🐙Git🌌Cosmos💫 - Azure Cosmos DB and Github Agent"
+Site_Name = '🐙GitCosmos🌌 - AI Azure Cosmos DB and Github Agent'
+title = "🐙GitCosmos🌌 - AI Azure Cosmos DB and Github Agent"
 helpURL = 'https://huggingface.co/awacke1'
-bugURL = 'https://huggingface.co/spaces/awacke1'
+bugURL = 'https://huggingface.co/spaces/awacke1/AzureCosmosDBUI/'
 icons = '🐙🌌💫'
-
 st.set_page_config(
     page_title=title,
     page_icon=icons,
@@ -43,6 +42,7 @@ st.set_page_config(
     }
 )
 
+
 # 🌌 Cosmos DB configuration - Where data goes to party!
 ENDPOINT = "https://acae-afd.documents.azure.com:443/"
 DATABASE_NAME = os.environ.get("COSMOS_DATABASE_NAME")
@@ -60,6 +60,8 @@ client = anthropic.Anthropic(api_key=os.environ.get("ANTHROPIC_API_KEY"))
 if "chat_history" not in st.session_state:
     st.session_state.chat_history = []
 
+
+
 # 🛠️ Helper Functions - The unsung heroes of our code
 
 # 📎 Get a file download link - Making file sharing as easy as stealing candy from a baby
@@ -207,6 +209,8 @@ def save_to_cosmos_db(container, query, response1, response2):
     except Exception as e:
         st.error(f"An unexpected error occurred: {str(e)}")
 
+
+
 # 🐙 GitHub functions - Where code goes to socialize
 
 # 📥 Download GitHub repo - Cloning repos like it's going out of style
@@ -228,24 +232,19 @@ def create_repo(g, repo_name):
 def push_to_github(local_path, repo, github_token):
     repo_url = f"https://{github_token}@github.com/{repo.full_name}.git"
     local_repo = Repo(local_path)
-
     if 'origin' in [remote.name for remote in local_repo.remotes]:
         origin = local_repo.remote('origin')
         origin.set_url(repo_url)
     else:
         origin = local_repo.create_remote('origin', repo_url)
-
     if not local_repo.heads:
         local_repo.git.checkout('-b', 'main')
         current_branch = 'main'
     else:
         current_branch = local_repo.active_branch.name
-
     local_repo.git.add(A=True)
-
     if local_repo.is_dirty():
         local_repo.git.commit('-m', 'Initial commit')
-
     origin.push(refspec=f'{current_branch}:{current_branch}')
 
 
@@ -254,14 +253,11 @@ def save_or_clone_to_cosmos_db(container, document=None, clone_id=None):
         timestamp = datetime.utcnow().strftime('%Y%m%d%H%M%S%f')
         random_component = ''.join(random.choices('abcdefghijklmnopqrstuvwxyz0123456789', k=8))
         return f"{timestamp}-{random_component}-{str(uuid.uuid4())}"
-
     max_retries = 10
     base_delay = 0.1
-
     for attempt in range(max_retries):
         try:
             new_id = generate_complex_unique_id()
-            
             if clone_id:
                 try:
                     existing_doc = container.read_item(item=clone_id, partition_key=clone_id)
@@ -276,15 +272,12 @@ def save_or_clone_to_cosmos_db(container, document=None, clone_id=None):
                     return False, f"Document with ID {clone_id} not found for cloning."
             else:
                 if document is None:
-                    return False, "No document provided for saving"
-                
+                    return False, "No document provided for saving"      
                 document['id'] = new_id
                 document['created_at'] = datetime.utcnow().isoformat()
                 new_doc = document
-
             response = container.create_item(body=new_doc)
             return True, f"{'Cloned' if clone_id else 'New'} document saved successfully with ID: {response['id']}"
-
         except exceptions.CosmosHttpResponseError as e:
             if e.status_code == 409:
                 delay = base_delay * (2 ** attempt) + random.uniform(0, 0.1)
@@ -293,7 +286,6 @@ def save_or_clone_to_cosmos_db(container, document=None, clone_id=None):
             return False, f"Error saving to Cosmos DB: {str(e)}"
         except Exception as e:
             return False, f"An unexpected error occurred: {str(e)}"
-    
     return False, "Failed to save document after maximum retries."
 
     
@@ -304,45 +296,37 @@ def archive_current_container(database_name, container_name, client):
         if os.path.exists(base_dir):
             shutil.rmtree(base_dir)
         os.makedirs(base_dir)
-
         db_client = client.get_database_client(database_name)
         container_client = db_client.get_container_client(container_name)
         items = list(container_client.read_all_items())
-        
         container_dir = os.path.join(base_dir, container_name)
         os.makedirs(container_dir)
-        
         for item in items:
             item_id = item.get('id', f"unknown_{datetime.now().strftime('%Y%m%d%H%M%S')}")
             with open(os.path.join(container_dir, f"{item_id}.json"), 'w') as f:
                 json.dump(item, f, indent=2)
-        
         archive_name = f"{container_name}_archive_{datetime.now().strftime('%Y%m%d%H%M%S')}"
         shutil.make_archive(archive_name, 'zip', base_dir)
-        
         return get_download_link(f"{archive_name}.zip")
     except Exception as e:
         return f"An error occurred while archiving data: {str(e)} 😢"
 
+        
 # 🔍 Search glossary - Finding needles in digital haystacks
 def search_glossary(query):
     st.markdown(f"### 🔍 SearchGlossary for: {query}")
-
     # Dropdown for model selection
     model_options = ['mistralai/Mixtral-8x7B-Instruct-v0.1', 'mistralai/Mistral-7B-Instruct-v0.2', 'google/gemma-7b-it', 'None']
     model_choice = st.selectbox('🧠 Select LLM Model', options=model_options, index=1)
-
     # Dropdown for database selection
     database_options = ['Semantic Search', 'Arxiv Search - Latest - (EXPERIMENTAL)']
     database_choice = st.selectbox('📚 Select Database', options=database_options, index=0)
-
     # 🕵️‍♂️ Searching the glossary for: query
     all_results = ""
-    st.markdown(f"- {query}")
-
+    st.markdown(f"- {query
+    
     # 🔍 ArXiv RAG researcher expert ~-<>-~ Paper Summary & Ask LLM
     client = Client("awacke1/Arxiv-Paper-Search-And-QA-RAG-Pattern")
-
     # 🔍 ArXiv RAG researcher expert ~-<>-~ Paper Summary & Ask LLM - api_name: /ask_llm
     result = client.predict(
         prompt=query,
@@ -352,7 +336,6 @@ def search_glossary(query):
     )
     st.markdown(result)
     st.code(result, language="python", line_numbers=True)
-
     # 🔍 ArXiv RAG researcher expert ~-<>-~ Paper Summary & Ask LLM - api_name: /ask_llm
     result2 = client.predict(
         prompt=query,
@@ -362,7 +345,6 @@ def search_glossary(query):
     )
     st.markdown(result2)
     st.code(result2, language="python", line_numbers=True)
-
     # 🔍 ArXiv RAG researcher expert ~-<>-~ Paper Summary & Ask LLM - api_name: /ask_llm
     result3 = client.predict(
         prompt=query,
@@ -372,7 +354,6 @@ def search_glossary(query):
     )
     st.markdown(result3)
     st.code(result3, language="python", line_numbers=True)
-
     # 🔍 ArXiv RAG researcher expert ~-<>-~ Paper Summary & Ask LLM - api_name: /update_with_rag_md
     response2 = client.predict(
         message=query,	# str  in 'parameter_13' Textbox component
@@ -381,19 +362,13 @@ def search_glossary(query):
         llm_model_picked="mistralai/Mistral-7B-Instruct-v0.2",
         api_name="/update_with_rag_md"
     )
-
     st.markdown(response2[0])
     st.code(response2[0], language="python", line_numbers=True, wrap_lines=True)  
-
     st.markdown(response2[1])
     st.code(response2[1], language="python", line_numbers=True, wrap_lines=True)
-
     return result, result2, result3, response2
 
 
-
-
-
 # 📝 Generate a safe filename from the first few lines of content
 def generate_filename_from_content(content, file_type="md"):
     # Extract the first few lines or sentences
@@ -404,6 +379,7 @@ def generate_filename_from_content(content, file_type="md"):
     safe_name = safe_name[:50].strip()  # Adjust length limit
     return f"{safe_name}.{file_type}"
 
+
 # 💾 Create and save a file
 def create_file_from_content(content, should_save=True):
     if not should_save:
@@ -413,12 +389,12 @@ def create_file_from_content(content, should_save=True):
         file.write(content)
     return filename
 
+
 # 📂 Display list of saved .md files in the sidebar
 def display_saved_files_in_sidebar():
     all_files = glob.glob("*.md")
     all_files.sort(reverse=True)
     all_files = [file for file in all_files if not file.lower().startswith('readme')]  # Exclude README.md
-
     st.sidebar.markdown("## 📁 Saved Markdown Files")
     for file in all_files:
         col1, col2, col3 = st.sidebar.columns([6, 2, 1])
@@ -435,7 +411,6 @@ def display_saved_files_in_sidebar():
                 os.remove(file)
                 st.rerun()
 
-
 def clone_record(container, clone_id):
     try:
         existing_doc = container.read_item(item=clone_id, partition_key=clone_id)
@@ -448,17 +423,14 @@ def clone_record(container, clone_id):
         new_doc['_etag'] = None
         new_doc['_attachments'] = None
         new_doc['_ts'] = None  # Reset timestamp to be updated by Cosmos DB automatically
-
         # Insert the cloned document
         response = container.create_item(body=new_doc)
         st.success(f"Cloned document saved successfully with ID: {new_doc['id']} 🎉")
-        
         # Refresh the documents in session state
         st.session_state.documents = list(container.query_items(
             query="SELECT * FROM c ORDER BY c._ts DESC",
             enable_cross_partition_query=True
         ))
-
     except exceptions.CosmosResourceNotFoundError:
         st.error(f"Document with ID {clone_id} not found for cloning.")
     except exceptions.CosmosHttpResponseError as e:
@@ -467,8 +439,6 @@ def clone_record(container, clone_id):
         st.error(f"An unexpected error occurred: {str(e)} 😱")
 
 
-
-
 def create_new_blank_record(container):
     try:
         # Get the structure of the latest document (to preserve schema)
@@ -477,52 +447,45 @@ def create_new_blank_record(container):
             new_doc_structure = latest_doc[0].copy()
         else:
             new_doc_structure = {}
-
         new_doc = {key: "" for key in new_doc_structure.keys()}  # Set all fields to blank
         new_doc['id'] = generate_unique_id()  # Generate new unique ID
         new_doc['createdAt'] = datetime.utcnow().isoformat()  # Set creation time
-
         # Insert the new blank document
         response = container.create_item(body=new_doc)
         st.success(f"New blank document saved successfully with ID: {new_doc['id']} 🎉")
-        
         # Refresh the documents in session state
         st.session_state.documents = list(container.query_items(
             query="SELECT * FROM c ORDER BY c._ts DESC",
             enable_cross_partition_query=True
         ))
-
     except exceptions.CosmosHttpResponseError as e:
         st.error(f"HTTP error occurred: {str(e)} 🚨")
     except Exception as e:
         st.error(f"An unexpected error occurred: {str(e)} 😱")
 
+
 # Function to preprocess the pasted content
 def preprocess_text(text):
     # Replace CRLF and other newline variations with the JSON newline escape sequence
     text = text.replace('\r\n', '\\n')
     text = text.replace('\r', '\\n')
     text = text.replace('\n', '\\n')
-    
     # Escape double quotes inside the text
     text = text.replace('"', '\\"')
-    
     # Optionally remove or handle other special characters that might not be JSON-safe
     # Here, we remove characters like tabs or non-ASCII characters (as an example)
     text = re.sub(r'[\t]', ' ', text)  # Replace tabs with spaces
     text = re.sub(r'[^\x00-\x7F]+', '', text)  # Remove non-ASCII characters
-    
     # Normalize spaces (strip leading/trailing whitespace)
     text = text.strip()
     return text
 
+
                 
 # 🎭 Main function - "All the world's a stage, and all the code merely players" -Shakespeare, probably
 def main():
     st.title("🐙Git🌌Cosmos💫 - Azure Cosmos DB and Github Agent")
 
-
-    
     # 🎲 Session state vars - "Life is like a session state, you never know what you're gonna get"
     if 'logged_in' not in st.session_state:
         st.session_state.logged_in = False
@@ -541,8 +504,6 @@ def main():
     if 'cloned_doc' not in st.session_state:
         st.session_state.cloned_doc = None
 
-
-    
     # 🔍 Query processing - "To search or not to search, that is the query" 
     try:
         query_params = st.query_params
@@ -577,13 +538,10 @@ def main():
                 st.error(f"HTTP error occurred: {str(e)} 🚨")
             except Exception as e:
                 st.error(f"An unexpected error occurred: {str(e)} 😱")
-
             st.stop()
     except Exception as e:
         st.markdown(' ')
 
-
-    
     # 🔐 Auth check - "With great keys come great connectivity"
     if Key:
         st.session_state.primary_key = Key
@@ -592,22 +550,17 @@ def main():
         st.error("Cosmos DB Key is not set in environment variables. 🔑❌")
         return
 
-
-    
     if st.session_state.logged_in:
         # 🌌 DB initialization - "In the beginning, there was connection string..."
         try:
             if st.session_state.client is None:
                 st.session_state.client = CosmosClient(ENDPOINT, credential=st.session_state.primary_key)
-            
             # 📚 Navigation setup - "Navigation is not about where you are, but where you're going"
             st.sidebar.title("🐙Git🌌Cosmos💫🗄️Navigator")
-            
             databases = get_databases(st.session_state.client)
             selected_db = st.sidebar.selectbox("🗃️ Select Database", databases)
             st.markdown(CosmosDBUrl)
 
-            
             # 🔄 State management - "Change is the only constant in state management"
             if selected_db != st.session_state.selected_database:
                 st.session_state.selected_database = selected_db
@@ -615,8 +568,6 @@ def main():
                 st.session_state.selected_document_id = None
                 st.session_state.current_index = 0
                 st.rerun()
-
-
                 
             if st.session_state.selected_database:
                 database = st.session_state.client.get_database_client(st.session_state.selected_database)
@@ -632,7 +583,6 @@ def main():
                 
                 if st.session_state.selected_container:
                     container = database.get_container_client(st.session_state.selected_container)
-                    
                     # 📦 Export functionality - "Pack it, zip it, ship it"
                     if st.sidebar.button("📦 Export Container Data"):
                         download_link = archive_current_container(st.session_state.selected_database, 
@@ -643,17 +593,13 @@ def main():
                         else:
                             st.error(download_link)
 
-
-                    
                     # 📝 Document handling - "Document, document, on the wall, who's the most recent of them all?"
                     documents = get_documents(container)
                     total_docs = len(documents)
-                    
                     # Add a slider to let the user choose how many documents to display
                     num_docs_to_display = st.slider(
-                        "Select number of documents to display", 5, 500, 20
+                        "Select number of documents to display", 5, 500, 10
                     )
-                    
                     # Adjust the document display logic based on the slider value
                     if total_docs > num_docs_to_display:
                         documents_to_display = documents[:num_docs_to_display]
@@ -662,12 +608,11 @@ def main():
                         documents_to_display = documents
                         st.sidebar.info(f"Showing all {len(documents_to_display)} documents.")
 
-                    
-                    
                     if documents_to_display:
                         # 🎨 View options - "Different strokes for different folks"
                         view_options = ['Show as Markdown', 'Show as Code Editor', 'Show as Edit, Save, Run AI', 'Clone Document', 'New Record']
                         selected_view = st.sidebar.selectbox("Select Viewer/Editor", view_options, index=2)
+
                         
                         if selected_view == 'Show as Markdown':
                             Label = '# 📄 Markdown view - Mark it down, mark it up'
@@ -697,8 +642,6 @@ def main():
                             content = json.dumps(doc, indent=2)
                             st.markdown(f"```json\n{content}\n```")
 
-
-                            
                             # ⬅️➡️ Navigation - "Left and right, day and night"
                             col_prev, col_next = st.columns([1, 1])
                             with col_prev:
@@ -711,6 +654,8 @@ def main():
                                     if st.session_state.current_index < total_docs - 1:
                                         st.session_state.current_index += 1
                                         st.rerun()
+
+
                         
                         elif selected_view == 'Show as Code Editor':
                             Label = '# 💻 Code editor view'
@@ -762,10 +707,6 @@ def main():
                                     except Exception as e:
                                         st.error(f"Error deleting document: {str(e)}")                                        
 
-
-                                        
-                                        
-
                                     
                         elif selected_view == 'Show as Edit, Save, Run AI':
                             Label = '# ✏️ Edit and save view - Edit with wisdom, save with precision'
@@ -775,8 +716,6 @@ def main():
                             num_cols = len(documents_to_display)
                             cols = st.columns(num_cols)
 
-
-                            
                             for idx, (col, doc) in enumerate(zip(cols, documents_to_display)):
                                 with col:
                                     
@@ -817,9 +756,7 @@ def main():
                                     with col_ai:
                                         if st.button("🤖 Run AI", key=f'run_with_ai_button_{idx}'):
                                             search_glossary(json.dumps(editable_doc, indent=2))
-                                    
-
-                        
+                                            
                                                 
                         elif selected_view == 'Clone Document':
                             st.markdown("#### Clone a document:")
@@ -913,8 +850,6 @@ def main():
                                         st.info("No documents to display. 🧐")
                                         
 
-
-                
             Label = '# 🐙 GitHub integration - Git happens'
             st.subheader("🐙 GitHub Operations")
             github_token = os.environ.get("GITHUB")
@@ -923,8 +858,6 @@ def main():
             new_repo_name = st.text_input("New Repository Name (for cloning)", 
                                         value=f"AIExample-Clone-{datetime.now().strftime('%Y%m%d_%H%M%S')}")
 
-
-            
             col1, col2 = st.columns(2)
             with col1:
                 if st.button("📥 Clone Repository"):
@@ -948,8 +881,6 @@ def main():
                     else:
                         st.error("Please ensure GitHub token is set in environment variables and source repository URL is provided. 🔑❓")
 
-
-            
             with col2:
                 if st.button("📤 Push to New Repository"):
                     if github_token and source_repo:
@@ -970,7 +901,6 @@ def main():
                     else:
                         st.error("Please ensure GitHub token is set in environment variables and source repository URL is provided. 🔑❓")
 
-
             
             st.subheader("💬 Chat with Claude")
             user_input = st.text_area("Message 📨:", height=100)
@@ -988,12 +918,9 @@ def main():
                     )
                     st.write("Claude's reply 🧠:")
                     st.write(response.content[0].text)
-
                     filename = generate_filename(user_input, "md")
                     create_file(filename, user_input, response.content[0].text)
-
                     st.session_state.chat_history.append({"user": user_input, "claude": response.content[0].text})
-
                     # Save to Cosmos DB
                     save_to_cosmos_db(container, user_input, response.content[0].text, "")
 
@@ -1005,8 +932,7 @@ def main():
                 st.text_area("You said 💬:", chat["user"], height=100, disabled=True)
                 st.text_area("Claude replied 🤖:", chat["claude"], height=200, disabled=True)
                 st.markdown("---")
-
-
+                
             
             # 📝 File editor - "Edit with care, save with flair"
             if hasattr(st.session_state, 'current_file'):
@@ -1023,8 +949,6 @@ def main():
                         file.write(new_content)
                     st.success("File updated successfully! 🎉")
 
-
-            
             # 📂 File management - "Manage many, maintain order"
             st.sidebar.title("📁 File Management")
             
@@ -1057,14 +981,10 @@ def main():
                         os.remove(file)
                         st.rerun()
 
-                        
-
         except exceptions.CosmosHttpResponseError as e:
             st.error(f"Failed to connect to Cosmos DB. HTTP error: {str(e)} 🚨")
         except Exception as e:
             st.error(f"An unexpected error occurred: {str(e)} 😱")
-
-
     
     if st.session_state.logged_in and st.sidebar.button("🚪 Logout"):
         Label = '# 🚪 Logout - All good things must come to an end'