Update run.py

run.py

@@ -1,12 +1,11 @@
-import chromadb
 import os
 import gradio as gr
 import json
 from huggingface_hub import InferenceClient
 import gspread
-from oauth2client.service_account import ServiceAccountCredentials
-from datetime import datetime
 from google.oauth2 import service_account
+from datetime import datetime
+import chromadb
 
 # Google Sheets setup
 scope = ["https://spreadsheets.google.com/feeds", "https://www.googleapis.com/auth/drive"]
@@ -55,76 +54,50 @@ json_data={
 }
 creds = service_account.Credentials.from_service_account_info(json_data, scopes=scope)
 
-#creds = ServiceAccountCredentials.from_json_keyfile_name('/home/user/app/chromaold/nestolechatbot-5fe2aa26cb52.json', scope)
 client = gspread.authorize(creds)
 sheet = client.open("nestolechatbot").sheet1  # Open the sheet
 
-def save_to_sheet(date,name, message):
+def save_to_sheet(date, name, message):
     # Write user input to the Google Sheet
-    sheet.append_row([date,name, message])
+    sheet.append_row([date, name, message])
     return f"Thanks {name}, your message has been saved!"
 
-    
 path='/Users/thiloid/Desktop/LSKI/ole_nest/Chatbot/LLM/chromaTS'
-if(os.path.exists(path)==False): path="/home/user/app/chromaTS"
+if not os.path.exists(path):
+    path = "/home/user/app/chromaTS"
 
 print(path)
-#path='chromaTS'
-#settings = Settings(persist_directory=storage_path)
-#client = chromadb.Client(settings=settings)
 client = chromadb.PersistentClient(path=path)
 print(client.heartbeat()) 
 print(client.get_version())  
 print(client.list_collections()) 
+
 from chromadb.utils import embedding_functions
 default_ef = embedding_functions.DefaultEmbeddingFunction()
-sentence_transformer_ef = embedding_functions.SentenceTransformerEmbeddingFunction(model_name="T-Systems-onsite/cross-en-de-roberta-sentence-transformer")#"VAGOsolutions/SauerkrautLM-Mixtral-8x7B-Instruct")
-#instructor_ef = embedding_functions.InstructorEmbeddingFunction(model_name="hkunlp/instructor-large", device="cuda")
-#print(str(client.list_collections()))
-collection = client.get_collection(name="chromaTS", embedding_function=sentence_transformer_ef)
+sentence_transformer_ef = embedding_functions.SentenceTransformerEmbeddingFunction(model_name="T-Systems-onsite/cross-en-de-roberta-sentence-transformer")
 
-client = InferenceClient("mistralai/Mixtral-8x7B-Instruct-v0.1")
+collection = client.get_collection(name="chromaTS", embedding_function=sentence_transformer_ef)
 
+inference_client = InferenceClient("mistralai/Mixtral-8x7B-Instruct-v0.1")
 
-#def format_prompt(message, history):
-  #print("HISTORY")
-  #print(history)
-  #prompt = "" #"<s>"
-  #c=1
-  #for user_prompt, bot_response in history:
-  #  if c<2:
-  #      prompt += f"[INST] {user_prompt} [/INST]"
-  #      prompt += f" {bot_response}</s> "
-  #      c=c+1
-  #prompt += f"[INST] {message} [/INST]"
-  #print("Final P")
-  #print(prompt)
-  #return prompt
+# Global variable to store the URL
+global_url = ""
 
 def format_prompt(message, history):
     print("HISTORY")
     print(history)
-    
-    # Initialize prompt with an empty string
     prompt = ""
-    
-    # Check if history is not empty
     if history:
-        # Extract the most recent user prompt and bot response
         user_prompt, bot_response = history[-1]
         prompt += f"[INST] {user_prompt} [/INST] {bot_response}</s> "
-    
-    # Add the new message to the prompt
     prompt += f"[INST] {message} [/INST]"
-    
     print("Final P")
     print(prompt)
-    
     return prompt
 
-def response(
-    prompt, history,temperature=0.9, max_new_tokens=500, top_p=0.95, repetition_penalty=1.0,
-):
+def response(prompt, history, temperature=0.9, max_new_tokens=500, top_p=0.95, repetition_penalty=1.0):
+    global global_url
+    print(f"Working with URL: {global_url}")  # You can use the URL here
     temperature = float(temperature)
     if temperature < 1e-2: temperature = 1e-2
     top_p = float(top_p)
@@ -136,42 +109,67 @@ def response(
         do_sample=True,
         seed=42,
     )
-    search_prompt = format_prompt(prompt,history)
-    results=collection.query(
-      query_texts=[search_prompt],
-      n_results=60,
-      #where={"source": "google-docs"}
-      #where_document={"$contains":"search_string"}
+    search_prompt = format_prompt(prompt, history)
+    results = collection.query(
+        query_texts=[search_prompt],
+        n_results=60,
     )
-    #print("REsults")
-    #print(results)
-    #print("_____")
-    dists=["<br><small>(relevance: "+str(round((1-d)*100)/100)+";" for d in results['distances'][0]]
-  
-    #sources=["source: "+s["source"]+")</small>" for s in results['metadatas'][0]]
-    results=results['documents'][0]
-    #print("TEst")
-    #print(results)
-    #print("_____")
-    combination = zip(results,dists)
+    dists = ["<br><small>(relevance: " + str(round((1-d)*100)/100) + ";" for d in results['distances'][0]]
+    results = results['documents'][0]
+    combination = zip(results, dists)
     combination = [' '.join(triplets) for triplets in combination]
-    #print(str(prompt)+"\n\n"+str(combination))
-    if(len(results)>1):
-      addon="Bitte berücksichtige bei deiner Antwort ausschießlich folgende Auszüge aus unserer Datenbank, sofern sie für die Antwort erforderlich sind. Beantworte die Frage knapp und präzise. Ignoriere unpassende Datenbank-Auszüge OHNE sie zu kommentieren, zu erwähnen oder aufzulisten:\n"+"\n".join(results)
-    system="Du bist ein deutschsprachiges KI-basiertes Studienberater Assistenzsystem, das zu jedem Anliegen möglichst geeignete Studieninformationen empfiehlt."+addon+"\n\nUser-Anliegen:"   
-    formatted_prompt = format_prompt(system+"\n"+prompt,history)
-    stream = client.text_generation(formatted_prompt, **generate_kwargs, stream=True, details=True, return_full_text=False)
+    if len(results) > 1:
+        addon = "Bitte berücksichtige bei deiner Antwort ausschießlich folgende Auszüge aus unserer Datenbank, sofern sie für die Antwort erforderlich sind. Beantworte die Frage knapp und präzise. Ignoriere unpassende Datenbank-Auszüge OHNE sie zu kommentieren, zu erwähnen oder aufzulisten:\n" + "\n".join(results)
+    system = "Du bist ein deutschsprachiges KI-basiertes Studienberater Assistenzsystem, das zu jedem Anliegen möglichst geeignete Studieninformationen empfiehlt." + addon + "\n\nUser-Anliegen:"   
+    formatted_prompt = format_prompt(system + "\n" + prompt, history)
+    stream = inference_client.text_generation(formatted_prompt, **generate_kwargs, stream=True, details=True, return_full_text=False)
     output = ""
     for response in stream:
         output += response.token.text
         yield output
-    #output=output+"\n\n<br><details open><summary><strong>Sources</strong></summary><br><ul>"+ "".join(["<li>" + s + "</li>" for s in combination])+"</ul></details>"
-    # Get current date and time
     now = str(datetime.now())
-    save_to_sheet(now,prompt, output)
+    save_to_sheet(now, prompt, output)
     yield output
 
-gr.ChatInterface(response, chatbot=gr.Chatbot(value=[[None,"Herzlich willkommen! Ich bin Chätti ein KI-basiertes Studienassistenzsystem, das für jede Anfrage die am besten Studieninformationen empfiehlt.<br>Erzähle mir, was du gerne tust!"]],render_markdown=True),title="German Studyhelper Chätti").queue().launch(share=True) #False, server_name="0.0.0.0", server_port=7864)
-print("Interface up and running!")
+def js_code():
+    return """
+    <script>
+        function getUrl() {
+            const url = window.location.href;
+            const xhr = new XMLHttpRequest();
+            xhr.open("POST", "/submit_url", true);
+            xhr.setRequestHeader("Content-Type", "application/json");
+            xhr.onreadystatechange = function() {
+                if (xhr.readyState === 4 && xhr.status === 200) {
+                    console.log("URL submitted successfully");
+                }
+            };
+            xhr.send(JSON.stringify({ url: url }));
+        }
+        window.onload = getUrl;
+    </script>
+    """
+
+def submit_url(url: str):
+    global global_url
+    global_url = url  # Save the URL in the global variable
+    print(f"Received URL: {url}")
+    return url
 
+iface = gr.Interface(
+    fn=response,
+    inputs=gr.Chatbot(value=[[None,"Herzlich willkommen! Ich bin Chätti ein KI-basiertes Studienassistenzsystem, das für jede Anfrage die am besten Studieninformationen empfiehlt.<br>Erzähle mir, was du gerne tust!"]], render_markdown=True),
+    title="German Studyhelper Chätti"
+)
 
+# Add a route to handle the URL submission
+iface.launch(share=True, js=js_code())
+
+@gr.routes.post("/submit_url")
+async def process_url(request):
+    data = await request.json()
+    url = data.get("url", "")
+    submit_url(url)
+    return {"status": "success"}
+
+print("Interface up and running!")