Update app.py

app.py

@@ -133,11 +133,11 @@ def process_directory(directory, partial_paths=None, file_paths=None):
             with zipfile.ZipFile(zip_file_path, 'r') as zip_ref:
                 zip_ref.extractall(tmpdirname)
             print(f"Extracted {zip_file_path} to {tmpdirname}")
-
+            
+            files = []
             if partial_paths:
                 files = get_all_files_in_folder(tmpdirname, partial_paths)
             else:
-                files = []
                 for root, _, files_list in os.walk(tmpdirname):
                     for file in files_list:
                         files.append(os.path.join(root, file))
@@ -155,7 +155,7 @@ def process_directory(directory, partial_paths=None, file_paths=None):
                 with open(file_path, 'rb') as f:
                     if file_ext in ['.rst', '.md', '.txt', '.html', '.json', '.yaml', '.py']:
                         text = f.read().decode('utf-8')
-                        print(f"Extracted text from {file_path}:\n{text[:200]}...\n")
+                        #print(f"Extracted text from {file_path}:\n{text[:200]}...\n")
                     elif file_ext in ['.svg']:
                         text = f"SVG file content from {file_path}"
                     elif file_ext in ['.png', '.ico']:
@@ -225,13 +225,14 @@ def rag_workflow(query):
 def initialize():
     global vectorstore, chunks, llm
 
-    download_gitlab_repo()
     partial_paths = ['kadi-apy-master/source/docs/setup/', 'kadi-apy-master/docs/source/usage/', 'kadi-apy-master/kadi_apy/lib/']
     file_paths = ['kadi-apy-master/docs/source/usage/lib.rst']
-    all_texts, file_references = process_directory(REPOSITORY_DIRECTORY, partial_paths, file_paths)
-    chunks = split_into_chunks(all_texts, file_references, CHUNK_SIZE, CHUNK_OVERLAP)
-    vectorstore = setup_chroma(chunks, EMBEDDING_MODEL_NAME, PERSIST_DIRECTORY)
-    llm = setup_llm(LLM_MODEL_NAME, LLM_TEMPERATURE, GROQ_API_KEY)
+    all_texts, file_references = process_directory(r'C:\Users\Anwender\Downloads\Neuer Ordner', partial_paths, file_paths)
+    chunks = split_into_chunks(all_texts, file_references, 512, 0)
+    print(f"Total number of chunks: {len(chunks)}")
+    # vectorstore = setup_chroma(chunks, EMBEDDING_MODEL_NAME, PERSIST_DIRECTORY)
+    # llm = setup_llm(LLM_MODEL_NAME, LLM_TEMPERATURE, GROQ_API_KEY)
+
 
 initialize()