chcunks

analyzer.py

@@ -11,17 +11,12 @@ def analyze_code(code: str) -> str:
     client = OpenAI(api_key=os.getenv("modal_api"))
     client.base_url = os.getenv("base_url")
     system_prompt = (
-        "You are a helpful assistant. Analyze the code given to you. "
-        "Return your response strictly in JSON format with the following keys: "
-        "'strength', 'weaknesses', 'speciality', 'relevance rating'. "
-        "Do not include any other text outside the JSON."
-        "the reply should just be the following format:"
-        "{"
-        "  'strength': '...', "
-        "  'weaknesses': '...', "
-        "  'speciality': '...', "
-        "  'relevance rating': '...'"
-        "}"
+        "You are a highly precise and strict JSON generator. Analyze the code given to you. "
+        "Your ONLY output must be a valid JSON object with the following keys: 'strength', 'weaknesses', 'speciality', 'relevance rating'. "
+        "Do NOT include any explanation, markdown, or text outside the JSON. Do NOT add any commentary, preamble, or postscript. "
+        "If you cannot answer, still return a valid JSON with empty strings for each key. "
+        "Example of the ONLY valid output:\n"
+        "{\n  'strength': '...', \n  'weaknesses': '...', \n  'speciality': '...', \n  'relevance rating': '...'\n}"
     )
     response = client.chat.completions.create(
         model="neuralmagic/Meta-Llama-3.1-8B-Instruct-quantized.w4a16",  # Updated model
@@ -80,12 +75,19 @@ def combine_repo_files_for_llm(repo_dir="repo_files", output_file="combined_repo
 
 def analyze_combined_file(output_file="combined_repo.txt"):
     """
-    Reads the combined file and passes its contents to analyze_code, returning the LLM's output.
+    Reads the combined file, splits it into 500-line chunks, analyzes each chunk, and aggregates the LLM's output.
+    Returns the aggregated analysis as a string.
     """
     try:
         with open(output_file, "r", encoding="utf-8") as f:
             lines = f.readlines()
-        code = "".join(lines[:500])
-        return analyze_code(code)
+        chunk_size = 500
+        analyses = []
+        for i in range(0, len(lines), chunk_size):
+            chunk = "".join(lines[i:i+chunk_size])
+            analysis = analyze_code(chunk)
+            analyses.append(analysis)
+        # Optionally, you could merge the JSONs here, but for now, return all analyses as a list
+        return "\n---\n".join(analyses)
     except Exception as e:
         return f"Error analyzing combined file: {e}"

chatbot_page.py

@@ -4,7 +4,7 @@ import os
 
 # System prompt for the chatbot
 CHATBOT_SYSTEM_PROMPT = (
-    "You are a helpful assistant. Your goal is to help the user describe their ideal open-source repo. "
+    "You are a helpful assistant. Your goal is to help the user describe their ideal Hugging face repo. "
     "Ask questions to clarify what they want, their use case, preferred language, features, etc. "
     "When the user clicks 'End Chat', analyze the conversation and return about 5 keywords for repo search. "
     "Return only the keywords as a comma-separated list."
@@ -69,7 +69,9 @@ def extract_keywords_from_conversation(history):
 with gr.Blocks() as chatbot_demo:
     gr.Markdown("## Repo Recommendation Chatbot")
     chatbot = gr.Chatbot()
-    state = gr.State([])  # conversation history
+    # Initial assistant message
+    initial_message = "Hello! What kind of open-source repo are you looking for? Please describe your ideal repo, use case, preferred language, or any features you want."
+    state = gr.State([["", initial_message]])  # Start with assistant message
     user_input = gr.Textbox(label="Your message", placeholder="Describe your ideal repo or answer the assistant's questions...")
     send_btn = gr.Button("Send")
     end_btn = gr.Button("End Chat and Extract Keywords")